The philosophy thread general discussion

It is if you want it to be free. You are describing a fully determined will, which is not the same thing.

Of course of will is determine. Your action are suppose to be determined by your will. And as I have explained they are determined by your will. What do you want that I haven't given?

When life gives you lemons, burn life's house down with the lemons.

If the will is itself determined, its actions will also be determined - so even though the subject will experience it as willing something, everything is still determined, with the processes generating the experience of the mind as just another cog in the machine, no freer than the rest.

You could say that the mind can want what it wants, but it can't choose what it wants.

Quod gratis asseritur, gratis negatur.

@Best of

Knowing a minds origin can suggest to us their preferences but, it can't tell us what they will do.

As for not being to able to decide what they want, they can do that after they already exist based on their original preferences. A mind that doesn't care about anything never will so preferences have to be build in.

When life gives you lemons, burn life's house down with the lemons.

If you're arguing what I think you're arguing—regardless of whether or not you're just playing devil's advocate—then I'm going to have to say that I disagree.

Even if the human mind is mechanical in nature, there's nothing we can do to escape from that. Choosing to not punish those who committed crimes is just as much a meaningless choice as the alternative.

To put it another way, even if you were to learn that you're the only sentient intelligence in existence and everyone else is just your illusion, you still have to live in the world. The Matrix might not be real, but it's inhabitants still have to get up in the morning.

I don't really know what a, "Free will," is, but even if choice is an illusion, we still might as well act like it isn't, because we're subject to that illusion, and no decision we make will free us from it.

Fire, air, water, earth...legend has it that when these four elements are gathered, they will form the fifth element...boron.

So you're saying that even if we knew absolutely everything about someone's brains and everything around them, we wouldn't be able to predict their behaviour based on our understanding of how their brain works and what input is given to the brain?

Because I that's what you're saying I'll disagree. It might be that in practice our models and experiments will never be good enough to get sufficiently accurate information to get exact predictions, but in principle it should be possible to predict what someone will do if we have a complete knowledge of the inner workings of their brains (and of incoming inputs to the extent that they're relevant to their next action.)

Quod gratis asseritur, gratis negatur.

@Best of

That was not what I meant. Knowing a persons genes and upbringing can tell us what sort of person they are but, we can't directly from that predict their actions without computing their brains.

When life gives you lemons, burn life's house down with the lemons.

But if we did compute their brains - well, that might not be the best word for it but you probably know what I mean - would you still say we can't predict what they'll do in a given situation?

Quod gratis asseritur, gratis negatur.

Basically to predict a person perfectly you would have to run the program. EDIT You're not really predicting them, so much as make them make the decision sooner.

Also the free part is helped by the fact that you do consider multiple options. And it is that consideration that determines which one you try to do.

Note that technically this means that there are robots with free will. Least then us, and they don't care about their freedom, but they have a little.

In particular a chess program only thinks about chess. It doesn't think about itself or its preferences, or its "thought" process. There are thing we can't think about either, at least ones that are too complex. While we can think about ourselves, and think about ourselves thinking about ourselves, we can't do it well pass some point, and past another point not at all. Our preferences are also black boxed, along with a lot of our other thought processes. And our ability to rewrite them a limited.

edited 25th Feb '14 3:32:27 PM by supermerlin100

You think a chess program has free will?

You can't predict what a chess program will do just from its source code. So yes it does have free will though it doesn't have the same level of agency as human.

When life gives you lemons, burn life's house down with the lemons.

That's a very strange definition of "free will." Why couldn't you predict the next move if you knew the source code and the position on the board? The search algorithm it uses isn't that complicated. You could carry it out by hand, it would just take a lot longer.

@De Marquis

That is the point. You have to run the process to get the result. Past->present->future you can't skip the middle step.

When life gives you lemons, burn life's house down with the lemons.

Are humans any different? Given a certain stimulus, you will react in a certain way. Granted, simulating this is presently beyond us, but we haven't found anything suggesting that our thoughts are random (Which would be pretty scary, when you think about it—there'd be nothing preventing someone from deciding to go on a murder spree. Having your decisions limited by your experiences is nice.)

edited 25th Feb '14 6:00:55 PM by Rem

Fire, air, water, earth...legend has it that when these four elements are gathered, they will form the fifth element...boron.

@High: OK, but if this process you have to run is a chain of cause and effect that originates with inputs from outside the mind of the person making the decision, then there's no freedom in it.

@Rem: Yes, that is what we are debating.

@De Marquis If you create a mind that has certain preferences, that doesn't violate their freewill. Their will is only violate if you control their actions directly bypassing their decision process.

To use the example of the chess program. If you program it to make a certain move then that is violating its "will". If you build an chess algorithm that makes a move base on some standard then it has a will. You can not predict the chess ai's without running its algorithm which determines the move it will make.

When life gives you lemons, burn life's house down with the lemons.

Again, that's a weird definition of freewill. If the algorithm is the only factor driving decision making, and I know the algorithm, and I know what the inputs to the algorithm could be, then I can predict what they will decide before they decide it, and in that case their decision seems completely determined to me. Am I missing something?

@De Marquis I would describe that as running another instance of the algorithm in advance. Like going back in time after seeing someones decision and not doing anything. Their decision still follow from there will and they won't know what decision they will make until they actually think it through.

When life gives you lemons, burn life's house down with the lemons.

The point is that your 'prediction' is necessarily so complex that it's indistinguishable from the real thing. So it's not a prediction anymore. You're not predicting someone, you're making a copy of them and then observing what the copy does. The copy still has to do the thing before you can know what choice it made.

edited 25th Feb '14 10:26:28 PM by Meklar

Join my forum game!

Yes that is what I meant.

When life gives you lemons, burn life's house down with the lemons.

I suppose you're free to postulate that it is not practical to predict the behaviour of a sufficiently complex organism, but who knows what sort of computing power we'll have in the future...

Quod gratis asseritur, gratis negatur.

@Best of

To predict something means making a simpler version of it. A similar enough model just is the thing. I never said that you can't predict a complex being but, rather that the simulation is them making the choice.

When life gives you lemons, burn life's house down with the lemons.

"You're not predicting someone, you're making a copy of them and then observing what the copy does. The copy still has to do the thing before you can know what choice it made."

I get it. A non-linear system is unpredictable for any simulation less complex than the system itself. That's why we dont already have sentient AI. But I dont need a perfect prediction to remove the possibility of a free will, I only need to show that the decision was fully determined- and if two identical systems both make the same decision based on the same inputs, then that counts as "fully determined" to me. Again, I ask, where is the freedom in this system?

edited 26th Feb '14 7:37:25 PM by DeMarquis

@De Marquis What matters is that that the decision is determined by the persons will. It is true that they can only make one choice, but that choice is determined by their will. A identical system will make the same choice because it has the same will.

edited 26th Feb '14 7:48:02 PM by higurashimerlin

When life gives you lemons, burn life's house down with the lemons.

So lets call it "Fully Determined Will" instead of "Freewill" and call it a day.