Wednesday, August 9, 2017

Happiness Is Not A Coherent Concept

[Epistemic status: Banging wooden spoon on pots and pans and shouting loudly]

Every time the subject of "happiness" comes up in social discourse around me—every time—I get frustrated by the fact that the most important fact that's been discovered about it still hasn't been integrated into most people's views on the world.

Happiness isn't real.

Let me expand on that.

What does it mean to be "real"?

Social science is, on some level, a fundamentally silly undertaking: We know that brains are made up of atoms, not abstract concepts like "happiness" or "sadness" or "anger." Social science is the project of finding the most useful and parsimonious abstract concepts to represent human behavior, even if they don't correspond exactly with physical processes that we know to exist.

To do good science, you have to entertain the hypothesis that your social science construct that you made up in your head does not make sense as a way of representing human behavior. You can't just assume that it's a reasonable concept because you thought of it—you have to rule out alternatives explicitly. To say that a construct is "real" means that you have ruled out alternatives, and found that it represents reality well.

You should apply similar standards to existing cultural concepts, not just things that you made up in your head. Scientific study examines "common sense" as well as surprising ideas, because there's a not-insignificant chance that any given piece of common sense is wrong.

So how do you check if your concept is the a highly parsimonious, predictive way of describing humans?

You come up with different ways of measuring it. You measure it on lots of different people. You check how the measurements correspond with each other, and how much they vary from person to person, and what other factors they're correlated with.

If your measurements correlate with each other, then that's a first sign that they're at least measuring related things, or perhaps measuring the same thing different ways.

If a thing can be measured several different ways, and a causal factor can push one in a direction but not the other, then you start to worry that the thing is not actually one thing, but several things.

This has happened with the concept of "happiness."

Results in "happiness" research

For example, if you are someone who values having a lot of money, and you are asked what your life satisfaction is ("How satisfied are you with your life overall, these days?"), it will be higher if you have more money, even in the extremely high ranges. However, having more money after a point does not affect your day-to-day happiness.

This suggests that life satisfaction and day-to-day happiness are not just different ways of measuring the same thing. They are measuring different underlying things, period.

Similar results can be seen when we measure remembered happiness vs. how happy someone felt during an experience: they seem to be fundamentally different things.

By any reasonable social science standard, happiness has been explicitly shown not to be a single node in thing-space that can be reliably measured. In that sense, it is not real.

Life satisfaction is real. Day-to-day pleasure is real. Remembered happiness is real, and experienced happiness is real. These things are measurable. They are often correlated. But they are not the same. "Happiness" is a broad category that can include all these things, and thus can't be measured or even pointed to very well.

(None of these results are particularly new. Kahneman devotes a substantial section of Thinking, Fast and Slow to describing different types of happiness and how they were established in the research literature.)

Why should I care?

This has implications not only for how we talk about happiness, but also for how we plan and live our lives.

Since different types of happiness are incommensurable, we can't substitute one for the other. You need to think about how you weigh life satisfaction against day-to-day happiness, in order to trade them off against each other. It doesn't make sense to make your life as pleasurable as possible without accounting for your overall satisfaction, and vice versa.

It also means that we need to be careful about increasing one type of "happiness" while accidentally decreasing another. Doing unpleasant activities sometimes brings more overall satisfaction; doing things that you don't enjoy that much in the moment may nonetheless result in happier memories over time.

This sounds like common sense, but it's sometimes forgotten when discussing research results summarized as "Money buys happiness, but only up to a point!", or "Weather doesn't make a difference to happiness!" When seeing such headlines, it's important to probe further: What exactly do you mean by "happiness"?

No comments :

Post a Comment