Dembski gets raked over the coals again

| 55 Comments

Cosma Shalizi has a sharp dissection of Dembski's peculiar abuses of information theory. Things don't look happy for poor Bill.

Dembski's paper seriously mis-represents the nature and use of information theory in a wide range of fields. What he puts forward as a new construction is in fact a particular case of a far more general idea, which was published in forty-four years ago. That construction is extremely well-known and widely used in a number of fields in which Dembski purports to be an expert, namely information theory, hypothesis testing and the measurement of complexity. The manuscript contains exactly no new mathematics. Such is the work of a man described on one of his book jackets as "the Isaac Newton of information theory". His home page says this is the first in a seven-part series on the "mathematical foundations of intelligent design"; I can't wait. Or rather, I can.

55 Comments

William A. Dembski Wrote:

I would go further than that and say that I value objective peer review. I always learn more from my critics than from the people who think I’m wonderful.

I think that is certainly the case here. I didn’t see any of the ID fringe say anything of the sort that Cosma has done.

So, Behe, Dembski et al. misrepresent information theory? There’s a surprise. They misrepresent and/or misunderstand just about every other field of science they ever talk about. And Phillip Johnson is still grappling with logic. The fact that these people have academic positions and advanced degrees almost defies logical explanation.

Not to detract in the least from William Dembski and the Discovery Institute, Renewing Science and Culture by Re-Inventing the Wheel, but this was also obvious to the math experts here at Panda’s Thumb.

***************

This is telling tales “out of school,” a big no-no, but Dembski was so clearly wrong that our very own PT math experts worried (privately) that THEY were wrong because how could the “Newton of Information” be so obviously full of shit?

I am just a poor little “stones and bones” archaeologist, but every time Demski comes in spitting distance of my professional experience, he is full of shit.

Maybe this will become a general trend.

FYI: I added a post on this over at ARN for discussion over there.

Cool Jack but do you really want to refer to the Blog posting as a ‘paper’?

I’m starting to feel pity for the dumbass.

To address Gary’s statement, indeed several of our math experts have been wondering why Dembski was touting a restricted Renyi measures of information (alpha=2) as something new or even relevant to ID.

Seems to me that the foundations of ID have been quite effectively removed by its critics. Speaking of which, I intend to address the latest book by Young+Edis (editors) called “Why intelligent design fails”. In thirteen chapters contributors Gert Korthof, David Ussery, Alan Gishlick, Ian Musgrave, Niall Shanks, Istvan Karsai, Gary Hurd, Jeffrey Shallit, Wesley Elsberry, Mark Perakh, Victor Stenger and of course Taner Edis and Matt Young show how shakey the foundations of ID really is.

A must read

Thanks, Pim. I fixed that.

Clueless Undergrad Wrote:

I’m starting to feel pity for the dumbass.

Tu stultus es!

Frieslander Wrote:

Seems to me that the foundations of ID have been quite effectively removed by its critics. Speaking of which, I intend to address the latest book by Young+Edis (editors) called “Why intelligent design fails”. In thirteen chapters contributors Gert Korthof, David Ussery, Alan Gishlick, Ian Musgrave, Niall Shanks, Istvan Karsai, Gary Hurd, Jeffrey Shallit, Wesley Elsberry, Mark Perakh, Victor Stenger and of course Taner Edis and Matt Young show how shakey the foundations of ID really is.

Wow, a compendium of lightweights.

If even “lightweights” can show that ID is a fraud, what does that make ID proponents?

Nice Robert, the usual ad hominems show that these ‘light weights’ once again have shown why ID is without much merrit. And given your reaction it seems you know this.

Wink…

Apparently Dembski isn’t producing good math, but could he be producing good science?

yeah, who cares what lightweights like David Wolpert say about things like NFL? It’s way over their heads.

If lots of science is based on evolution, and it’s so well understood it’s used in engineering, and all IDers can do is pretend they aren’t creationists, use math incorrectly, and insult evolutionists, I think I’ll just have to be satisfied with that.

dr. Lane Wrote:

Apparently Dembski isn’t producing good math…

According to whom? I have yet to see a qualified mathematician comment on William Dembski’s mathematics. As for Rosenhouse, if I want the topological properties of the knot formed by my shoelaces, I’ll contact Rosenhouse because that’s about all he is good for; probability and statistics is out of his league.

PZ Myers Wrote:

If even “lightweights” can show that ID is a fraud…

That is an open question, as far as I am concerned.

From ARN:

According to Cosma Shalizi, the variational information measure I describe in my paper “Information as a Measure of Variation” is not new (for his criticism, see here). I never claimed to be “introducing” a “new” measure of information (do a find-command in Adobe Acrobat on these or similar terms and you’ll see that confirmed). Indeed, I was fully prepared that the measure had been introduced already. That’s because (a) the mathematical machinery I was using is old (I was fully aware of that) and (b) this information measure is derived very naturally. To see that I’m not making this up, on July 28, 2004 I sent Jeffrey Shallit, Wesley Elsberry, Richard Wein, Thomas Schneider, Victor Stenger, and Mark Perakh (along with other critics) the following email (please note the second to last paragraph):

quote: ——————————————————————————–

Dear Critics,

As I indicated in the preface to No Free Lunch, “my strategy in writing this book has been to include just enough technical discussion so that experts can fill in the details.” Because the experts have not done this, the burden remains on me to fill in the technical details.

To begin to redress this burden, I’m planning a series of seven articles, collected together under the rubric Mathematical Foundations of Intelligent Design. Attached is the first installment: “Information as a Measure of Variation.” Subsequent planned articles are:

“Uniform Probability and Bayesian Methods” “Displacement” “Specification” “Universal Probability Bounds” “Complex Specified Information” “The Law of Conservation of Information”

I expect to place some of these articles in the mainstream statistics/probability/complexity literature. Once they are all finished, I plan to collect the articles in a straight mathematical monograph (perhaps with CUP in one of their lecture note series or with the IMS). Once this monograph is done, I plan to produce a second edition of The Design Inference (the book is now dated, needs to be cleaned up, and make its mathematical underpinnings clearer). I’ve already been in touch with CUP about this (though, sadly, Terry Moore, my editor at CUP, recently passed away).

I’m writing you because I want your scrutiny on this project. Certainly, I want the mathematics to be correct; I don’t want to reinvent the wheel; and I want to be sure that the results are not trivial or being misinterpreted (by either side!). Also, I see this as a crucial phase in the debate over my own contribution to design-detection methods: whether my program ultimately succeeds will depend on being able to fill in the technical details that till now I’ve only addressed in part.

Future installments will be posted on my website (www.designinference.com) as well as on the ISCID website (www.iscid.org, look under the Archives).

Best wishes, Bill Dembski

——————————————————————————–

I didn’t expect my usual critics (e.g., Shallit, Elsberry, Wein, Schneider, Stenger, and Perakh) to point out where the measure would have been introduced because the paper is beyond their expertise. So it was passed on to Shalizi, who does have the requisite expertise and who noted that Alfred Renyi had in fact introduced it earlier (I gave the Hilbert space, or L^2, version, but Renyi developed a more general version). Thanks to Shalizi, I’ll cite Alfred Renyi in future versions of my paper.

So what’s the problem? “Information as a Measure of Variation” is intended to be the first chapter in a monograph on the mathematical foundations of intelligent design. This chapter is supposed to lay some conceptual groundwork. As far as I’m concerned, it continues to do that job admirably. Mark Perakh, in his book Unintelligent Design (p. 90) criticizes my identification of information with the negative logarithm to the base two of a probability as “amateurish.” In “Information as a Measure of Variation” I show that my “simplistic” definition generalizes in ways that Perakh can no longer follow mathematically and that, as I intend to show in future installments, is deeply significant for science.

Out steps Perakh and in steps Cosma Shalizi to mount much the same criticism, though now at a higher level of sophistication. This time the charge is that there is no new mathematics in my paper. If I thought there were new mathematics here, I would have submitted the paper to a mathematics journal. Instead, I submitted it to Complexity. As anybody who has read this journal knows, its audience is quite diverse, with articles accessible to lay readers as well as quite technical articles (mine will fall on the technical side, perhaps too technical for Complexity). At any rate, it is not a mathematics journal, and I submitted my paper there fully conscious that its significance would lie in the insights it might generate for understanding evolving natural systems.

Shalizi’s criticism of my paper epitomizes why I got out of research mathematics. As a postdoctoral fellow at MIT in 1988, I was struck by how much of mathematics research was spent extending and refining (usually by some very small epsilon) existing results with no regard whatsoever for why those results were significant except for extending and refining existing results. If this sounds circular, it is. Shalizi’s charge amounts to: “How disgusting! Your variational information measure was developed over forty years ago, and look at all the progress since then.” My interest is in using this measure, applying it as a conceptual and analytical tool for understanding certain aspects of the natural sciences.

As I see it, the merit of my paper consists in the following:

Showing that the negative logarithm to the base two of a probability is the canonical way to conceive of information, and that what the communication engineers have done by always averaging this measure distracts from the privileged place this measure has (or should have – see especially the quotes by Fred Dretske in my paper). . Showing how this very simple way of conceiving information generalizes naturally and canonically to a more sophisticated way of understanding information, namely, in terms of the variational information. Note that this is the canonical extension. Alfred Renyi’s generalization over and beyond the variational information is not canonical. Think of a curved manifold embedded in Euclidean space. Euclidean space has a canonical metric, and that metric extends only one way to the curved manifold in terms of geodesics. Sure, other metrics exist on the manifold that preserve the topology. But to extend the metric canonically, there is only one way to go. I argue a similar point in my paper. . Introducing a canonical continuity measure (the continuity spectrum) for informational paths with respect to the variational information. The canonicity of this measure is the crucial point. The Kantorovich-Wasserstein metric, on which the continuity spectrum is based, is the only way to fully extend the metric structure on a metric space to the probability measures sitting on that metric space. I argue this in my paper and, accordingly, argue that this gives a preferred way of understanding continuity of informational paths (i.e., a preferred information geometry). Shalizi dismisses my point by simply saying that the topic of information geometry has been well-explored. But the question remains whether the information geometry that I introduced (yes, here I am claiming priority) is just reinventing the wheel. Shalizi gives no evidence of familiarity with the Kantorovich-Wasserstein metric. What’s more, the article by Streater that he cites, though describing a number of information geometries, does not describe the one I developed. Further, Shalizi’s claim that “the continuity spectrum seems to be nothing more than a confused (and admittedly conjectural) grope towards the idea of distance and divergence measures” seems to concede that there is something new (albeit confused) but at the same time fails to acknowledge that the continuity spectrum is a perfectly well-defined mathematical notion (i.e., for a given informational path, the continuity spectrum has a given mathematical form). At any rate, Shalizi has presented nothing like a cogent argument to show that the notion cannot have useful applications.

In conclusion, I want to thank Cosma Shalizi for the time he devoted to criticizing my paper. In future versions of it, I’ll be citing Alfred Renyi and acknowledging Shalizi for the reference. Also, I would ask Shalizi, when he has the chance, to supply me with the proof of the conjecture on page 13 of my paper (a conjecture relating to the continuity spectrum). I’ll be happy to remove the conjecture and replace it with an actual theorem along with his proof (giving Shalizi full credit).

Or perhaps I’ve proven the conjecture myself already, and am simply using it to gauge whether critics of my paper are in fact qualified to criticize it.

Apparently, Mr. R. O’Brien can not follow the argument under discussion.

As Daddy used to say, “Tuf shit, boy.”

Happily, I didn’t come from an academic family.

Syntax Error: mismatched tag at line 7, column 49, byte 286 at /usr/local/lib/perl5/site_perl/5.16/mach/XML/Parser.pm line 187.

Billy, Reputable journals do not publish warmed over hash. If you had ever known about the relevant literature, you would have cited it in the first place. You didn’t because you couldn’t.

My next article exposing your weak low ball is titled “Dembski: Hammered and Screwed.”

The EF fails on complexity (hammers) and fails on specification (the screwdriver). You are screwed.

My next article exposing your weak low ball is titled “Dembski: Hammered and Screwed.”

The EF fails on complexity (hammers) and fails on specification (the screwdriver). You are screwed.

As a lowly social “scientist,” you are not remotely qualified to refute William Dembski. Why don’t you stick to the poorly designed sample surveys and unfounded conclusions that are the hallmarks of your field.

As a matter of fact, I am published in medicine, chemistry and biology, as well as anthropology. Indeed, I found the anthropology reviewers are much more demanding than any others. You childishly disparage work that you have no inkling about, and have made no contributation.

I have done with you. Adios.

Doesn’t feel so good when the insults are turned on you, does it?

We need a sign in the pub that says “Do not feed the Trolls”. If kids like R.O. want to spout off like ill behaved junior high students, just ignore them. If he had any evidence for ID he would have presented it by now.

From 6284

I have yet to see a qualified mathematician comment on William Dembski’s mathematics.

Surely that should tell you something.

Robert O’Brien: I have yet to see a qualified mathematician comment on William Dembski’s mathematics.

Dr. Lane: Surely that should tell you something.

Yep, that I have a higher standard than other people here.

In conclusion, I want to thank Cosma Shalizi for the time he devoted to criticizing my paper. In future versions of it, I’ll be citing Alfred Renyi and acknowledging Shalizi for the reference.

Wait, he didn’t reference who or what the base formulation is and just pretended that he invented it from scratch. Please tell me you are joking. This is dishonest to the highest order. It would be like deriving a special form of the Navier-Stokes equations and then not say as much in the paper. In fact, he should start with the generalized form, at least mention it is passing, and then describe the specialization that is being addressed directly. I would think someone who has done writing for technical journals for many years should know that by now.

G3 Wrote:

If kids like R.O. want to spout off like ill behaved junior high students, just ignore them.

You must have me confused with Steve Story, your resident “Mini-Me.”

I’m pleased that Dr. Dembski has joined the dialog, since - whatever opinions people might have of his writings - they are the subject of much discussion here.

I, for one, would be delighted if he could speak for himself when questions come up about his work, as his supporters here have, to put it politely, not done a good job.

Will I be hauled before the star-chamber of Darwinian Orthodoxy if I suggest we engage him with as much courtesy and as little hostility as possible?

In his prompt reply to Shalizi Dembski specifically mentions my name a few times. In his opinion, I am not capable of understanding the math of his new paper, so now “Perakh is out and Shalizi is in.” On the other hand, he asserts that his new paper does the intended job “admirably” (plus a few more remarks asserting the immense importance of his contribution to science). I have already commented on Dembski’s self-admiration and contempt for his critics several times before, and quoted the maxim attributed to Leo Tolstoy. Its gist is that the value of a person is like a fraction wherein the numerator is the person’s actual talents and achievements and the denominator is what the person thinks of himself. In Dembski’s case, the numerator may be estimated differently depending on who is making the estimation. Many detractors of Dembski think that it is quite low, while his admirers like O’Brien or Dembski himself seem to think it is very high. However, even if the numerator is high, the large value of the denominator will make the fraction approaching zero. And in Dembski’s case the denominator is obviously very large. He also seeems to be confident that the critics of his work (he listed six including me) are incapable of judging his great mathematical achievements. I admit that the material so finely addressed by Shalizi is not part of my ken, as I am a physicist rather than a mathematician, and Shalizi is certainly much better qualified to analyze Dembski’s paper. However, I had no problem with understanding Dembski’s new paper, and I doubt that mathematicians like Shallit, Levitt, Wein or Erik had any such problem. In fact, I’ll soon post an essay on definitiions of complexity, which, although it has no relation to Dembski’s new paper, contains references to Renyi’s divergence, Kullback-Leibler divergence and related matters. I am not pretending to be an expert in this field, but I can assert Dembski that I can understand his mathematical effort and appreciate Shalizi’s critique. Dembski’s new paper was subjected to a brief discussion on two closed fora and we came to a consensus that the critique of the paper (as per Wilson and Shalizi) should best be entrusted to a professional mathematician with a specific expertise in that field and Shalizi performed the task in a very convincing manner. Now Dembski tries to say that he did not claim his result to be new and that he was perfectly aware that the math he used was old. Hard to believe - were this the case he certainly would gave referred to Renyi and others who followed Renyi. He says that in the subsequent versions of his paper he will refer to Renyi and to Shalizi. It is hard to imagine how he will be able to refer to Renyi without undermining the pretense of making a contribution to information theory. It seems obvious that the Isaac Newton of information theory was quite ignorant of the vast literature which started with Renyi over forty years ago and contains many papers wherein Renyi’s ideas have been utilized and further explored and developed.

Perhaps I’ll also comment very briefly on the posts by O’Brien. As seems to follow from his exchange of messages with Wolpert, O’Brian is incapable of reasonably interpreting what others say. I have communicated with Wolpert rather extensively and know his opinion of Dembski - in David’s view there is nothing of substance in Dembski’s mathematical exercises, and I pointed to that in my critical comments to Dembski’s work. O’Brien’s calling my critique “rabid” or “frothing-at-the mouth” is not only rude but also false. In my critique of Dembski I have never stooped to using any rude words like those routinely used by O’Brien. Such epithets can though legitimately be applied to O’Brien own posts which are usually rude and intended to insult those who do not share his admiration of Dembski. O’Brien tries to find in Wolpert’s response to his inquiry something more positive in regard to Dembski’s output, but this only shows that O’Brien simply, intentionally or unintentionally, misinterprets Wolpert’s quite clear words to suit his (O’Brien’s) admiration of Dembski. Perhaps because of the lack of arguments of substance, O’Brien resorts to insults and other forms of underhanded assaults. With friends like O’Brien Dembski needs no adversaries. It is useful to have posts like those from O’Brien on this blog as they demonstrate what kinds of adversaries (lacking qualifications which they replace with arrogance and impudence) we sometimes have to encounter. Cheers! Mark

Robert Obrien Wrote:

According to whom? I have yet to see a qualified mathematician comment on William Dembski’s mathematics.….

Well, Mr. O’Brien – you certainly seem to be full of yourself today… Care to exercise your formidable intellect and take apart this critique of Dembski’s work?

Robert O’Brien continues his ‘defense’ of Dembski with: ‘According to whom? I have yet to see a qualified mathematician comment on William Dembski’s mathematics. As for Rosenhouse, if I want the topological properties of the knot formed by my shoelaces, I’ll contact Rosenhouse because that’s about all he is good for; probability and statistics is out of his league.”

Nice ad hominems, ignoring the actual arguments. Typical…

O’Brien provides us with a fascinating insight into how ID proponents handle criticism, especially when the criticism is hard hitting and valid.

Thanks Robert.

What I find fascinating is that the ‘Isaac Newton of information technology’ is unaware of the work by Renyi. If Dembski’s paper is mostly a reinvention of the wheel then I wonder if journals would be interested in publishing such especially when the author seems unaware of the state of information technology?

Dembski asked for critical comments, now that he got them he is doing a lot of spinning. Fascinating behavior showing in action his comments in “DEALING WITH THE BACKLASH AGAINST INTELLIGENT DESIGN’

Such as

‘putting critics to use’

Critics and enemies are useful. The point is to use them effectively. In our case, this is remarkably easy to do. The reason is that our critics are so assured of themselves and of the rightness of their cause. As a result, they rush into print their latest pronouncements against intelligent design when more careful thought, or perhaps even silence, is called for. The Internet, especially now with its blogs (web logs), provides our critics with numerous opportunities for intemperate, indiscreet, and ill-conceived attacks on intelligent design. These can be turned to advantage, and I’ve done so on numerous occasions. I’m not going to give away all my secrets, but one thing I sometimes do is post on the web a chapter or section from a forthcoming book, let the critics descend, and then revise it so that what appears in book form preempts the critics’ objections. An additional advantage with this approach is that I can cite the website on which the objections appear, which typically gives me the last word in the exchange. And even if the critics choose to revise the objections on their website, books are far more permanent and influential than webpage

Irony alert highlighted… What do ID proponents think of Dembski’s approach here?

‘the principle of radiating confidence’

Instead, we need to be mentally and emotionally tough enough to withstand their attacks and to avoid being cowed.

Although in my opinion Dembski may have confused arrogance with confidence.

it may be best for critics to stay quiet until Dembski’s ‘writings’ appear in a paper, although I wonder if critics can wait that long, before showing the many flaws in his arguments.

I, for one, would be delighted if he could speak for himself when questions come up about his work, as his supporters here have, to put it politely, not done a good job.

And don’t think for a moment that THAT is not a two way street. I notice that steve, for example, was silenced as soon as Dembski posted. As it should be.

(Not trying to be combative about it, it’s just a thought for consideration.)

FL

G3 has the best suggestion: ignore the trolls (unless and until there’s anything of substance to respond to).

The lack of substance speaks eloquently for itself - a lot of sound and fury, signifying nothing.

A question for Dembski (or his supporters): Was he aware of Renyi’s work or not? And if he was aware of Renyi’s work, would it not have been appropriate to provide for the usual references to his work?

So far, all I know is this:

Trolls or no trolls, Dembski has said something worth checking out; Shalizi has said something worth checking out; Jack Krebs (alone!) has had the sensibilities to call ARN’s attention to Shalizi and ask for response, (which his post and subsequent ARN discussion was worth checking out), and Dembski has had the sensibilities to come to PT and offer commentary, and that too was worth checking out.

Perakh’s comments are of interest, to be checked out, but his remarks are not coming from a professional mathematician’s perspective such as Dembski and Shalizi.

So, I guess I’ll keep primarily reading Dembski and Shalizi, trying to understand as much as possible about what they are talking about, while secondarily reading ARN and PT fwiw.

(As for PvM’s Renyi question, I would think that a “Was he aware…” question would be best asked of the person whose awareness one is trying to gauge._

FL

Isn’t thins the same Dembski tactic we’ve seen over the years?

1. Write a convoluted mongraph intentionally using unnecessarily unintelligible jargon such as “mathematical machinery”

2. Be challenged on the monograph on the basis that it merely a special case of a well understood and well known rule.

3. Make two inconsistent responses to the challenge by claiming that he knew of the existence of the general rule and assumed everybody else did too, but thanked Shalizi for pointing out the similarity and promising to give credit in the future.

4. Claim that his earlier work is dated, impliedly suggesting that he can not longer be cricized for “dated” work.

5. Calim tha this work is “incomplete and therefore he cannot be criticized on this work either.

6. Suggest that he is near a breakthrough: “Also, I see this as a crucial phase in the debate over my own contribution to design-detection methods: whether my program ultimately succeeds will depend on being able to fill in the technical details that till now I’ve only addressed in part.”

7. But also state that he’s only addressed part of the problem and that it will be fullu addressed in future work, so he can’t be criticized for incomplete work.

8. Finally demand that others do his work for him, or suggest that nobody understands his work anyway: “Also, I would ask Shalizi, when he has the chance, to supply me with the proof of the conjecture on page 13 of my paper (a conjecture relating to the continuity spectrum). I’ll be happy to remove the conjecture and replace it with an actual theorem along with his proof (giving Shalizi full credit).

Or perhaps I’ve proven the conjecture myself already, and am simply using it to gauge whether critics of my paper are in fact qualified to criticize it.”

Barbara Forrest and Paul Gross call this tactic of Dembski’s, “the treadmill” in Creationism’s Trojan Horse.

It’s nice to see a good example here.

FL: As for PvM’s Renyi question, I would think that a “Was he aware … “ question would be best asked of the person whose awareness one is trying to gauge._

Conceivably, that’s why PvM prefaced his query with: “A question for Dembski…”

A few additional comments regarding Dembski’s response to Shalizi. He sent his article to “Dear critics” and explicitly requested a scrutiny. Now he claims that all six dear critics are not qualified to judge his work. Typical Dembski-style consistency.

Regarding our qualifications, I cannot speak for Shallit, Levitt, Schneider, Stenger, and Wein, but I suspect Dembski might be surprised if he actually found out the level of knowledge and understanding of the subject matter these people may have. This suspicion is based on a simple fact: unlike Shallit, Levitt, or Wein, I am not a mathematician, but I read Dembski’s paper without problems, and the other dear critics most probably are better versed in the subject than I am.

In relation to the above, I’d like to tell something else. About one year ago appeared a collection of papers titled From Complexity to Life, edited by Gregersen. It contains papers by Chaitin, Bennett, Stewart, Morowitz and others, and also a paper by Dembski on his beloved complex specified information. Many of the papers in that book were of much interest for me and I wanted to review the entire collection. I started writing reviews of several of the papers there, and after a while I had reviews of Bennett, Stewart, and Davis half-ready. I was proceeding slowly as other projects distracted from that job. Then Norman Levitt published an excellent review of that collection which slowed down my work even more. The material rested on my disk waiting its turn whenever its time would come. Among the partial reviews almost ready was that of Bennett’s article devoted to the definitions of complexity. (BTW, surveying various definitions of complexity, Bennett did not mention at all Dembski’s “contribution” to the topic at hand, despite Dembski’s being a contributor to that collection and the author of many-many words about complexity in Dembski’s numerous publications).

It probably would continue resting on my disk, but the discussion of Dembski’s new paper and its critique by Shalizi prods me to separate the review of Bennett’s paper and publish it right now as a stand-alone piece. The reason is simple: although that review has no relation whatsoever to Dembski’s new article, it contains some material wherein I refer to Renyi and some subsequent publications further exploring Renyi’s ideas. I admit that when I perused Dembski’s paper, I did not realize that his Variational Information is in fact Renyi’s divergence of the second order. I realized that when Shalizi pointed to this fact. However, the review in question (which I intend to post to Talk Reason as soon as I manage to convert it into PDF so the equations could be reproduced properly) shows that if I, not a mathematician, have had some (admittedly very limited and rather shallow) knowledge of Renyi ideas, mathematicians like Shallit, Levitt, or Wein (and I would add Erik who is not listed by Dembski, and Jason Rosenhouse) may be expected to have a better knowledge of this stuff than I. On the other hand, Dembski, acclaimed as a great expert and innovator in information theory, would be expected to know all this material “along and across.” In fact, he thanks Shalizi for pointing to Renyi and promises to add proper references, which makes his asertions that he knew about all of that without Shalizi sound quite implausible.

One more detail. Dembski’s Variational Information turns out, as Shalizi pointed out, the second-order Renyi divergence. Could it be that Dembski at least pioneered the use of that particular version of Renyi divergence? Let us see. In the mentioned review of Bennett which I am going to post shortly, there is a discussion of the so called LMC complexity. This measure was suggested in 1995 by Lopez-Ruiz, Mancini, and Calbet (the reference is given in my review of Bennett). It is a function of the second-order Renyi entropy via an exponential term. Of course Renyi antropy and Renyi divergence are not exactly the same, but the fact is the Renyi’s generazied measures of the second order have been used before as well. In my review I did not mention this fact because it has no relation to the topic of my review, but I am mentioning it here to show that Dembski’s measure not only is a well known particular case of a more general measure, but that even this particular version had been indirectly utilized before.

The final remark: Regardless of the merits or shortcomings of Dembski’s newest mathematical exercise, what its relevance to Intelligent Design can possibly be? I see none. Perhaps some of the ID advocates, who inundate this blog with spiteful remarks, will for change say something of substance, in particular pointing to the relevance of all that Dembski’s great math to ID?

FL writes

So, I guess I’ll keep primarily reading Dembski and Shalizi, trying to understand as much as possible about what they are talking about, while secondarily reading ARN and PT fwiw.”

Meanwhile, those of us back on Earth with finite amounts of time will wait until it’s reasonably clear that Dembski isn’t a garbage-talking charlatan. I expect that time will never come. But please keep us posted FL on when you think Dembski has finally succeeded! I’m counting on you. I want to be as close to ground zero as possible when the greatest revolution in the history of scientific thought reaches fruition.

And surely, FL, you agree that the mathematical and unassailably logical proof of an inconceivably intelligent entity would the greatest revolution in the history of science. Right?

A selection relevant to this discussion from one of the greatest books of all time, The Hitchhiker’s Guide to the Galaxy:

Now it is such a bizarrely improbable coincidence that anything so mind-bogglingly useful could have evolved purely by chance that some thinkers have chosen to see it as a final and clinching proof of the NON-existence of God. The argument goes like this: ‘I refuse to prove that I exist,’ says God, ‘for proof denies faith, and without faith I am nothing.’ ‘But,’ says Man, ‘The Babel fish is a dead giveaway, isn’t it? It could not have evolved by chance. It proves you exist, and so therefore, by your own arguments, you don’t. QED.’ ‘Oh dear,’ says God, ‘I hadn’t thought of that,’ and promptly disappears in a puff of logic. ‘Oh, that was easy,’ says Man, and for an encore goes on to prove that black is white and gets himself killed on the next zebra crossing. Most leading theologians claim that this argument is a load of dingo’s kidneys, but that didn’t stop Oolon Colluphid making a small fortune when he used it as the central theme of his best-selling book, “Well, That about Wraps It Up for God.”

I believe the PvM quotation was prefaced,

A question for Dembski (or his supporters):

On another question, I would hope to be near ground zero too, GWW, if such a revolutionary day as the acceptance of within ID in the scientific community should in fact arrive. Just have to wait and see.

Scientific revolution is always a slow bake, one can only check the oven temperature and peek inside on occasion.

And surely, FL, you agree that the mathematical and unassailably logical proof of an inconceivably intelligent entity would the greatest revolution in the history of science. Right?

Well sure it would be, but I honestly don’t think it would take all that to create some serious scientific paradigm-shifts anyway. If ID advocates keep on going and find themselves successful a few years hence, that might would do it.

FL

Oops, typo. Should be:

“acceptance of ID within the scientific community”

(in fact, it should actually say “acceptance of ID within the majority of the scientific community”).

FL Wrote:

Scientific revolution is always a slow bake, one can only check the oven temperature and peek inside on occasion.

But as far as ID is concerned there is no evidence of any scientific revolution. Apologetic revolution perhaps but when it comes to science ID has been shown to be without much merrit. And this is not just evidenced by the many critics who have shown in full detail the many fundamental problems with ID as a science but also with an absence of ID relevant work by ID proponents.

There is just nothing to speak of…

FL writes

If ID advocates keep on going and find themselves successful a few years hence, that might would do it.

Do what?

And by “few years” do you mean the usual “few years” (i.e., three or four) or the geological “few thousand years”?

FL Wrote:

Scientific revolution is always a slow bake, one can only check the oven temperature and peek inside on occasion.

It seems to me that ID movement’s approach is

1) Make a recipe but not test it or use it 2) Hold the bake sale

caerbannog Wrote:

Well, Mr. O’Brien — you certainly seem to be full of yourself today … Care to exercise your formidable intellect and take apart this critique of Dembski’s work?

I have already seen that critique and have had an e-mail correspondence with Dr. Wolpert concerning it (the most relevant part of which was posted to a previous thread).

This quote from Shalizi is priceless, but I doubt Shalizi is aware of it,

But he might have thought to look Patrick Billingsley’s wonderful 1965 book on Ergodic Theory and Information,.…

Billingsley was Dembski’s PhD advisor for his doctorate in mathematics!

caerbannog wrote:

Well, Mr. O’Brien … Care to exercise your formidable intellect and take apart this critique of Dembski’s work?

O’Brien: I have already seen that critique and have had an e-mail correspondence with Dr. Wolpert concerning it (the most relevant part of which was posted to a previous thread).

“caerbannog” didn’t ask whether you had seen it; (s)he asked if you’d care to “take it apart” (i.e. analyze, critique, show us where it’s mistaken…)

For those who may not have seen, or may not recall, the correspondence Mr. O’Brien refers to, you might have formed the impression, from this response to caerbannog, that that correspondence somehow Dr. Wolpert is more sympathetic to Dembski and ID than Dr. Perakh would have you believe. Here is the relevant quote from that correspondence:

[David Wolpert:] My skim on Dembski is that he has nothing to agree with or refute - there is nothing concrete to respond to, one way or another… PS. For the record though, I am suspicious of the ID movement. It has trouble passing the smell test, IMO.

I’m at a loss as to why Mr. O’Brien thinks, in light of the critique itself and this correspondence about it, that Dr. Wolpert’s opinion differs in any way from Dr. Perakh’s.

hey Steve, you might want to add your last name, or people will think you’re me.

Russell Durbin Wrote:

“caerbannog” didn’t ask whether you had seen [Dr. Wolpert’s critique]; (s)he asked if you’d care to “take it apart” (i.e. analyze, critique, show us where it’s mistaken … )

No, I do not.

About this Entry

This page contains a single entry by PZ Myers published on August 10, 2004 4:10 PM.

It’s Tangled Bank time again! was the previous entry in this blog.

Tangled Bank #9 is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.

Categories

Archives

Author Archives

Powered by Movable Type 4.381

Site Meter