JCC If you don't believe in free will, you are bad and should feel bad

y'all need jesus

 

Hey I have nothing against Plato.

 

GrandAdmiralJello, Saintheart: Sorry I've fallen behind in this thread. I'll try to jump back in now. I don't think I quite made my point with the MacBeth discussion. Or rather, it go too particular to that character, and wasn't about the general case.

You've both made the point that it's not "legitimately" heroic to do something that appears risky if the outcome is in fact known in advance, and no true risk was involved. I can accept that logic. But I think there is another aspect to fatalism you haven't addressed yet. Just as one can be destined to succeed at something, one can also be destined to fail. So how do you account for the actions of a person that still feels compelled to try their hardest, even in a situation that they feel it is impossible to alter the outcome? If there is literally no benefit to investing more energy in something, why do they do it? I would argue that this sort of perseverance in the face of futility is a kind of bravery in its own right.

I guess it goes back to Jello's discussion about whether it should be considered a character trait or a quality of specific acts done by an individual. If it is a character trait, then I would think a brave person would tend to be "brave" in all circumstances, even ones where the outcome is foreordained against him, and it has no point. In fact, these instances might be the only one that help you distinguish truly brave people, as otherwise there are clear incentives to attempting bravery so as to alter the outcome.

Is this reasonable? How do you guys view these kinds of situations?

 

Ah, you sound like Camus in "The Myth of Sisyphus." "The struggle itself towards the heights is enough to fill a man's heart." I tend to agree with you.

 

Nice thread. Some of my thoughts regarding free will:
I'm pretty open minded and flexible regarding the meaning of free will. I might agree with some definitions but not others. Mainly I think it exists, after a fashion, in a very limited and cynical sense, sort of.

My main problems with it relate to the logic of spacetime, the evidence of neuroscience, and other stuff too complicated for me to understand but for some reason I don't feel hindered in trying to describe. Briefly then,

• spacetime - people are apparently unable to change or influence events in the past. It's as if the past is fixed. The present is future past. The future will be future past. But it really looks like we can influence the present. What's THAT all about?
• neurosciency stuff - it's pretty conclusive that our subconscious determines the outcome of our self perceived conscious decisions well ahead of time. It's probably still US making those decisions (just not the 'us' we imagine it to be), and we can't deny that our decisions are better than chance (otherwise we wouldn't be here, etc) so it really looks like we have some power to reason and predict... But besides the problem of the inflexibility of time, there's the stuff that I have problems defining...
• wibble - we could design a robot with a finite number of preprogrammed responses to stimuli that, given a sufficiently large table of responses and a better-than-chance level stochastic algorithm, would appear to exhibit behaviour indistinguishable to a creature with free will.
• wobble - evolution could easily have done that already. We know we make models that we fool ourselves are real (often knowingly), and we're probably stuck this way for the time being. Unless we're not. I don't know.

As to belief in free will relating to how charitable or nice you are, pfft.

 

Actually, no you couldn't. The most theoretically sophisticated non-deterministic Turing machine possible would still have a finite number of total states and couldn't respond to every possible situation the way a person theoretically could. You could create an incredibly sophisticated illusion but there would always be a gotcha stimulus.

 

why? do we have an infinite number of total states? do we know that for sure?

 

The number of possible connections in a brain must be finite. Bloody huge, but finite.

 

im not taking a side, btw, i just legit dont know anything about this topic.

im just asking questions //glenn beck

 

Okay, but the brain isn't necessarily mappable to a Turing machine the way a computer, even an NTM computer, is. It could be, in which case I would concede one could make such a free will simulating machine (Of necessity because "any brain process" would then be computable), but that's an open question.

 

turing machines are gay

 

Turing was prosecuted for homosexuality in 1952, when such acts were still criminalised in the UK. He accepted treatment with female hormones (chemical castration) as an alternative to prison. Turing died in 1954, 16 days before his 42nd birthday, from cyanide poisoning. An inquest determined his death a suicide; his mother and some others believed it was accidental.[7]On 10 September 2009, following an Internet campaign, British Prime Minister Gordon Brown made an official public apology on behalf of the British government for "the appalling way he was treated."
In May 2012, a private member's bill was put before the House of Lords to grant Turing a statutory pardon.[8] In July 2013, it gained government support;[9] however, instead of calling for the second reading of the bill through the House of Commons,[10] the government opted for a posthumous pardon under the royal prerogative of mercy, which was signed on 24 December 2013 with immediate effect.[11][12]
christ thats effed

 

Strikes me as suitably meta given the thread topic.

 

A loooong time ago I did some reading about AI, and there was this theory that a brain could be approximated by certain kinds of self-referential logic loops . . . logic sentences that "know" about their own existence. These things are supposed to write themselves into existence using a few simple rules, much like a fractal "grows itself." Consciousness is said to be an emergent property of a sufficiently-complex collection of logic loops.

I remember thinking at the time that this was interesting, but how would we be sure that whatever "emergent properties" such a system might develop would resemble human cognition in the slightest? What if we accidentally created dolphin consciousness? Or insane axe murderer consciousness?

Sorry, Ramza, if I'm talking out my ass about this.

 

it was more i just kind of knew in passing that turing was gay and posted a joke about it and then i wikied for confirmation/more info and found out he had been basically badgered to death over it and got sad so i made another post

but jello i think you're just sore that the british government did something bad and later had to admit it and symbolically reverse themselves

 

I did say I wasn't very good at articulating my ideas. Turing machines are your business, I'm trying to describe a table of values/instructions/reflexes with a master program that selects strings of them in response and in anticipation of events/stimuli. You could describe simple life forms in this way, and I don't see why you can't describe ever more complex life forms in this way. Consciousness (or the illusion of it) could be a by-product of a sufficiently complicated and flexible selection system. Like a virtual second opinion, or second guess protocol, or... ..something.

 

wait how does one accidentally ingest cyanide?

 

Eating a lot of decomposing corpses.

 

jeffrey dahmer's disease

 

You are describing a non-deterministic Turing machine. As it stands, right now, it would not work. Again, I concede that if the brain is Turing mappable it could be made to work, but that question is open.

This... could work but I think the computer scientists that would've been making such a speculation are being a bit cocksure. Mostly because we don't yet know if it's possible to construct a logical language that would allow self-reference in a non-paradoxical form (It starts to get really easy to invoke a Gödel sequence when you get too wonky and that leads to hilarious but extraordinarily nontrivial problems.) The answer you get depends on the researcher you ask - Stanford is saying "Yes," MIT is saying "No," logicians are saying "Shouldn't we make sure quantifiers aren't deterministic before trying to worry about self-recognition?" and Hollywood is saying "Well I suppose it did make 2.3 times its budget at the box office, we could do a remake."

 

Ramza
I don't know what 'Turing mappable' means exactly, but I'm guessing it means 'possible to sample and emulate' in some way. So I guess you're agreeing with me but acting all clever or summink.

Don't you agree that a brain is finite and not boundless? There's a physical limit to the number, size, detail and frequency of calculations, a limited 'space' allocated to memory storage, models of reality must be limited in complexity by the size and efficiency of a brain, the range and value of the sensory input, and the effectiveness of interpretation of sensory data. We're bound by that, we cannot be beings of infinite potential or any of that Chopra woo.

There must be a finite number of bits required to realistically sample and emulate a brain, because the brain is made of a finite volume of material that only does a finite variety of stuff. Find the brain scanning equivalent of the Nyquist rate and all the detail you'd want would become available, assuming you had the means to scan and correctly interpret it.

It doesn't have to be a human brain, unless your definition of free will wont extend that far. Any decision making, problem solving creature would do. An insect brain might be a lot simpler to emulate. And it doesn't even have to be a perfect emulation to make it indistinguishable from an insect with 'real' free will. Given how many dodgy hacks and engineering disasters evolution throws up, I'd imagine an intelligent designer could realise a more efficient code or mechanism to achieve the equivalent appearance of free will.

I remember a hippy hypothesis I once had, that what we imagine as free will/consciousness is like us navigating the multiverse of fixed (but varied) spacetime events. Maaan. But that's a vaguely similar proposition to what I'm saying now, that we are limited by circumstance and can only have a finite number of real/realistic choices to imagine we choose from. We're fixed but not necessarily fated (within reason). Probably.

 

Not possible.

 

No, see, this is precisely why I keep using the phrase Turing mappable, because it's independent of finiteness. A Turing machine is a theoretical data construct used to determine whether or not an algorithm produces something called a computable sequence, handy because it works in a pretty general context. Turing mappable means that you can take the space of algorithms and stick them to Turing machines. It doesn't have a lot to do with sampling and emulation so much as comparison, as they're a proof of concept thing, though if the space isn't Turing mappable you're hosed in terms of creating a perfect computational reconstruction, so as a loose frame of reference that works.

And now, to reiterate - finiteness does not imply Turing computability, and hence the brain having finite dimension does not imply the brain's space of all possible algorithms is Turing mappable. It could be, but we don't know. I'm not invoking some mystical shlock, here, I'm talking about an active field of mathematical, logical, and theoretical computer scientific research, and the fact that we can dream up sequences that do not satisfy Turing computability requirements (Basic propositions like "Is statement X provable or not?" is the classic example, since it's what Turing machines were invented for) in and of itself poses a tremendous obstacle to ever answering the question. And as your proposed device is a non-deterministic Turing machine, and hence well-known to be bounded and Turing mappable, then as everything stands, right now, nobody knows for sure, but I suspect the answer is no due to how wonky everything can get.