From: jwgh@earthlink.net (Jacob Haller)
Newsgroups: talk.bizarre
Subject: FTSD: The Incompetent Watchmaker
Date: Tue, 1 Dec 1998 22:48:41 -0500
X-ELN-Insert-Date: Tue Dec  1 19:55:11 1998
X-Newsreader: MacSOUP 2.3.3
Organization: EarthLink Network, Inc.
X-Posted-Path-Was: jwgh
Lines: 63
X-ELN-Date: 2 Dec 1998 03:48:42 GMT
Message-ID: <1djdqii.1iu0mid1j424aoN@1cust80.tnt35.nyc3.da.uu.net>

It started as a joke, really.  How can you figure out if the principle
of scientific induction actually works?  The whole edifice of science
rests upon it, yet it itself has never been (can never be?) conclusively
proved.

Well, obviously, we reasoned, if you want to find out if something is
true or not, then, scientifically, you have to design an experiment to
test it.  If the test turns out one way, then you know that the
proposition is false (or true, depending on how the experiment is
designed); if the other way, then it may not prove anything
conclusively, but at least it's another piece of information to add to
your store.  ("There!  You're using induction there!" one of us crowed.
"OK, then, a positive result is meaningless.  The principle stands," the
other grumbled.)

An intellectual game, then, and if not one seriously taken then one at
least taken to absurd lengths for the sheer fun of it.  What experiment
could you make?  Would it be possible to design a control?

We came up with this:

Create a device which is designed to create output with a given pattern.
Flip the switch, turn the crank, pull the lever, whatever is required to
start it up.  Observe.  If it doesn't follow the given pattern, then the
hypothesis that induction holds--that past pattern is a predictor of
future performance--is weakened.  Otherwise, the inductive hypothesis is
left, if not strengthened, then standing at least.

The circular nature of this amused us, and we could have left it at
that.  But the idea stuck with me, and I decided (as a lark) that I
would create a device, designed solely to fit the requirements of the
experiment.  Ultimately I decided that a small device with a digital
display that counted from 1 to 10 and back down again would be just the
thing; small enough to carry around, and a nice little conversation
piece when I was in a certain type of crowd.  I thought it would be
amusing.  (Amusing!)

The problem was that after the first couple of weeks the thing stopped
working properly.  One day I took it out of my pocket and caught it
counting down from fifteen.  It acted normally for a while, then started
to display prime numbers only.  Finally I couldn't even predict if it
was going to act predictably or not.  I examined the workings minutely,
made sure everything was in working order.  But finally I had to admit
the truth.

You know those times when you swear you left something in one place but
find it in another?  You remember someone having blue eyes but the next
time you look they're green?  The person you met the other day who
seemed so nice is now clearly a total twit?

Oh, induction may have worked fine once upon a time.  But the universe
is running down, and the ordered rules that used to serve us all so well
are starting to wear thin.  For all I know. you could wake up tomorrow
and find everything gone.

I have the experimental data to prove it.

-jwgh
-- 
"Many people in Manhattan believe that if they travel outside of
Manhattan proper the world is flat and they may very well fall off the
map and end up in Brooklyn or Queens or worse."
   - Kinky Friedman, /God Bless John Wayne/