zondag 16 november 2014

For the Sake of Pleasure Alone

In the past few years I have made acquaintance with several people who, despite being both smart and educated, continue to worship the false god of preference utilitarianism.  I too consider myself both smart and educated, but worship hedonistic utilitarianism instead.  Sometimes I wonder if our differing views aren't merely different representations of the same thing.

I have been wanting to explore this for a long time, but never knew where to start.  Today, less is more, and I will just take an excerpt from Not for the Sake of Happiness (Alone) and go from there:
It is an undeniable fact that we tend to do things that make us happy, but this doesn't mean we should regard the happiness as the only reason for so acting.  First, this would make it difficult to explain how we could care about anyone else's happiness - how we could treat people as ends in themselves, rather than instrumental means of obtaining a warm glow of satisfaction.

Second, just because something is a consequence of my action doesn't mean it was the sole justification.  If I'm writing a blog post, and I get a headache, I may take an ibuprofen.  One of the consequences of my action is that I experience less pain, but this doesn't mean it was the only consequence, or even the most important reason for my decision.  I do value the state of not having a headache.  But I can value something for its own sake and also value it as a means to an end.

For all value to be reducible to happiness, it's not enough to show that happiness is involved in most of our decisions - it's not even enough to show that happiness is the most important consequent in all of our decisions - it must be the only consequent.  That's a tough standard to meet.
Yes, there are multiple reasons for taking the ibuprofen, and it would be a stretch to say that the other reasons boil down to maximizing expected happiness.  But just because they are reasons for acting one way or another does not mean there is something valuable there.  This strikes me as a confusion of terminal and instrumental values.

I can spend hours and hours aimlessly clicking around on Facebook or aimlessly walking around the grocery store.  Anyone with a similar affliction can confirm that these activities yield more dolor than hedon; why then do I keep doing them?  Whatever the cause, I don't feel at all compelled to elevate it or any aspect of its outcomes to Terminal Value.

This feels like a strawman because writing a blog post has higher expected value than aimlessly clicking around on Facebook.  But if that's the crux, then it's a confusion over instrumental and terminal values.

The more substantial disagreement is that I don't see why our conception of terminal values should explain our behaviour.


In Terminal Values and Instrumental Values, Yudkowsky writes:
Consider the philosopher who asserts, "All of us are ultimately selfish; we care only about our own states of mind.  The mother who claims to care about her son's welfare, really wants to believe that her son is doing well - this belief is what makes the mother happy.  She helps him for the sake of her own happiness, not his."  You say, "Well, suppose the mother sacrifices her life to push her son out of the path of an oncoming truck.  That's not going to make her happy, just dead."  The philosopher stammers for a few moments, then replies, "But she still did it because she valued that choice above others - because of the feeling of importance she attached to that decision."


When we object that people sometimes do sacrifice their lives, the philosopher's reply shifts to discussing Expected Utilities over Actions:  "The feeling of importance she attached to that decision."  This is a drastic jump that should make us leap out of our chairs in indignation.
And this is basically how I see it, except that I don't try to be cool and cynical and deep by making accusations of selfishness, and that my stammering wouldn't be relevant to the argument.  The mother is an adaptation-executor just like the rest of us; endowed with reinforcement-learned genes that make no distinction between expected utility and actual utility or between instrumental and terminal values.  Worse, her adaptations have been optimized with respect to an irrelevant goal: inclusive genetic fitness.

If the terminal value function we are looking for should explain our behaviour, then there you have it: inclusive genetic fitness.

(As I am writing this, it occurs to me that I may see it differently from the paraphrased philosopher after all.  I would not have to make the jump from talking about utility to talking about expected utility because I would be talking about expected utility from the start.)


Years ago when I first started reading Yudkowsky's writings, I came upon
How to Convince Me That 2 + 2 = 3:
I find it quite easy to imagine a situation which would convince me that 2 + 2 = 3.  Suppose I got up one morning, and took out two earplugs, and set them down next to two other earplugs on my nighttable, and noticed that there were now three earplugs, without any earplugs having appeared or disappeared [...].
(Note: I have omitted essential parts of his argument because I'm not going to argue with it anyway.)

I remember reading over that second sentence several times.  You have two earplugs, you add two earplugs, and you end up with three earplugs.  That's impossible!  What on Earth is he not telling me?  I think it took close to a dozen fruitless re-readings of the first few paragraphs of the post before it dawned on me.

It's unimaginable, and that's the point.  The hypothesis that two plus two equals four is tested possibly more often than any other, and has never had evidence -- no matter how unexpected -- cast doubt on it.  We simply can't see things being any other way.

And so I feel about pain and pleasure being the only terminally valuable things.  I can't even fathom how something could possibly be terminally valuable if it were neither pleasurable nor painful.  But I would know it if it happened, and then I would be convinced.


I feel like I should taboo "terminally valuable", but I am not sure what to replace it with.  I terminally value good sensations because they are good to me.  I terminally disvalue bad sensations because they are bad to me.

I could go on for a while, rephrasing it in different ways, all conveying simply "I like what I like and I dislike what I dislike".

That statement is tautological but not useless.

There is no objective morality.  Even if there were, there would be nothing to enforce it.  If there were something to enforce it, this whole discussion would be pointless.

On the flip side, there is plenty of subjective morality.  Different people like different things.  Different people want different things.

The nice thing about subjective morality is that it's nobody's business but your own.  But it's a leaky abstraction.  We share an objective reality, and what happens there violates most people's subjective moralities.

That much is obvious.

Now suppose we want to do something about this.  Ideally, we'd like for everyone to get what they want.  Failing that, we can make them want what they get.

That last option is one that Yudkowsky seems to be heavily opposed to.  If he wants chocolate ice cream, then he doesn't want to want vanilla ice cream, because he doesn't want vanilla ice cream.  He shudders at the thought of eating vanilla ice cream.  Eww!

But I hope that it is uncontroversial to say that there is nothing valuable about chocolate or vanilla ice cream per se.  The value comes from the experience of eating it and liking it.

There is no objective morality, and so you should consider your attachment to your current wants (whatever they are) to be instrumentally irrational.  In fact, your wants are worse than useless: there is an opportunity cost to insisting on chocolate ice cream when vanilla ice cream is much more abundant and your wants are easily changed.

The extent to which you insist on keeping your wants the way they are determines your... say... preferential footprint.  If, as I assumed earlier, we want for everyone to get what they want, then we want to minimize our preferential footprints.

Doing so will leave only that which we cannot help but value: pleasures and pains.  If it turns out we can do away with those as well, then great, but just like I don't expect two plus two to stop equalling three, so I don't expect pain to stop hurting.


So far I have assumed that we want to cooperate so that everyone gets what they want.  Given that the conception of value presented here is in some sense selfish, it is a whole 'nother minefield whether and how this cooperation can occur.  I believe I have written plenty of disagreeable things already; let's leave it at this for now.

Geen opmerkingen:

Een reactie posten