From: Iain Strachan (iain.strachan@eudoramail.com)
Date: Fri Nov 17 2000 - 13:10:08 EST
On Sat, 16 Nov 2002 18:34:51 Glenn Morton wrote:
>
>I am finally reading Dembski's Intelligent Design. A certain claim caught my
>attention.
>
> lNow a little reflection makes clear that a pattern need not
>be given prior
>to an event to eliminate chance and implicate design. Consider the following
>cipher text:
>
> nfuijolt ju jt mjlf b xfbtfm
>Initially this looks like a random sequence of letters and spacessinitially
>you lack any pattern for rejecting chance and inferring design.
> lBut suppose next that someone comes along and tells you to treat this
>sequence as a Caesar cipher, in which each letter has shifted one notch down
>the alphabet. The deciphered sequence then reads,
>
> methinks it is like a weasel
>
>Even though the pattern (in this case, the decrypted text) is given after
>the fact, it still is the right sort of pattern for eliminating chance and
>inferring design. In contrast to statistics, which always identifies its
>patterns before an experiment is performed, cryptanalysis must discover its
>patterns after the fact. In both instances, however, the patterns are
>suitable for inferring design.n William Dembski, Intelligent Design,
>(Downers Grove, Illinois, 1999), p. 132
>
>I decided to test this concept. I had a random number generator create
>random letter sequences and then I looked for Caesar cyphers to turn them
>into something meaningful. Using this criterion, my computer is an
>intelligent creature who is trying to communicate with me by DESIGN.
>
>xeckqbfumq
>wasitabrat (was it a brat)
>
>gpizuwbgtu
>ontheslope (on the slope)
>
>wxbukwoors
>amadhatter (a mad hatter)
>
>ijfqbjwdih
>isthisapig (is this a pig)
>
>jyybxozrbd
>beeverfive (be ever five--something said of a dead five year old)
>
>yzusizpzqb
>isthisasin (is this a sin)
>
>It is not that hard to determine a caesar cypher to turn a randomly
>generated sequence into a meaningful, short message. I won't claim that all
>long sequences can be so treated but it is interesting that random sequences
>can be given meaning where none was intended. Thus, the question is how do
>we determine design in the face of this phenomenon?
>
The answer to your question is pretty straightforward. You have had
the random number generator look for 10 letter sequences. The phrase
METHINKSITISLIKEAWEASEL contains 23 letters. The expected length of
time for a random number generator to generate this will be
astronomically longer because it will scale exponentially with the
length of string. The fact it's a Caesar cypher is irrelevant;
without trying all 26 possibilities, it will always take 26 times as
long to get a meaningful string; a constant scaling. But the scaling
with length of string is exponential.
The key to Dembski's argument is that the event should exhibit low
probability. Obviously it's pretty likely that a meaningful 10
letter sequence can come up in a realistic space of time. Difficult
to calculate how long it would take a computer with just random
numbers to generate a meaningful 23 letter sequence, but consider the
following example instead. Suppose the same person wins the jackpot
in the 6-from-49 ball lottery for two weeks in succession. Someone
has to win it every week, but once person X has won it one week
(about which no one is surprised), it acts as a specifier (in the
sense of Dembski's "Specified Complexity". If that specified person
wins it next week, then we suspect foul play. We only expect to see
such an event once in 13 million weeks. Well, it could be
coincidence, of course. Then supposing the same person wins it a
third week. The expected time now is 169 million million weeks.
Now Dembski actually specifies a very conservative "universal
probability bound" of 10^-150, based on a calculation of the number
of events in the cosmos. It has been challenged on theoretical
grounds, but consider this; an event of that order of magnitude is
equivalent to the same person winning the lottery THIRTEEN weeks in a
row. I'm sure long before that, people would suspect a conspiracy
(i.e. "design" rather than randomness).
To illustrate how the "length of string" scaling also works, I can't
resist quoting one of Bob Newhart's funniest sketches, called "An
infinite number of monkeys". He considers the infinite (in reality
finite but large) number of mokneys typing out Shakespeare's works,
and how, for practical purposes you would have to have humans
watching over them to see if anything worthwhile was produced. The
comments of one of these humans goes along the following lines:
"Just been looking at Number 32. Oh boy. I don't think that poor
fella is ever going to come up with anything worthwhile .... But
hold on a minute! Take a look at Number 33! I think this is
something famous:
TO .... BE .... OR .... NOT .... TO ... BE ... THAT ... IS .. THE
...... BZORNENPLATZ.
Iain.
Join 18 million Eudora users by signing up for a free Eudora Web-Mail
account at http://www.eudoramail.com
This archive was generated by hypermail 2.1.4 : Sun Nov 17 2002 - 22:53:19 EST