Wesley R. Elsberry posted Entry 1277 on August 2, 2005 07:55 PM.
Trackback URL: http://www.pandasthumb.org/cgi-bin/mt/mt-tb.fcgi/1275

Over on his weblog, William Dembski has a post making reference to an article on a means of “fingerprinting” textured surfaces, like paper. It is an interesting article. But look what Dembski has to say about it:

The Logic of Fingerprinting

Check out the following article in the July 28th, 2005 issue of Nature, which clearly indicates how improbability arguments can be used to eliminate randomness and infer design: “‘Fingerprinting’ documents and packaging: Unique surface imperfections serve as an easily identifiable feature in the fight against fraud.” I run through the logic here in the first two chapters of The Design Inference.

Well, it is a little troubling how to proceed from this point. Did Dembski fail to read the article? Is Dembski simply spouting something that ID cheerleaders can nod sagely about without regard to whether it happens to accord with reality? Whatever excuse might be given, the plain fact of the matter is that the procedure and principles referred to in the short PDF Dembski cites have nothing whatever to do with Dembski’s “design inference”, and cannot be forced into the framework Dembski claims.

(Continue reading… on Antievolution.org)

Commenters are responsible for the content of comments. The opinions expressed in articles, linked materials, and comments are not necessarily those of PandasThumb.org. See our full disclaimer.

Comment #40998

Posted by ts on August 2, 2005 8:48 PM (e)

Take, for example, Dembski’s claims that Visa credit card numbers or phone numbers represent CSI concerning the cardholder or phone owner (see Intelligent Design. p. 159).

Ah, so Dembski equates “intelligent design” with “produced by a random process” (such numbers are generally produced by a pseudo-random number generator, but since the random number generator is normally seeded with the clock time when the program was started, and which numbers are assigned to which customers depends on the order in which they are assigned, which depends on a raft of external events, they are truly non-deterministic). Well, I can go with that.

This makes clear the fact that even Dembski doesn’t understand when and how to deploy his own notions of “design inference”.

That’s not quite right; Dembski understands when and how to deploy them to shore up support from his equally inept acolytes, get funding, bamboozle reporters, provide cover for C student politicians who never read a book, and so on.

Comment #41016

Posted by Don on August 2, 2005 10:47 PM (e)

“But, in fact, work in probability went on before Dembski’s odd notions were published, goes on now without reference or reliance on those notions, and will continue into the future when Dembski’s work is no more than a footnote in reviews of socio-political oddities.”

So brutal!

…and accurate!

Comment #41034

Posted by Stephen Erickson on August 3, 2005 1:10 AM (e)

I had no idea Dembski was the first person ever to compute a posterior probability. I’m sure he’s very disappointed the authors of this paper failed to properly cite his groundbreaking work!

Comment #41063

Posted by Richard Wein on August 3, 2005 7:59 AM (e)

Dembski claims that the article “clearly indicates how improbability arguments can be used to eliminate randomness and infer design”. In fact, the article doesn’t even clearly indicate how improbability arguments can be used to eliminate randomness, let alone say anything about inferring design.

Certainly, the article gives one example of rejecting a probabilistic hypothesis on the basis of small probability. But that’s just what virtually any application of statistical hypothesis testing does. This example is nothing special, and actually says nothing at all about the statistical reasoning employed (and that includes the “supplementary information” at the Nature web site).

The article also says nothing at all about inferring design. It proposes a method for determining whether one piece of paper has been substituted for another. But both possibilities (substitution has occurred or not) could be the result of intelligent agency.

As has been pointed out before, Dembski’s method of design inference is merely a god-of-the-gaps argument. It tells us to infer design when we have rejected all the non-design hypotheses we could think of. Certainly, statistical hypothesis testing is one way to reject hypotheses. But statistical hypothesis testing owes nothing to Dembski and is quite independent of his god-of-the-gaps approach to design inference.

Comment #41065

Posted by SirL on August 3, 2005 8:05 AM (e)

If Dembski’s design inference is as revolutionary as he claims, there’s a lot of money to be made in digital watermarking. Somebody should point that out to him, because we really can’t let all that talent go to waste…

Comment #41088

Posted by Stephen Erickson on August 3, 2005 10:35 AM (e)

Everyone who ever published a p-value owes a world of debt to William Dembski.

Comment #41092

Posted by PvM on August 3, 2005 11:11 AM (e)

There are times when Dembski starts to sound more and more like a Salvador… Becoming a victim of his own arguments…
Wesley, as usual you have done an excellent job at exposing the grandstanding of Dembski. From the moment you forced Dembski to admit the existence of apparant versus actual CSI, he never seems to have recovered. Now he is arguing that random processes can be designed… False positives, which were initially claimed to not exist, but over time were accepted, are what make the design inference ‘scientifically vacuous’ or as Dembski puts it ‘useless’.

From 1996:

I argue that the explanatory filter is a reliable criterion for detecting design. Alternatively, I argue that the Explanatory Filter successfully avoids false positives. Thus whenever the Explanatory Filter attributes design, it does so correctly.

via 1999

“On the other hand, if things end up in the net that are not designed, the criterion will be useless.”

to 2001

Now it can happen that we may not know enough to determine all the relevant chance hypotheses. Alternatively, we might think we know the relevant chance hypotheses, but later discover that we missed a crucial one. In the one case a design inference could not even get going; in the other, it would be mistaken. But these are the risks of empirical inquiry, which of its nature is fallible. Worse by far is to impose as an a priori requirement that all gaps in our knowledge must ultimately be filled by non-intelligent causes.

Funny how these arguments lead to but one conclusion. The criterion is useless. While people have objected to this conclusion, Dembski’s own latest seems to strongly support the conclusion.

No wonder Dembski has chosen to ignore most of what critics say… No wonder, more and more people come to realize that ID is scientifically vacuous.

Comment #41095

Posted by minimalist on August 3, 2005 11:31 AM (e)

So, it’s official: ID these days consists of picking random articles out of the literature and shrieking “SEE? DESIGN!” without much – or any – justification. If you’re lucky, it may be a quick gloss of the basic idea of the paper that demonstrates they read as far as the abstract; most of the time they horrendously misunderstand or deliberately misstate the article’s conclusions to make it look like it lends support to ID.

This is pretty much all I see on ID blogs or IDEA’s sites whenever they feel like addressing the science at all. (Unless they think Dembski’s nyah-nyah responses to Jeffrey Shallit constitute “scientific debate”)

But golly, if there’s so much support for their ideas in the experimental literature, surely it would be a snap to set up their own course of research to test their ideas. Right?

Comment #41097

Posted by C.J.O'Brien on August 3, 2005 12:13 PM (e)

Dembski : Mathematics
Old Lady : Rummage Sale

Comment #41098

Posted by Jim Wynne on August 3, 2005 12:26 PM (e)

When I was in high school I worked in an store that sold TVs, and often some dolt would buy a 27” set and then be surprised when the box wouldn’t fit in his car. In those cases, we suggested that the TV be taken out of the box, in which case we could usually get it into the back seat. One guy refused to take my word for it that the box wouldn’t fit (it wasn’t even close) and insisted that we try to jam the thing into the back seat. After a few predicably futile attempts, I convinced him that we should take the set out of the box, and he reluctantly agreed. I asked him if he wanted to keep the box, and he said he did, and I asked him if he wanted me to flatten it for him (as it wouldn’t fit in the car otherwise) and he declined. After I got back in the store, I looked outside and saw the man trying to fit the empty box into the car. I tell this story only because I think of it every time I read about Dembski pulling something out of his butt and trying to make it fit through his filter.

Comment #41099

Posted by RBH on August 3, 2005 12:30 PM (e)

Stephen Erickson wrote

Everyone who ever published a p-value owes a world of debt to William Dembski.

Even those of us who published a p-value before Dembski was born?

RBH

Comment #41101

Posted by RBH on August 3, 2005 12:35 PM (e)

minimalist wrote

So, it’s official: ID these days consists of picking random articles out of the literature and shrieking “SEE? DESIGN!” without much — or any — justification. If you’re lucky, it may be a quick gloss of the basic idea of the paper that demonstrates they read as far as the abstract; most of the time they horrendously misunderstand or deliberately misstate the article’s conclusions to make it look like it lends support to ID.

It hasn’t “come to this” – “this” is the way it’s been since the beginning; witness the Bibliography the DI pilgrims presented to the Ohio State Board of Education three years ago.

RBH

Comment #41102

Posted by Russell on August 3, 2005 1:15 PM (e)

Stephen Erickson wrote:

Everyone who ever published a p-value owes a world of debt to William Dembski.

I believe “Stephen Erickson” is a pseudonym for Shirley Eugeste. (Note the initials.)

Comment #41113

Posted by rdog29 on August 3, 2005 2:36 PM (e)

So two random surfaces have a very low probabilty of being indistinguishable from each other - big deal. What the hell does that have to do with CSI?! Not much - Even a dummy like me can see that.

Dembski must really be scraping the bottom of the barrel.

Comment #41119

Posted by Shirley Knott on August 3, 2005 3:06 PM (e)

The problem is that when Dembski and his ilk scrape the bottom of the barrel, they’re reaching up to the limits of their appendages. They so consistently rise to the level of their own incompetence…
Were it not so tragic it would be funny to watch.

Hugs,
Shirley Knott

Comment #41140

Posted by Dene Bebbington on August 3, 2005 5:15 PM (e)

What’s instructive about this matter is that nobody, including Dembski himself, is applying the notion of CSI rigorously to anything. In the meantime, others are using probability productively. All Dembski can then do is grab onto their coat tails trying to make it look like it somehow vindicates his work.

Comment #41842

Posted by RBH on August 8, 2005 1:39 AM (e)

I’ve just gotten around to reading the actual Nature paper, having rescued that issue from the pile on my desk into which it was slowly sinking. I cannot fathom how Dembski thinks it is an example of anything having to do with his purported design detection methodology. Dembski introduces his post saying

Check out the following article in the July 28th, 2005 issue of Nature, which clearly indicates how improbability arguments can be used to eliminate randomness and infer design:

That is simply bizarre: there is no design inference at all going on in that paper. The technology described allows one to conclude (via correlations of repeated scans of random surface imperfections) that a blank piece of paper is either the same piece of paper the instrumentation looked at three days ago or is a different piece of paper. The only improbability that enters is that of the peak value of the correlation value, a perfectly standard statistical testing procedure that predates Dembski’s The Design Inference by many decades. Is Dembski claiming that he invented statistical significance testing and estimating alpha?

I am slowly becoming convinced that when he refers to his own “design” detection methodology, Dembski literally does not know what he’s talking about any more (if ever he did).

RBH