Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

Marius comments on Not for the Sake of Happiness (Alone) - Less Wrong

49 Post author: Eliezer_Yudkowsky 22 November 2007 03:19AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (94)

Sort By: Old

You are viewing a single comment's thread. Show more comments above.

Comment author: Marius 27 March 2011 09:13:03PM *  1 point [-]

You misunderstood the first point. I did not claim you succeed at tasks you are good at. I claimed that if you define desire by "what you do", and simultaneously believe that "satisfying your desires -> happiness", then succeeding at the tasks you attempt would cause happiness. Yet that is an incomplete descriptor of happiness.

Additionally, I obviously agree people have competing desires. But this makes it impossible to use "what I did" as a measurement of "what I want". For instance, if I want to run but don't, it may be due to laziness (which is hardly a "desire for slack"), fear (which is not merely a "desire to avoid risk or embarrassment"), etc.

Your lottery description is inconsistent with other accomplishments and pleasures. For instance, people who marry [the right person] do not simply become habituated to the new pleasures and establish a new baseline. People with good or bad jobs do not become entirely habituated to those jobs - they derive happiness and unhappiness from them every day. The lottery is a different story from these, and you'll need to come up with a better explanation as to why it is different. My explanation is that we derive happiness from earning success, but not from being given it arbitrarily, and that regardless of one's desires human nature tends to behave that way.

This is my first counterexample to your puzzle: regardless of whether one has a desire to have to earn success (and most people desire not to have to earn it), we are made happy by earning success. Other examples: we are made happy by hard work (even unsuccessful hard work), by being punished when we deserve it, by putting on a smile (even against one's will), and by many other things we don't desire and some that we try to avoid.

Comment author: Polymeron 27 March 2011 10:08:11PM *  2 points [-]

Thank you; you've made some very good points that deserve a proper reply. However it's getting late here and I will need more energy go over this properly. I'll definitely consider this.

As a quick opener, because I think there's an open point here: It seems to me that all emotions serve as behavioral feedback mechanisms. But even if I am mistaken on that, and/or happiness is not desire fulfillment feedback, what would you think its evolutionary role is? It's clearly not an arbitrary component. Not to make the fallacy that any explanation is better than no explanation, I would nevertheless be interested in playing off this hypothesis against something other than a null model - a competing explanation. Can you offer one?

Comment author: Marius 28 March 2011 03:23:39AM 1 point [-]

I agree that emotions do serve as behavioral feedback mechanisms, but that's not all they do. They have complex social roles, among other things, including signaling, promotion of trust, promotion of empathy, etc. This social role is probably just as important in the case of happiness as the marker of "needs satisfied". In the case of grief, the social role is probably far more important than any feedback role. In addition to these roles, happiness contains an element of contentedness: "you are at a local maximum, and would be better off staying at this local maximum than risking matters to satisfy more needs". Thus, many slaves are content until they see the chance at freedom. There is a joy in great/beautiful/religious things that science currently lacks a good explanation for. There may be many other roles for happiness, as well.

Comment author: Polymeron 28 March 2011 04:02:31PM 1 point [-]

I have to agree that happiness (and other emotions) have come to have a strong signaling component. I'm now even more interested than before about the mechanism by which it operates - just what triggers this emotion. I've also been thinking quite a bit about grief, which didn't fit as a pure feedback mechanism (otherwise you'd expect to have the same emotion for a person going away for life and that person dying), and your comments on that finally drove the point home.

I will need to consider all this further and revise my hypothesis. Thanks again for the insight!