I think it is a mistake, when reading Dembski, to assume that he ever meant any of what he writes to be taken seriously by serious-minded people. Rube-fooling is the business model. This is the guy whose book, The Design of Life, attributes the finding of homology between reptilian jaw bones and the mammalian ear ossicles to a mere “bone count,” after all. He cannot possibly expect credibility among educated folks when he does things like that.
1 Like
In December 2019 I made a similar argument, in a post at Panda’s Thumb entitled Is Algorithmic Specified Complexity useless for analyzing evolution? There simply is no reason for thinking that values of that “randomness deficit” indicate anything of relevance to understanding evolution.
2 Likes
He published in some sort of philosophy of religion journal. It certainly never got any mathematical review. Appearance are he never gave it a second read himself.
Good article! I especially liked …
Natural selection cannot work in a population that always contains only one individual. To model the effect of natural selection, one must have genetic variation in a population of more than one individual.
… because in my discussions with Eric Holloway he never seemed to grok the concept that populations have variability. (I’ve have some nice discussions with Eric - we disagree, but we get along OK.)
1 Like
Which illustrates a major defect in how we teach evolution–by not starting with (often completely ignoring) existing variation, we all but promote the false assumption that populations are clonal and have to “wait” for new mutations to happen, when in reality it operates almost entirely on existing inherited variation, which anyone can observe.
This defect in turn provides an invitation for IDcreationists to focus their politicoreligious attacks on the randomness of mutations…
3 Likes
He’s doing it again! Next week he has a new book, with Casey Luskin (have you ever noticed that Casey Luskin is like Billy Barty? It’s impossible to mention Luskin without getting a laugh.), and this new book is published by what appears to be a deeply crazy Christian publishing company, “Harvest House.”
I assume the book will begin with Dembski’s profound apologies for all the lies he’s told over the years, and then will move on to presentation of a thoroughly credible case for ID Creationism.
3 Likes
Dembski’s argument from Complex Specified Information (or his very-closely-related argument from Specified Complexity) is not illogical, contrary to what some people here have said. It is basically an argument that if we have a scale of adaptedness, and the probability of seeing adaptedness as high as, or higher than what we observe is very small, then the probability of achieving that by ordinary evolutionary process is implausible. Question: probability under what null distribution? One way would be to consider all possible genotypes equiprobable. When you do that, it turns out to be easy to show that there are cases where ordinary processes of mutation and natural selection can increase specified information, with no obvious barrier to achieving 500 bits of it (see for example my 2007 article in Reports of the National Center for Science Education).
That would dispose of the CSI argument, but Dembski in 2006 clarified that he did not intend to use the equiprobable distribution as his null distribution. Instead he meant that the probabilities should be computed as the probabilities of outcomes under ordinary evolutionary processes. This is hard to compute, but more importantly leaves us in the following situation.
- We want to see whether the degree of adaptation is improbably high under normal evolutionary processes. We want to use Dembski’s argument.
- So we compute the probability of adaptation this good or better under those processes. Dembski does not give us any guide to how to do that.
- If that probability turns out to be very low, we want to express this as showing that CSI is present. So we compute -log2(P) and see whether it shows that CSI is present.
- If it is, we can conclude that the probability of adaptation this good or better is so low as to be implausible that it could have happened by ordinary evolutionary processes.
You may see the problem here. We have already, at step 2, seen what the probability is, and can judge whether it is so low as to render normal evolutionary processes implausible. Doing the CSI step (step 3) does not add anything to that except an impressively sciency and informationy label. It is redundant. And Dembski gives us no clue as to how to do the probability calculation. My take on this issue will be found in my 7 April 2013 post at Panda’s Thumb.
5 Likes
Your pieces on this really are excellent. Marvelous explanations.
I have always thought of Dembski less as “illogical” and more as being the man who, having worked out an equation that shows that bumblebees cannot fly, decides not to reevaluate his work but instead to rail at the bees. That isn’t, strictly speaking, illogical: one can take the view that it is the bees who are breaking the rules and that one’s math is perfectly sound. But it does defy probability, which is the sort of thing you might expect a man who claims to have his math right to understand.
2 Likes
If reality and your mathematical model don’t match, question your model rather than reality. Alan Kleinman anyone?
5 Likes
5 posts were split to a new topic: HcQ is not an Information argument
A post was merged into an existing topic: HcQ is not an Information argument
This is an example of painting a bulls-eye on a target after shooting the arrow.
Specification, in the ID argument, is functionality, not knowing the sequence in advance.
It seems to me that’s what ID-creationists are doing with “functional” biopolymer sequences.
Function is in the eye of the beholder and contextually dependent, just like it is for biological polymers.
There is no reason in principle why the sequence of letters you wrote couldn’t be functional in some context, and I could just as well argue that you typing that sequence made it functional in that you relied on it’s functioning as an example of gibberish.
Heck, we could even say certain geological structures serve functions from certain perspectives. Depressions in the Earth’s surface serve the function of providing reservoirs for fresh water, which serve the function of hosting particular ecologies and so on. Again we can then take any particular example and show that it is extremely unlikely to obtain this particular type of thing by just randomly pulling from the total space of all possible arrangements of an equal number of atoms, structures that perform the function of being reservoirs for freshwater ecologies. And yet we also know there is a natural process that produces such extremely rare structures given certain initial and ongoing geological conditions on the surface of a rocky planet, in the same way biological evolution is a process that produces extremely rare functional sequences of biological polymers.
2 Likes
The point is, Specifies does not mean “knowing the sequence in advance”.
In the first “Looking for sources on the information argument” thread, I used the following definition of information from Websters:
the attribute inherent in and communicated by one of two or more alternative sequences or arrangements of something (such as nucleotides in DNA or binary digits in a computer program) that produce specific effects
This is the definition that Stephen Meyers uses in the book “Signature in the Cell”, which is what I offered as a source.
So, the information is sequential and functional. Capable of producing an effect. For letters of the alphabet, it would mean valid words and sentences. For sequences of DNA, it would be producing functional proteins/rna products/regulation.
If you want to refute the ID information argument, you should actually address it.
SARS-CoV-2 D614G mutation is sequential information.
The novel sequence is highly specified to enhance the binding of the viral spike to human cell ACE2 receptors.
Greater Covid infectivity leading to immense personal and social consequence.
Question - was the specified D614G sequential information the product of a mind, or came about as a mutational variation which has appeared in independent SARS-CoV-2 lineages and become dominant by strong environmental selection?
3 Likes
In theory, both propositions could be simultaneously true.
N.B. - The theological proposition is closely linked to the problem of theodicy, on which I do not feel qualified to expound.
Regards,
Chris
Dave, science isn’t so much about making and refuting arguments; it is about rigorously trying to falsify the empirical predictions of one’s own hypothesis. Do you have any hypothesis, and if so, what testable, empirical predictions does it make?
You seem to have accidentally offered one that Ron just tested. 
4 Likes
The mutation is a change to the sequential information of the spike protein gene.
The S gene still codes for a spike protein, albeit, one that is more effective at binding to human cell receptors.
There is evidence that it was the product of a mind in the form of human engineering. But, It could also have come from a mutation, either conditional or random and increased in frequency due to environmental selection.
Neither ID nor creationism argues that a point mutation can’t occur randomly and affect the function of a protein.
We aren’t conducting scientific experiments here, are we? We are on an internet discussion forum. Making and refuting arguments.
As far as a hypothesis of my own goes (and I’m not a scientist conducting experiments of my own), it would be that science would continue to uncover more and more systematic behavior of living organisms. Where mutations aren’t so much random as programmatic or conditional. This is not to say that random mutations can’t or don’t happen.
One example of this is Lenski’s cit+ e coli. What at first seemed random has been shown by further research to be very repeatable when the right conditions are met…
He is clearly talking about creating and testing hypotheses, not conducting experiments.
This is not a hypothesis. In addition this prediction was made centuries ago.
Do you even know what random means in the context of biological evolution?
What seemed random at first?
3 Likes
D614G literally means a point substitution/mutation from aspartate (D) to glycine (G) at position 614. Stop saying nonsense please.
4 Likes