Tuesday, June 28, 2011

Descartes Demon, The Problem of Induction, and the Death of Certainty

Well, after a whole bunch of biblical studies posts, I figure it's time to get back to some philosophy, keep it interesting around here.  In many ways, the history of modern philosophy can be looked at as an attempt to ground reality in some sort of certainty.  The big names in philosophy, like Descarte, Leibniz, Kant, Schopenhauer, etc. were trying to build a system from the ground up, so that we can be sure of everything we know.  Once we get the foundation, they reason, we can start securing to that foundation, and in that way build up a massive tower of knowledge from a few simple premises.  It was a noble goal, and it is very interesting to chart the path of this quest, and to see the shambles it is in today.

Descarte has been called the father of modern philosophy, because he did something very interesting when he started his quest into philosophy.  What if I was just put into this world knowing nothing, and I had to figure out what is real going solely by my reasoning powers?  If I strip away every belief it is possible to doubt, what will I be left with?  These are the questions he set out to investigate.  They name this type of skepticism "Cartesian Doubt" because of this contribution he made.

It was from this doubt that he comes up with probably the most well known philosophical statement of all time.  Descarte realized that there was one thing he couldn't doubt, even if he assumed all his senses were unreliable.  From the cold dark depths of utter scepticism, a small voice rings forth.  "I think, therefore I am!"  What he saw was that to deny his own existence was to believe a contradiction, because without existing he couldn't perform the act of denying.  He had it!  This was the capstone upon which he could build his system.  The goal to derive all knowledge from a set of a-priori truths we can know intuitively is known in philisophical circles as "rationalism".

There was a slight hitch though.  Even acknowledging his own existence, he had to admit it was possible that everything else he believed was a lie.  It was totally possible that there was an evil demon, incredibly powerful, who was manipulating Descarte to believe in the existence of other people, a physical world, etc.  There is another version of the argument called the "brain in a vat" argument.  Descarte believed he could get around the implications of this argument by his peculiar versions of the causal and ontological argument, then saying that a good God wouldn't deceive us in such a fashion.  In that way he was able to reclaim most "common sense" knowledge and ground his philosophy in certainty.

It wasn't too long before people began to realize that Descarte's foundation had some cracks in it.  He hadn't eradicated skepticism, and a new group of philosophers were soon to come along and undermine much of the ground he had gained.  George Berkely argued quite convincingly against the existence of a physical world (check out "Three Dialogues between Hylas and Philonous" and think about how you would engage in the conversations).  David Hume came along after that and took the wind completely out of the rationalist sails.

Hume is an awesome writer, and his ideas are clear and devastating to the search for a solid foundation for philosophy.  Many of the best arguments for skepticism can be traced back to him as the first to really eloquently lay them out.  For instance, it was he who introduced the "is - ought" gap in ethics, as well as an argument against the miraculous that is still talked about today, though in slightly modified form.  I want to talk about another one of his contributions that hit me like a brick when I first grasped the sense of his argument.  It's known as the problem of induction, and I won't be able to do it justice, but I shall try.

Hume was an empiricist, which meant that he believed all truth was gained through experience, and that we can't know anything a-priori.  One might say that empiricism developed as a reaction to the rationalism espoused by Descarte, as well as by the philosophers Spinoza and Leibniz.  Now-a-days, there aren't many pure empiricists left, however it has left an indelible impression in the fields of philosophy in general, and has been helpful in developing modern science.  This empiricism created a problem for him when he looked at the relationship between cause and effect. 

He asked himself how we can know that the future is going to be like the past.  He wanted to find the connection that exists between the cause and the effect in an event.  These two questions are intricately related, as once we find that "necessary connection" we can be assured that things will continue as they are right now, for like effects will follow from like causes.  When he looked into this area, he found something disturbing.  He uses the example of a pool ball.  If we were dropped to the earth with all our faculties, but no knowledge, and saw one pool ball rolling towards another, we would have no way of knowing what would happen when the two hit.  Would they stick together?  Would the one pass through the other?  Would they explode?  Any of these are possible until the 2 balls strike, and we see the change in momentum from one to the other.

Furthermore, there is no way to determine the effect simply from the cause.  Examine a pool ball all you want, it will tell you nothing about what happens in a collision without actually watching it collide.  This goes for all of our "inductive" reasoning.  You can't be able to derive what something will do in the future without observing what it has done in the past and then making a prediction.  There is no logically necessary connection between the two.  The more we see an effect issue from a similar cause, the more sure we are that the one is causing the other.

The problem here that Hume points out is that all our causal reasoning is formed completely in our mind.    The events are all the same.  There is nothing different happening when the ball hits the first time than when it hits the 100th time.  The only difference is in how we perceive the event.  You learn in critical thinking class that correlation doesn't imply causation (<---hint: click this one) but it turns out that causation is simply the assumption we make on the basis of correlation.  It gets worse though.  The arguments that are used to try to solve the problem typically just argue in a circle.

It goes something like this:  We know the future will be like the past because past futures have been like past pasts.  Do you see the trick there?  The argument is that things will continue to operate as they always have because they always have operated that way.  It really isn't an answer at all.  It's the same thing with arguing that probability theory solves the issue.  That goes like this:  Though there is no logical connection between cause and effect, we can add up the times that these 2 events happen together and approach a nearly probabilistic certainty that it will happen in the future the same way.  This is subject to the same criticism as before though.  The probability calculation is only valid if the future operates the same as the past, and we have no way of knowing that will be the case.

There is a way out of the problem posed by Hume, but it involves the death of that quest for certainty that had so characterized the modern philosophers.  We can admit that there is no way of knowing that things will always continue on as they have, but keep living as though they will.  The scientific method operates on the assumption of the uniformity of nature.  It has been incredibly useful, but it is simply an assumption.  I think it is a perfectly reasonable thing to believe, and will continue to believe it.  It is simply one of those things we believe that we can't really have any proof for.

P.S.  I put belief in the uniformity of nature in the same category as belief in free will and the material world.  Arguments can cause us to doubt them, but they are a "basic" belief that we can't consistently live doubting.  G.E. Moore argued that we have more reason to doubt the academic skeptical arguments than we do to doubt that the world is pretty much how it seems.  For further reading, I would really recommend either Hume's Treatise of Human Nature Vol. 1, or his "Enquiry Concerning Human Understanding."  They are both really good works, and free on Project Gutenberg and elsewhere.

No comments:

Post a Comment