Re: [asa] Doug Groothuis v. William Dembski

From: Michael Roberts <>
Date: Thu Jan 01 2009 - 11:55:03 EST

This is what I have long thought about these probabilistic arguments. Unless
one has good reason to ascribe meaning to the numbers.i.e probability then
the whole exercise is worthless. I can only give a general objection but it
is good to see it from one with good maths

----- Original Message -----
From: "Iain Strachan" <>
To: "David Opderbeck" <>
Cc: <>; "Rich Blinne" <>;
Sent: Thursday, January 01, 2009 4:45 PM
Subject: Re: [asa] Doug Groothuis v. William Dembski

> On Thu, Jan 1, 2009 at 4:23 PM, David Opderbeck <>
> wrote:
>> Yes, but, I don't want to be hasty in dismissing the analogia entis
>> altogether. First, I think we do have intuitive a priori knowledge of
>> God
>> -- a key question is the degree to which that natural knowledge is
>> surpressed by sin (Romans 1). Second, it seems reasonable to assume that
>> something created will bear some characteristics that are analogous to
>> characteristics possessed by its creator -- e.g., beauty, rationality,
>> order. If we see beauty, rationality and order in nature, and we have at
>> least some dim sense that there must be a God, it seems reasonable to
>> suggest that these characteristics point towards a creator-God who also
>> possesses those characteristics. A weak form of the argument from
>> design --
>> "small i.d." if you will -- seems to me a reasonable argument to make.
> I don't disagree with the "small i.d." argument you're making. The
> problem arises if one wants to use Dembski's rigorous probabilistic
> arguments (be it via explanatory filter or via CSI). If one is going
> to make rigorous probabilistic arguments then one must be able to
> assign hard numbers to the a priori probability in order to make
> inferences ( Bayes's theorem).
> On a second thought - I am not sure how CSI is defined in general, but
> it seems to me that just the existence of a DNA molecule with no real
> useful information in the base sequence has what might be described as
> Complex Specified Information just from the base-pairing of
> nucleotides. Consider the double-helix, say of 100 nucleotides. For
> each of the 100 on one strand there are four possibilities, hence
> 4^100 possibilities. Now, given complete randomness, one could say
> the same for the other strand, so there are 4^200 possibilities. But
> as we know an A only ever pairs with a T and a C only with a G. Thus
> ANY 100 base-pair sequence is highly specified in the sense that there
> can only be 4^100 out of 4^200 sequences. This simply arises out of
> the chemical bonding properties of the nucleotides, and not from any
> intelligent agent. It seems to me that it is both complex (you need
> 200 bits of information to write it down) and highly specified ( the
> general case would require 400 bits). Hence I don't really see how
> CSI in this case could be an indicator of design.
> Iain
> To unsubscribe, send a message to with
> "unsubscribe asa" (no quotes) as the body of the message.

To unsubscribe, send a message to with
"unsubscribe asa" (no quotes) as the body of the message.
Received on Thu Jan 1 11:55:20 2009

This archive was generated by hypermail 2.1.8 : Thu Jan 01 2009 - 11:55:20 EST