jack

Do we have free will?

Forget your preconceptions. If you ask, "what does it look like to have free will" or "what does it look like NOT to have free will", it rapidly becomes a lot more obvious what you actually mean by that question.

What does it mean to NOT have free will? Well, think of examples:

* Someone asks you to pick a number between 1 and 10. You felt free to choose any number! But it unbeknownst to you beforehand, it turned out that you were more likely to pick 7 than any other number
* You are trying to concentrate on something. But there's some delicious food / a person you're fond of / a persistent worry in your brain which keeps stopping you
* You're drunk, tired, too young, or otherwise impaired, and can't fully understand the options and have to choose between them based on a very limited understanding

All of those where you're TRYING to do something, but something stops you. Often something INSIDE YOUR BRAIN. And I mean, that bit is still you. "Being bad at choosing random numbers" is just as much a part of you as anything else.

I used to stop the argument there. That's who you ARE. It's still that 'you' choose. If you don't like that you're not a pure abstract reasoning machine, that's not a "don't have free will" problem.

I used to jokingly phrase it as a syllogism

Premise: You are actions are predicted by deterministic rules.
Premise: You control your actions
Conclusion: You ARE deterministic rules.

But that's not actually the whole story. Sometimes those "something in your brain stops you" happens more than others. Choosing to do something else instead of eating when you're hungry is HARD, but you usually can if you try hard enough. Some things are harder to suppress. And you can't choose to "not sleep" or "not breathe" by strength of will however much you try.

Sometimes more free will, sometimes less.

What about all those OTHER times. Everyone knows that sometimes something overrides your attempt to decide something. But the rest of the time, when it feels like you have a free choice, do you really?

Well, there's some other exceptions. But... we know a lot of the time, your brain is running on heuristics that produce weird answers like "the first choice offered" or "the choice most like what you remember from childhood". Even when we don't know that it's half-assing the analytical reasoning, it probably is some of the time. That's just who we are.

What about what when people say "no free will". Like, an implacable force forcing you to certain outcomes? Well, if you count the laws of physics, then yes, we already covered that. If you mean something else, then, "there's no way to tell, but probably not".

What about practical consequences? Well, yes, act like you have free will. And what about other people? Well, sometimes they do and sometimes they don't -- interact with them in ways that work, not ones that meet some theoretical standard of "fair".

A recent conversation about Defence against the Dark arts teachers made me realise I use "evil" in two different ways. Sometimes I mean, "doing something bad on purpose". Sometimes I mean, "doing harm to other people". The greatest harm is often done by people who are indifferent to it. But people who maliciously hurt others are awful is a special way.

A couple of the professors were very indifferent-evil. They didn't set out to hurt people, but they didn't see any of the awful things they did to people. Others were malicious-evil, they were killing people all over the place.

And of course, it's more complicated by that. Most people who cause harm by inattention SHOULD notice, and exist somewhere on a scale from "I'm 8 and I haven't broken away from the worldview I'm immersed in" to "I'm really really really wilfully ignorant, and I must be actively avoiding thinking about this."

But insofar as it's helpful to be able to think about bad things, it's useful to realise that they often overlap, but when I say "very evil" I might mean one of two different things.

Until recently, I tended to follow an automatically utilitarian approach. After all, people would say, if one action does more good than another, how can it possibly be true than the second is the "right" one?

But it started to bother me, and I started to think maybe you needed both something like utilitarianism and something like virtue ethics.

I imagined using the same language to describe different ways of getting something right that’s somewhat less subjective and emotional than morality, say, the right way to write reliable computer programs.

Some people would say “here is a list of rules, follow them”. But this approach sucks when the technology and understanding get better, because you keep writing code that saves two bytes if it’s compiled on a 1990's 8-bit home computer, because that was a good idea at the time.

Other people would say, “choose whichever outcome will make the code more reliable in the long run”. That’s better than “follow these outdated rules”, but doesn't really tell you what to do hour-by-hour.

In fact, the people who do best seem to be those who have general principles they stick to, like “fix the security holes first”, even if it’s impossible to truly estimate the relative plusses and minusses of avoiding an eventual security breach versus adding a necessary new feature now. But they don’t stick to them blindly, and are willing to update or bypass those principles if the opposite is obviously better in some particular situation.

My thought, which I’m not yet very sure of, is the same applies to morality.

Diax's Rake

Diax's Rake says "Don't believe something simply because you want it to be true". It's from Anathem -- I'm not sure if there's a real-world name?

It sounds obvious, but in fact I keep coming across it in contexts where I hadn't realised "believing in something because I wanted to" was what people were doing.

For instance, the most common argument that there's an absolute standard of morality seems to be "But if we didn't, it would be really terrible. Blah blah blah Hitler." But that seems to be an argument for why it's not desirable to live in that world, but it offers no reason other than sheer optimism to think that we do live in that world.

But another case seems to be free will. Why do people think we have free will. It seems like the most common argument is "But if we didn't, it would be terrible! Our lives would be pointless, and we wouldn't be able to philosophically justify prison sentences." But again, that seems to be "we WANT to have free will", not "here's a reason to think it's LIKELY we have free will".

Free Will

However, that's somewhat misleading. I feel like at some point society started toying with the idea that "have free will" or "not have free will" made no seriously falsifiable assertions, even in principle.

At which point, some people said "Look, our future actions are basically predetermined by the physics of our minds. 'free will' is basically a meaningless concept."

And other said, "No, wait. Look at what we associate with 'free will': rights, responsibilities, choices, law, etc, etc. We do have all that, we don't care if it's predetermined or not. I think 'not having free will' is basically a meaningless concept."

And the thing is, THEY'RE BOTH RIGHT. "free will" being meaningless and "not having free will" being meaningless are exactly the same statement, they just SOUND like they're opposed. They're somewhat opposed: they agree how the world works, but disagree whether "free will" is an appropriate description to use to describe it.

And arguing about "should we use this word or not" is almost always pointless, with people regressing to assuming that they're still arguing for the concept they used to assocaite with the word, without recognising that the other people don't disagree, they're just doing the same thing.

Many people who know more about philosophy than me seem to be self-defining as compatibilists (the idea that free will and determinism aren't contradictory?) If someone says they're compatibilist, I generally find I completely agree with how they say the universe works. But I don't understand the assertion that free will exists. Is there a basis for that? It's not just pandering to people who have a really intense intuition that free will is a well-defined concept that exists, at the expense of alienating people who at some point because convinced it doesn't?

Something I forgot to add in the previous post is that a very similar question applies to time travel.

Most stories[1] dealing with time travel implicitly take a stance on history being deterministic (either a single world track, or a steady loop in a multiverse of world-tracks), or being changeable.

But the narrative typically only follows one character through one worldtrack. Precisely because, if you start asking questions like "if you change history, what happens to all the people who were already there" and "can a narrative jump about between worldtracks", your brain will melt with indecision. It's almost the same question as having the ability to make multiple virtual copies; if there are lots, then which one is "the" one the story follows?

[1] Stories being the best proxies we have for "how we would deal with blah in real life"

The other day I drew an analogy which at the time felt very silly, but in retrospect I like a lot more. Many people have attempted to define a rational ethical system of some sort -- preferably one which stems from self-evident premises, but at the least, one which is consistent, and based on reasonably plausible premises. Which I would agree would be ideal, but given that none of the proposed systems have worked without advising massively unworkable or repugnant outcomes, I'm pessimistic there is a perfect system to be found.

It was implicit in this approach that you were attempt to decide an answer to questions of the form "given a choice between A and B (or A and inaction) which one should you choose?" And yes, there are definitely further questions to ask, but telling you how to act is a lot of what an ethical system is.

Then I drew the analogy. When google first became the pre-eminent search engine, someone commented that the problem they had solved was not finding which pages were relevant to the search terms (which had already yielded considerably to prior human effort), but to find which top ten of the potentially relevant pages were actually most useful.

Likewise, there are vast swathes of area where most popular systems of ethics would agree. And yes, there will be endless fiddly corner cases involving hypothetical moral dilemmas about people tied to trolley-card tracks where people will disagree -- and perhaps always disagree. The question to me is, what problem do we face more in everyday life: choosing between a few clear alternatives which solution is right? Or prioritising, out of millions of potential actions which are all good, which is most urgent?

The first is more dramatic. "Should I protect my family, or uphold the law?" But day-to-day the second is probably more common. But seeing it as an ethical dilemma may be actively unhelpful. If so, you're inclined to view any choice solely in terms of the opportunity-cost, discounting the effort spent choosing. Of 3,000,000 good things you might potentially do, if you do the 2000th best, you're WRONG because doing the 1st best would have a greater utility, even if only by a little. And yet, comparing utilities is not always guaranteed to work (?) Several modified ethical systems have tried to weight things to compensate for this.

But if you think of it as a todo-list, it immediately sounds more reasonable. Of these, things, which do you want to do first? You obviously don't want to ONLY do things at the bottom of the list. But if you do a few at the top, and a few at the bottom that are convenient to you, and so on, it's hard to argue with that. You've thrown away the internal and external expectation that you will do the perfect thing, freeing you to try to do good things as fast as you're able.

There are many moral dilemmas of the form "given this unpalatable choice, how would you rate the choices?" Although I'm blessed with rarely facing such hard choices in real life (which does tend to give you a more pragmatic and less idealistic view) I find it interesting to track how my answers change over time.

A typical example would be: "A group of people from your village are hiding from insurgents who will slaughter you all if you're found. You're caring for a patient who has just become delirious and you can't shut them up and their cries are going to bring the soldiers. Do you keep him as quiet as possible and hope the soldiers somehow overlook the group? Or smother him and hope he somehow survives?"

There are many variations on this theme, for instance:

* Are you in a position of authority over the patient?
* Are you in a position of authority over the group?
* Is the dilemma horrendously contrived? (eg. "do you push person X onto the train track to stop a train full of people going over a cliff")
* Are the outcomes presented as certainties or high probabilities?
* Is the patient a volunteer or a soldier? An adult? A child? A baby?
* Does the patient stand a high chance of survival if you're caught? Any chance?

My responses now would generally be:

1. Death is bad.

( Read more... )

If someone proposes an abstract argument with apparently impeccable premises and yet a difficult-to-accept conclusion (eg. Pascal's wager), it may be interesting for two reasons.

(1) There may be a serious chance it's true and you need to decide if it's valid or not, and whether or not you need to start accepting the conclusion.

(2) You may be pretty sure the conclusion is false, because you have watertight reasons you trust more than this argument however superficially plausible, but you can't understand WHY this argument is wrong, and want to do so, in order to rebut similar arguments made by yourself and others in the future.

It's normally clear which category something falls into, although we often don't explicitly say that. (1) is probably more important, if it involves ACTUALLY changing your mind about something, but saying "that's false" is not a reason to stop thinking about it if you can't explain WHY it's false.

For instance, in a recent discussion on free will, I realised that I was reasonably sure that (a) what choices I make in the future are inherent in the current state of the universe and my brain and (b) in general we should definitely ACT as if we have free will and it will Just Work. That pretty much settles the practical questions in my mind. But the questions of how to deal with what everyone believes about it, and what everyone feels about it, are still very much there.

Preamble

At some point I came to the realisation that there is a distinction between statements about how the world IS (facts which can be decided by evidence and reason) and statements about how the world should be, about aims and ethics, (which can be refined and studied with evidence and reason, but ultimately have to stem from somewhere else).[1]

This is an important distinction because if someone (or you yourself) says "X is clearly wrong", and attempts to justify it solely in terms of observation about facts, then you can be fairly sure that they either made a fallacy or introduced a hidden moral assumption somewhere -- often a FALSE one. It sounds obvious put like that, but people have spent thousands of years making that sort of statement.

Digression

There is a grey area concerning statistical facts and rules of thumb, eg. deciding if economic strategy X will be a good thing for the people of a country. That is, in principle, a factual question, but it may not actually be decidable on the evidence available (either because the evidence is too hard to gather, or it's so unlikely that it's not worth spending the time to analyse it in full), which means people have to take a provisional opinion about the best strategy, which is likely to be a mix of generalisations from the evidence that IS available, and of different priorities in terms of the desirability of the likely results.

However, I don't think that's conceptually a separate category. It's that factual questions are themselves very murky, and we spend most of our time operating with reasonable certainty, not absolute certainty, whether morality has anything to do with it or not.

Realisation

However, it now seems to me there IS an overlap. A set of statements which are somewhat purely factual, yet also somewhat faith/morality based.

For instance, statements that might fall into this category, and I generally agree with:

-Advancing human knowledge is of benefit both for me and for society in general
-Cooperation with other humans is of benefit both for me and for society in general
-Honesty is of benefit both for me and for society in general

And statements that I may not agree with:

- Royalty and nobility have an obligation to rule well
- God exists, and is the source of morality

In all cases I think people would say that, in principle, the statement is subject to factual testing. In principle, we might discover that, say, actually increasing human knowledge makes our lives worse in most ways. However, we have a large emotional commitment to holding to the belief as long as it's at all possible to do so.

This is interesting, because for purely factual questions, I would say it's incorrect to hold to them in the face of convincing evidence (it's often correct to hold to them provisionally while considering the evidence, but not to fight the evidence). People often do this, and I generally consider it the sign of a weak argument (or at least, an argument that needs a lot more work).

To put it in other terms, holding to the belief in the face of evidence fails the objectivity test of asking "what evidence WOULD it take to convince you?" as the response is often "I can't imagine anything that would", which is an indication that even someone SAYS its a factual belief, they're actually holding it on faith.

However, with the examples I described, I have to say, I do have to admit I would change my mind faced with OVERWHELMING evidence, but also I would cling to them as long as possible. And, even though it goes against my instincts for "how to be objective", I think that's the right thing to do.

Postamble

Realising I do have beliefs I cling to against evidence makes me much more sympathetic to other people who do so: with my previous belief structure I would basically have dismissed the idea.

On the other hand, there is still the question of how to choose such beliefs. Pure morality beliefs, like "murder is bad" and "harm is bad" I think most people agree on, even if they assign different weights to them. But with both, there is the question of how to decide on them.

To some extent they are plucked straight from our underlying drives. Why is murder bad? Because we instinctively think it is (because evolution or God made us that way, or because society trained us to). But which of these do we accept, and which do we dismiss as spurious? Most people accept harming other humans is bad. Most people accept harming stuffed toys with big eyes produces a negative emotional reaction, but isn't inherently bad. People disagree about humans from other tribes, about animals, about aliens. Is there a way of deciding, or do we have to accept that we were given no objective guide, and must adopt one on faith?

Footnotes

[1] And, um, some fundamental assumptions like "logic works" and "Occam's razor" which are arguably factual, yet also faith-based. I'm not sure how that fits into the system.

Arguing duality with Rachel's brother.