Friday, April 30, 2010

Muddled Thinking about Free Will

Over at his blog, in support of the existence of "free will", David Heddle says (in a comment):

If free will is an illusion, then deterrents are an illusion. How can a deterrent make me choose not to commit a crime, unless I have the facility of choice?

I think this is more muddled thinking about free will. I still don't know what free will is supposed to mean, exactly, and I don't think anyone else does, either.

But, ignoring this, let's address Heddle's claim. Could deterrents work on humans if they have no free will? I think the answer is clearly, yes. Let's pretend that humans are soulless computational machines, shaped by evolution, who act based on a very very very complicated algorithm that takes sensory impressions as inputs and produces actions as outputs. Let's say that this algorithm tends, generally speaking, to try to ensure survival and pleasure and reproduction of the individual. Now the human machine suddenly sees resources, for the taking, that belong to another. The human machine does a cost-benefit analysis to "decide" whether to take the resources or not. In the absence of a known deterrent, such as a dangerous dog or future incarceration, the human machine may decide to take the resource. In the presence of a deterrent, the human machine may make another decision.

How does this involve "free will"? You can call this decision-making "free will" if you like. But then a thermostat has free will, too.

It's not at all surprising to anyone who thinks about computation for a living that a complicated algorithm can result in different behavior based on different inputs. The mystery to me is why Heddle thinks this says anything about free will.

28 comments:

XiXiDu said...

I have put my thoughts on this subject together here:
http://spacecollective.org/XiXiDu/5759/Free-will-as-nonlinear-transformational-effectiveness

It includes, amongst other opinions, the take on free will according to a neuroscientist, a computer scientist, a AI researcher and philosophy in general.

I hope that roundup might be of use to someone.

Yob said...

As I recall, this question is often asked in philosophy classes, only the questions focus on the value of punishment (after the fact) in the absence of free will. After all, doesn't punishment assume that a person should, and could, have chosen a "better" course of action?

Shifting the discussion to focus on judicial action as deterrent completely changes things, as you rightly point out.

Valhar2000 said...

The mystery to me is why Heddle thinks this says anything about free will.

Because Jesus died for our sins, you silly boy!

Diogenes said...

Yes, the religious right makes utterly muddled, self-contradicting statements about free will. The other recent example being Rebecca Bynum's horrible article in New English Review in which she asserted that scientists are just like Islamic radical terrorists, because they both are "materialists" against free will who envision man as a slave.

The one consistent statement the religious right makes about free will is that whatever scientists they've put a contract hit on [evolutionists, neurologists, climatologists, quantum physicists etc.] are against free will, and out to destroy your freedom.

The bottom line is that the religious right asserts that human behavior can *only* be influenced by religious belief. If you're a Christian, you'll obey the law and love your fellow person. If you're an atheist or deist or Unitarian, you'll shoot up your school.

However, if you think human behavior could *also* be influenced by material conditions... economics... poverty... hunger... greed... capitalist competition... Oh no! If you think human behavior is influenced in any way by those things, you're a "materialist" and out to enslave mankind.

cody said...

Well said.

Ty said...

Ooooh, totally going to steal the thermostat comparison the next time I get in a free will debate.

Timothy said...

Ugh, bad logic on both Heddle's part and on the part of the original article's author.

"True free will, if it exists, is inherently supernatural. By its very definition it involves circumventing nature. The universe's differential equation is leading you to perform action A, but you rise up against nature's next time-step and choose B instead. "

Yes, the 'problem' is that if your choice is based on something natural, then it's determined by something natural, and humans don't like that idea. But Heddle just pushes the problem back one step further. If some supernatural mechanism allows you to choose B instead, what was that choice based on? Why did you choose B over A? Any reason you give is going to determine the output of the human machine, it's just that the determining factor will now be supernatural rather than natural. Whoop-dee-doo, we're right back where we started.

Since we're on the subject of free will, I was wondering, Dr. Shallit, what your opinion is of the idea some people are espousing that certain quantum events are uncaused/undetermined by anything. Not that I think the answer contributes anything to the free will issue, but people often bring the two up in association. I find the idea of non-deterministic events to be completely unfathomable, and I'm wondering what your take is on the matter.

Frank said...

"You can call this decision-making "free will" if you like."

Who says it's decision-making in the first place? Maybe we just think that we're making decisions.

Paul C. Anagnostopoulos said...

Let's make sure we say we are talking about libertarian free will when we say the concept is incoherent. There are various compatibilist free will concepts that make some amount of sense.

~~ Paul

Gingerbaker said...

My impression is that 'Free Will' is merely a medieval apologetic invention to counter the question of why does not god not show himself and improve the lot of humanity, instead of remaining hidden and allowing evil.

If god was to eliminate the ills of mankind, goes the argument, then man would not be able to exercise 'his Free Will'. Hence suffering.

Arguing about 'Free Will' is an idiotic endeavor, one that gives great glee to the religiously motivated, because any argument reinforces their framing that god actually exists.

Frank said...

Gingerbaker, how can you say free-will is a medieval invention? It's at least 2000 years old. It must be as old as the notion of reward and punishment. Without free-will, according to believers in r & p, r & p would be meaningless (at least quite unfair).
If free will is an invention, I think that this reason for it is stronger than your reason for it.

David said...

The mystery to me is how you missed the boat completely. Of course if we are automatons then if you change the inputs the response will change. When I wrote: "then deterrents are an illusion." It was not an argument, as you seem to think, that behavior cannot be modified. It was an argument that the imposition of the deterrent itself is predestined by the universe's differential equation. We are not outside the system imposing a deterrent on the inside; we're on the inside too. If you modify a behavior because you imposed a deterrent, both the modification and the imposition were predetermined—it is that sense that the deterrent is an illusion.

I thought computer types were good at levels of abstraction—isn't that what OOP is all about?

Jeffrey Shallit said...

David:

If I misunderstand your point, perhaps it is because you express your point incoherently. You wrote

How can a deterrent make me choose not to commit a crime, unless I have the facility of choice?

and I answered it. A deterrent can make you choose not to commit a crime, as you admit. So your question is silly.

P. S. The plural of "automaton" is "automata".

KeithB said...

I keep thinking about what I have heard about Doublas Hofstedder's (sp!) thoughts about this. Especially the book where he talks about running his late wife's ego* program along side his own. In other words, he could choose a movie based on what this "dead-wife-program" tells him about her preferences as opposed to his own. Could a Christian have incorporated the moral code he has been inculcated with as a separate ego that contends for a spot in the decision making hierarchy?

*I really should get this book and read it.

David said...

Well, that's one explanation. But there are other explanations, less advantageous to you--such as you ignored the entire post and focused on an isolated comment which, in light of the post, could not have meant when you claimed. (You know, the quote-mining explanation.)

P.S., according to dictionary.com "automatons" is an acceptable plural. If you are going to be a pedantic ass, you have to be an accurate pedantic ass.

Jeffrey Shallit said...

according to dictionary.com "automatons" is an acceptable plural.

Ah, someone who thinks they can deduce typical usage from looking in a dictionary.

By all means, feel free to use "automatons" but don't be surprised if people laugh when you say it.

David said...

Shallit,

Ah, someone who thinks they can deduce typical usage from looking in a dictionary.

By all means, feel free to use "automatons" but don't be surprised if people laugh when you say it.


I'll interpret that as: "I'm wrong, but I don't have the grace to admit it, even to the extent that I'll criticize someone who *gasp* uses a dictionary to support his claim of the plural form of a word. Then I'll offer the self-affirming: "don't be surprised if people laugh..."

I'll remind you that you didn't write: "We use automata in computer science," but rather "The plural of 'automaton' is 'automata'" --which is demonstrably false in its universality--leaving you with no comeback beyond, in effect, "well what do dictionaries know, anyhow?"

Diogenes said...

I don't think Shallit was a pedantic ass on this point. It's fair to correct people's grammar. There are some common grammatical mistakes that drive me to conniptions.

The credit card commercial that says: "There's no restrictions" Plural/singular mismatch AAARRGGGGHHH!!!!

Jeffrey Shallit said...

David:

I thought Christians were supposed to be modest. Yet you seem to want to take credit for everything that goes on in my head as well as your own. Good luck with that.

XiXiDu said...

Muddled Thinking about Grammar

Nobody has any authority when it comes to grammar. Natural language evolved and does evolve. Once you design your own language you can nit-pick about people being wrong in their use of your language. That is, as long as they are not trying to improve it in a critical fashion, since otherwise you are either claiming copyright infringement or simply can't stand critique. My humble opinion :-)

Luckily there does exist no holy book of grammar.

Jim Lippard said...

Philosophical libertarian free will has nothing to do with legal or criminal responsibility or punishment, as has been forcefully argued by Stephen J. Morse of the Univ. of Pennsylvania Law School faculty.

The Atheist Missionary said...

Interesting site. This is my first visit here.

IMHO, free will is a mirage: http://www.atheistmissionary.com/2009/07/michael-vick-determinism-and-mirage-of.html

Miranda said...

The Atheist Missionary says:
"IMHO, free will is a mirage"

What forced you to come to that humble opinion?

XiXiDu said...

The problem is that people think about free will in terms of cause and effect. If you read the link I posted in the first comment above, you'll see that free will can be about the ability to do as you want and not to choose what you want.

Now I'm uneducated, but nobody yet told me what's the problem with this idea about free will and why anybody would want to be free from forces opposed to be being a focal point of forces that does imprint reality with a pattern of volition created by the forces of nature but shaped and applied by the consciousness middleman that is our subjective first-person perspective.

More here:
http://spacecollective.org/XiXiDu/5759/Free-will-as-nonlinear-transformational-effectiveness

It's just my own solution that I'm shaping on the journey towards a more informed worldview.

Matteo said...

If there is no free will, Shallit had to write this blog post. By the same token, idiot commenters on YouTube had to write their inanities. If there is no free will, then why should I prefer reading one to reading another? Both are equally well-determined events, and there are far more idiot commenters on YouTube than intelligent (if wrong) writers on the philosophy of consciousness. Given that there are far more YouTube idiots, shouldn't I just pay them the most attention, since they are quite evidently the most evolutionarily successful?

But, then, I have no real choice in the matter. If Shallit wants to maintain that no one would willingly choose to read his blog, then who am I to argue?

Jeffrey Shallit said...

If there is no free will, Shallit had to write this blog post.

You're very confused. The universe could have a stochastic element (and probably does; viz. radioactive decay) and hence people's actions might not be deterministic; yet they might still lack free will.

Anonymous said...

Here's a way to think about the free will/determinism paradox that I've found helpful:

Shuffle a deck of cards. What's the probability that the top card is the ace of spades?

According to elementary probability theory, 1/52, but in real life, it's either 1 or 0 depending on whether the top card is the ace of spades or not. It either is or isn't; cards are macro objects not subject to quantum indeterminacy, and with a good enough high-speed camera, I could have watched your shuffle and told you precisely whether the top card was or wasn't an ace of spades.

Now think about the choice of whether or not I visit Dr. Shallit's blog. The decision can be completely deterministic, but that doesn't necessarily mean I know what the decision is until its made. Just like I don't know the true probability of the top card being an ace of spades until I look, I don't know whether I'm going to choose to visit Recursivity until I do. And once I've made the choice to visit the blog, I can't rewind and make the opposite choice any more than I can put the top card back, draw again, and get a different result.

In many cases where people invoke "randomness," the only true source of unpredictability is a lack of knowledge of the initial conditions or transformative rules of a system.

Until you can take a brain state, clone it, and have each instance arrive at a different decision based on the same input, there is no reason to suspect that the brain is anything but a deterministic, wholly material computational device (with an admittedly complex and inscrutable architecture), at least with respect to free will and determinism.

Of course, the alternative to compatibilism is that "free will" means "determined without respect to material phenomena," or more simply, "completely at random." Is that the moral, humane, dignified picture of the human person that the religious are trying to push?

--Dan L.

Timothy said...

Matteo said, If there is no free will, then why should I prefer reading one [Dr. Shallit's blog post] to reading another [inane comments on YouTube]?

Because human (and many other smart animals') brains are built to prefer stimulation over a lack thereof. We are drawn to novel stimuli over inanity. If we had actual free will that wasn't constrained by the hardware or software of our brains, then we would be able to make ourselves be more interested in inanity than novel stimuli. But we cannot. The program that is our mind can only do what evolution programmed it to do.