Sunday, 10 April 2011

Philosophy and happiness?

Taken all together how satisfied or dissatisfied are you with your life as a whole these days on a scale of 1-10?

Me: I literally have no idea what that question means.

Sociologist: Yes, we find that philosophers say that quite a lot. But ordinary people seem quite capable of understanding it and giving a sensible answer.
Philosophers, and other academics, have been arguing about the nature of the good life for a very long time, but one thing we rarely take very seriously is subjective happiness. Perhaps we should.

I recently had the pleasure of attending a workshop on happiness by Ruut Veenhoven, leading expert on happiness and director of the fascinating World Database of Happiness. He argues that happiness is actually rather easy to measure and should be central to the analysis of quality of life. The typical philosopher is inclined to be sceptical about both, but Veenhoven's case is surprisingly persuasive.

He begins by noting the various dimensions of quality of life we might consider in the assessment of a person's quality of life: i) opportunities and resources; ii) personal talents and psychology; iii) valuable achievements; iv) happiness. The first 3 are traditional academic objects of study, resulting in all sorts of lists of different items that get quantified as quality of life indexes. 

Sometimes these indexes are restricted to particular aspects of one dimension, like city quality of life indexes, or stretch further like the Quality Adjusted Life Years indexes used in health economics. The influential Human Development Index covers all three dimensions in a very simple way by measuring a single item in each dimension and weighting them equally: GDP per capita (general purpose resources); literacy (an important personal talent); and longevity (a valuable achievement).

Veenhoven points out that while such lists include items that are more or less easy to measure (like literacy), interpreting their significance is somewhat problematic because it is not obvious that the things you managed to measure adequately represent general achievement in that dimension. A functional literacy rate of 90% is better than 80%, yet what does it really say about how well people's lives are going in general?

In contrast happiness is easy to interpret, but - one might think - impossible to measure. Veenhoven's breakthrough is in demonstrating that measuring happiness is actually relatively easy. One can just go up to people and ask them and they will tell you. (The number of people who take the 'don't know' option, presumably including the world's philosophers, is only around 1% of respondents.) The survey method is robust across different different people (countries, cultures, languages, gender, age) so that it seems possible to find out how happy people are in different circumstances around the world (Denmark and Costa Rica are the world leaders; Sub-Saharan Africa fills the bottom). And also to analyse the main determinants of happiness by regression analysis: governance, gender equality, democracy, and wealth (but not income inequality, as many people might expect).

Now the question is what to make of this new found tool. Philosophers should be somewhat humble in approaching this, I suggest, particularly in light of our common failure of ordinary understanding I mentioned at the start. Philosophers may, by our training, have become particularly focussed on rationally derived standards of well-being - the unexamined life is not worth living, and so forth - to the exclusion of the more affective sense of well-being. We cannot simply dismiss such feelings of happiness because we don't feel them, or feel their importance, as non-philosophers overwhelmingly seem to. We need to base our response on objective arguments, not our subjective ignorance.

One important issue is that once happiness is recognised as easy to interpret and easy to measure, it may easily swamp our other ways of thinking about quality of life. This would be the opposite danger to the one Veenhoven identified in the common disdain among academics towards taking account of happiness. Simply because something is easy to measure in numbers doesn't make it any more significant. Simply because some aspects of the good life - like justice - are hard to measure doesn't mean they are any less significant. Indeed, Veenhoven's own research shows that there is considerable variation in how people perceive the importance of happiness as one among other valuable dimensions of life. So philosophers need not surrender to the power of large-scale surveys. There is still important work to do in clarifying the different dimensions of a valuable human life and debating how they should be considered in relation to each other, for example in individual and social choice situations.

Yet the appearance of this credible information about a rather neglected dimension of human life raises some immediate issues for moral philosophers to address. Now that we know something about the international distribution of happiness and its determinants, should we see low happiness as an injustice in the same way that the deprivation of basic needs like food and shelter in a world of plenty cries out for action? For example should we consider that everyone has a right to happiness that governments have a duty to meet? If so perhaps happiness should be included as a 4th dimension on the Human Development Index, an extra item on the Millennium Development Goals, or an additional distinct right in UN Declaration of Human Rights.

What should be the role of technocrats? Is there a risk of enthroning a ruling elite of sociologist-kings? Veenhoven's research shows that although people are pretty good at telling how happy they are with their lives, they are rather poor at telling what makes them happy, or the happiness consequences of the life choices they are contemplating (e.g. having children - not good). He sees this as evidence of the primitiveness of affective happiness - the capability to recognise it when we have it does not require the more abstract reasoning skills required to understand why we have it. 

This raises issues concerning how knowledge of the determinants of happiness should be used. Should people be informed about the typical happiness outcomes of different life choices in order to allow rational choice? Should governments act discreetly to tweak institutions and choice framing so that people are naturally led to make choices that will generally maximise their happiness (a variation on libertarian paternalism)?

4 comments:

  1. Ha! For a philosopher, you're quick to adopt the mindset of an economist. Just because someone throws a subjective number on something doesn't make it meaningful. Each respondent who answered "7" to Veenhoven's question, for instance, necessarily had a different conception of what that answer meant.

    Moreover, if Veenhoven asked everyone to answer the same question a month, week, or even a day or hour later, I feel confident in predicting that he would have gotten different answers out of many of the same people.

    As the street philosopher Biggie Smalls put it: Cash rules everything around me. Thus, economists and their little statistician helpers are very in vogue right now. Don't be fooled. Qualitative analysis is what really counts if you want serious answers, not mathematical shortcuts.

    ReplyDelete
  2. Actually, Veenhoven is a sociologist. Economists have in fact long rejected any attempt to measure subjective happiness objectively (Lionel Robbins made this official back in 1932 in his influential "Essay on the Nature and Significance of Economic Science"). That's why economists are so obsessed with indirect measures of welfare, like choices and money, but not happiness.

    But I had the same immediate reactions as you about Veenhoven's survey methodology, since the credibility of his project rests on happiness being a relatively inter-temporally and inter-subjectively stable concept. Veenhoven argues that asking people about happiness as satisfaction rather than momentary feelings works against the first: people do in fact give pretty much the same answer from day to day. He also uses various techniques to calibrate the inter-personal understanding of the questions asked (e.g. experiential sampling, different formulations of the questions, etc) and argues that this exercise shows that people do have approximately the same understanding of happiness. I find him quite convincing on this, but of course a proper assessment of his methodology would require a lot more familiarity with the happiness literature and statistical techniques than I have to offer.

    All in all Veenhoven considers that 'measurement error' probably explains around 5-10% of happiness survey results (which he notes is comparable to the measurement error for 'objective' things like GDP).

    ReplyDelete
  3. Correction - it was Wu Tang who sang C.R.E.A.M., not Biggie. Wow, it's surprising to me that economists had the humility to officially not to try to measure happiness - a lesson sociologists are obviously happy to ignore.

    I still feel as though Veenhoven's numbers are pretty meaningless, even if he can tease consistency out of his respondents. Maybe they're a rough indicator that other sociologists should check out the lifestyles of Costa Ricans and Danes for any particularly awesome stuff going on.

    The policymaking shenanigans you're suggesting at the end of the post sound like they're straight from one of my favorite books, "Brave New World." When some global government starts engineering or "informing" people about the determinants of their own personal happiness, we will be perilously close to Huxley's dystopia.

    In fact, BNW is based on a society wherein every individual attained maximal happiness because he/she was biologically engineered to enjoy and excel in his/her role in society. To be honest, it sounds pretty awesome to me, but I have a feeling the real-world version won't be as good as the book version.

    ReplyDelete
  4. Correction- Biggie also said "Cash rules everything around me" in Notorious Thugs in 1997. Noob.

    ReplyDelete

Comments will be moderated for civility and relevance