Friday, September 27, 2013

Dennett Breakdown

A while back I wrote about a speech given by Daniel Dennett, the popular philosopher, that I found particularly frustrating.  I have been meaning for a long while to respond to his arguments more thoroughly.  The original video was over an hour long, so I found a shorter video on the same topic to address, point by point.  Even so, I have clipped it down to the essentials, as Dennett has a particular knack for not arriving at the point.

INTERVIEWER: People usually think of free will as opposed to determinism -- of course, by determinism, we mean the idea that basically the future of this universe is inevitable [...] and that includes people's brains; they're deterministic and so free will is kind of an illusion because of determinism [...].  That's the tradition argument against free will.  You're saying the two are compatible in some meaningful sense of both terms?

DENNETT: First of all, I want to say: that phrase, "the future is inevitable" just doesn't mean anything.  The future's going to happen, whatever it is, and that's true whether determinism is true or indeterminism it true, there's going be a future.

The statement "the future is inevitable" has (to me) a perfectly obvious meaning and implication.  It certainly doesn't mean the trivial truth that "there will be a future" (as Dennett supposes), because that is simply part of the framework in which we are having this discussion.  But perhaps there is some subtlety that we are not grasping when we make this particular statement?  Or perhaps misunderstanding things and disproving the wrong sense of them is his modus operandi.

Now in what sense could you talk about the future being inevitable?  I don't know, but what we have is particular events being inevitable, or particular types of events, and in order to see what the word inevitable means, you have to take it apart!  And oddly enough, even though the word trips off the tongue of everyone who talks about free will and determinism, hardly anybody's ever looked at it.  But of course what it means is unavoidable, [...] that's all the word means.

Here is the Merriam-Webster definition of inevitable:

Inevitable:  incapable of being avoided or evaded

Thanks for the clarity!

But now, to avoid something, this is something that an agent does, an avoider.  I mean agent in the broad sense of being an actor that has some sensory capacities and some goals, and acts in the world to accomplish its ends.  Now, are there agents that can avoid things?  Sure, tons of them.

[...]

But now, that means that the whole concept of inevitability gets its meaning from a perspective in which there are agents, that might want to avoid something, and it might be in their power, or it might not.  Now if we start looking at particular worlds with particular agents and particular circumstances in them, we can now start saying "well, in this world, what things are avoidable, what kinds of things are avoidable by this agent, given its powers and its circumstances?".  The answer might be "well, if you throw a brick at it, it can duck", because there's enough light to see the brick, and its nervous system is good enough, and its reflexes are fast enough 

[...]

Now, in order to be able to talk this way, in order to be able to partition the universe into things that are inevitable for that agent, or evitable by that agent, we have to have a way of talking about evitability and inevitability in a deterministic world.  Since there's plenty of evitability in deterministic worlds that we define, the implication that determinism implies inevitability is just false, it's just a mistake.  It's thousands of years old, it's never been pointed out, it's just a mistake.

WAIT.  You're saying that... okay, I get it.

The roundabout conclusion that "determinism does not mean inevitability" suffers from poor definitions; ironic from a man who spends so much time criticizing them.  Strictly speaking, Dennett is not actually wrong; only deeply misleading.  He has re-framed the question as "Can we envision a deterministic world in which agents avoid certain outcomes?".  The answer to his question, as he poses it, is "yes".

But the "agents" Dennett refers to are merely more complicated constructs within the framework of determinism; just arbitrarily elaborate machines.  Dennett has asserted only that deterministic structures do follow the rules of determinism.

The problem comes down to the way he partitions his deterministic world into "agents" and "circumstances".  In this context, agents can be said to "predict" the future and "avoid" some of the possible circumstances.  But the reality is that the agents and the circumstances are fundamentally equivalent, governed by the exact same rules, distinguishable only by some outside bias towards more complicated or discrete machinery.

But by calling these contrivances "agents", Dennett has empowered himself to inflict personal pronouns (and personal motives) upon them.  In effect, has has strongly implied (without saying it) that these "agents" are somehow more than the deterministic constructs that he has shown so elegantly to be deterministic.

He has in effect divided up his mechanical world, placed it into different colored boxes, and is drawing broad conclusions based on the colors of the boxes.

If professor Dennett had meant simply to clarify that the question "does determinism implies inevitability" could be more clearly worded, that would be one thing.  But to select for himself a convoluted vocabulary in which the question becomes trivial and then to make out like "thousands of years" of philosophy has been horribly misguided in answering that question, is simply intellectually dishonest.

What Dennett will not admit directly is that what he calls "free will" is just a backhanded description of the simple phenomenon that machines can be arbitrarily complex.  This, as Chesterton would put it, is a revelation not of thought, but of syntax.

I do not propose here that Daniel Dennett is wrong.  Only that if he is right, he has no business parading about words like "freedom" as if they still meant something to him.  He is singularly adept at speaking for hours about determinism and free will, forcefully shoving words into their awkwardest possible definitions.  Unfortunately, lackadaisical strolls through a morass of bad characterizations will not make determinism any easier to swallow.

Because however eagerly he may connive his way into saying he gets free will, the only bits of free will that are worth having are free, and will.  Determinism plainly gets you neither.

But perhaps determinism could find it in his heart to buy Daniel Dennett a dictionary.

Tuesday, April 9, 2013

Dennet's Determinism

Let me preface this post by reassuring you all of my extraordinary humility.  The primary goal of this endeavor is to clarify my own ideas to myself, and when they are wrong I hope you tell me where and why.

The other night I was having trouble getting to sleep, so I watched a lecture by Daniel Dennet on the topic of free will.  Dennet is an atheist philosopher who Richard Dawkins refers to several dozen times in his book "The God Delusion" -- by the by, I am halfway through another of Dawkins' books, this one on evolution, and I like it very much.

The overwhelming feeling that I got from Dennet's lecture was that he could have explained exactly what he meant in the span of five minutes, but instead took a confusing hour and a half.  For a man to say so little in the span of so much time is an accomplishment in the same sense that golf is a sport: both are just obnoxiously difficult.

Dennet's basic claim is that determinism does imply that everything is determined, but does not imply inevitability.  He shows (or declines to show) this in a roundabout way by refusing to define the sense in which he means inevitability, and to be quite honest I still don't know what he thinks the word is supposed to mean.  For a while he claims that evolutionary biology can explain free will, but he neither explains what he means by free will or what evolutionary biology has to say about it.  For a good five minutes he exchanges the word "avoidable" for the more esoteric "evitable" without explaining what it is that the former means that the latter doesn't.  For another while he goes into an extended analogy about computers playing chess, the purpose of which I cannot hope to divine.

I have come to the conclusion that either Dennet is extraordinarily confused or extraordinarily confusing.  This isn't to say that I do not understand what he meant, only that I did not understand why he meant it.  If strict determinism is true, there can be no escape from the hard narrow groove of our inevitable future.  There might be an overwhelming amount of chaos, which looks like randomness but isn't, or an underwhelming amount of genuine quantum randomness, which only seems like a way out if you prefer losing at roulette to losing at checkers.

I wish Dennet had something more to say, but then if you taken determinism as fact, there is nothing more to say.  There is no free will and there is no free thought, unless we redefine the terms cleverly enough to trick ourselves for a short while into pretending there might be.  It is the only answer to the only question.

The question is "why", and the answer is "because".