Re: [asa] Atheist finds God thru Behe's books....

From: Dennis Venema <>
Date: Wed Oct 14 2009 - 01:13:31 EDT

The geneticist can't resist weighing in here:

Rich is correct: Behe built his whole book on a flawed understanding of elementary statistics. I've read all of EoE, and Behe is wrong. He assumes you need all needed mutations (two or three) in one fell swoop. Lenski's work neatly shows up Behe's premises to be flawed, but so did real malaria research BEFORE Behe published.

Have you ever had to mark a paper where a bright student thought they had a great idea, researched it reasonably well (but missed a few key references), presented reasonable-sounding argument with good grammar, etc - BUT had a completely failed premise that you could have pointed out with a 5 minute conversation? Now you know how I felt reading EoE.

There is a reason why peer review by competent scholars in the field is so important.


On 13/10/09 10:00 PM, "Schwarzwald" <> wrote:

Thanks, Rich. So what are the odds that Behe provides which are vanishingly small? If he didn't say impossible, that would be helpful to know - and it would help confirm that Stats 101 claim. As far as I know, Behe isn't saying that such a mutational event has odds so small as to never expect it to occur in nature - what he's saying is that the number of times such mutational events would have to pop up renders "random" mutation as being very unlikely to achieve what it's claimed/assumed to have, in the time frame we know it has to have. In this case, in extremely supportive laboratory conditions, after thirty thousand generations and in a single line, the cells were able to use citrate as a food source (and it's known that E.coli can digest citrate - but it needed an ability to transport citrate under oxic conditions).

Again, re Lenski:

"Such a low rate suggests that the final mutation to Cit+ is not a point mutation but instead involves some rarer class of mutation or perhaps multiple mutations. The possibility of multiple mutations is especially relevant, given our evidence that the emergence of Cit+ colonies on MC plates involved events both during the growth of cultures before plating and during prolonged incubation on the plates."

Seems pretty reasonable to me. Do you find that going overboard?

On Wed, Oct 14, 2009 at 12:38 AM, Rich Blinne <> wrote:

On Oct 13, 2009, at 10:23 PM, Schwarzwald wrote:

How does Behe take away support for his claims from the paper? I provided the link for you - Behe can speak for himself on this. I just provided some quotes and some additional commentary. And you, of course, can respond to what Behe says as well. I also asked where Behe claims that these things are impossible - he may have done so and I simply missed it, so I'd like to see this quote.

Behe makes a Stats 101 boo boo. He takes the probability of a single mutation and squares it for a double mutation. This is an implicit assumption of statistical independence. <> The very title of the Lenski paper reveals that this is NOT true. He claims that the mutation is contingent, the probability of the second event is dependent on the first one thus breaking the assumption of statistical independence. Thus, Lenski refutes Behe. My impossible claim is the odds that Behe wrongly computes in EoE are vanishingly small and to use the vernacular the event are impossible. It's like saying quantum tunneling my fist through the wall is impossible. The odds are non-zero but so close to zero to be effectively impossible.

Rich Blinne
Member ASA

To unsubscribe, send a message to with
"unsubscribe asa" (no quotes) as the body of the message.
Received on Wed Oct 14 01:11:09 2009

This archive was generated by hypermail 2.1.8 : Wed Oct 14 2009 - 01:11:09 EDT