Fisking Dembski: Ulam and the Wistar Conference

Of course, it is no surprise that similar claims and arguments can be found in other creationist literature such as the book “ By Design Or by Chance?” written by Denyse O’Leary (p93-94)

tacitus:

Of course, it is no surprise that similar claims and arguments can be found in other creationist literature such as the book “ By Design Or by Chance?” written by Denyse O’Leary (p93-94)

I stand in awe that you made in through 94 pages of Denyse O'Leary's writing. That is dedication beyond the call of duty.

I'm glad to see a concerted effort to rescue Stan Ulam from the clutches of the prevaricating creationists.

Dembski's favourite tune is "I heard it on the quote mine"

Checked my institution's library, but no dice. Maybe other commenters or lurkers could do the same.

Failing that, I'll chip in to help you purchase a copy if you'd like. Got a PayPal account PvM?

My libary has it somewhere in the dungeons.

Title
Mathematical challenges to the neo-Darwinian interpretation of evolution : A Symposium held at the Wistar Institute of Anatomy and Biology, April 25 and 26, 1966 / edited by Paul S. Moorhead, Martin M. Kaplan ; Editorial assistance: Pamela Brown.

Publisher
Philadelphia, Pa. : Wistar Institute Press, 1967.

Dembski’s favourite tune is “I heard it on the quote mine”

Richard Wein:

Dembski’s favourite tune is “I heard it on the quote mine”

"...and I'm just about to lose my mind."

Many thanks for saving Stan Ulam from the crazies. I had always considered him to be full of good sense, as well as math genius.

I'm looking at the original monograph (How to Formulate Mathematically Problems of Rate of Evolution? by Dr. Stanislaw M. Ulam) right now, so a few additional quotes are in order. Needless to say, the stuff Dembski does not quote is far more revealing than the stuff he does.

From the very next paragraph (any transcription errors are mine):

"But, I believe that the comments of Professor Eden, in the first five minutes of his talk at least, refer to a random construction of such molecules and even those of us who are in the majority here, the non-mathematicians, realize that this is not the problem at all.

A mathematical treatment of evolution, if it is to be formulated at all, no matter how crudely, must include the mechanism of the advantages that single mutations bring about and the process of how these advantages, no matter how slieght, serve to sieve out parts of the population, which then get additional advantages. It is the process of selection which might produce the more complicated organisms that exist today.

As for myself, I have done a bit of very schematic thinking on the mathematics of such a process and I want to make some remarks to you which certainly are not, as one of the speakers addressed before, correct in a realistic sense but might be relevant for the approach to some quasimathematical discussion at least. The philosophical and general methodological remarks made by various speakers so far can form the basis of what can be, sometime in the future, mathematized. What I am going to do will consist, as it were, of picking out various items from the comments made so far and try to show how, perhaps in some remote future, mathematical schemata can be formulated."

Did Dembski give an accurate representation? What do you think?

I wonder if there are some dedicated (perhaps paid?) quote-miners, who toil diligently combing through any possible source of anything that can be taken out of context and misrepresented. Once a single quote has been mined, then it gets copied forever (sometimes with mutations), so it wouldn't take too many miners.

This presents a hauntingly familiar model: These miner-demons are presented with countless possibilities, but select only those that fit the required creationist environment. These selected quotes get fixed in the population of creationist literature, and gradually change as creationism must keep re-inventing its external appearance to survive.

There seems little chance that Dembski himself ever read the Wistar proceedings or any of the follow-ups. The money quotes were extracted and made available by some ancestor process.

Flint: : These miner-demons are presented with countless possibilities, but select only those that fit the required creationist environment.

Flint...thanks.

A perfectly presented parody of profound prevarication.

I recommend taking a look at the Wistar Proceedings if you've got some time on your hands. From what I remember, they are a fairly interesting read, mostly because of the recorded comments after (at least some of) the papers.

W. Kevin Vicklund:

Richard Wein:

Dembski’s favourite tune is “I heard it on the quote mine”

"...and I'm just about to lose my mind."

"Money, money, YEAH!"

When it comes to Dembski, believe half of what you see, and none of what you hear.

PvM quotes from Some Elementary Attempts at Numerical Modeling of Problems Concerning Rates of Evolutionary Processes, Stanislaw Ulam and R. Schrandt (LA-4573-MS, December 1970): "What was really intended was indications from our computations - simple minded as they were - that a process involving only mitosis, in absence of sexual reproduction, would be indeed much too slow."

Ulam and Schrandt were obviously predicting Dembski and ID.

--my apologies if this has been addressed somewhere in these threads and I just missed it---

I'm no statistician, but can't Ulam's rejected argument be run against pretty much everything that exists? Unless the universe is strictly deterministic, every event has a less than 1.0 probability of occurring. Any current state of affairs required many billions of preceding events to bring it about, so if we multiply together all those probabilities, we see that every current state of affairs has a vanishingly small (a priori) chance of occurring.

And if such a small probability should lead us to conclude that the event could not have occurred, then we should conclude nothing in the present has really occurred.

Ummm...I hope that this counts as a simple reductio ad absurdum against any such line of reasoning, not just when it is applied to evolutionary biology.

I still don't really get their (Dembski, et.al.) point. Even if the possibility for a mutation approaches zero, once the population become sufficiently large, those mutations will happen.
I think I have heard the average person has about 100 billion cells in his body. How many single cell organisms could have first populated the Earth's oceans? If we have trillions (most likely many more) reproductions in a day, that becomes many mutations - every day.
Give this process millions or billions of years to happen and things will evolve.

Why can they not see this?

In a stimulating discussion of several evolutionary phenomena, Dr. Mayr demonstrated the power of the qualitative concepts developed by biologists in explaining the biological world. It also became clear that the mathematical formulation of these concepts will not be easy. Perhaps the consensus of the conference was best expressed by Dr. Waddington in his closing statement when he said: I think we have approached each other to some extent. I hope the biologists have shown the physicists that evolutionary theories are not totally vacuous. I think the physicists have shown us that they are certainly as yet very incomplete, and I think we are ready to realize they are very incomplete. Possibly we now know slightly better in which directions they are incomplete.

The mathematicians participating in the conference subconsciously made the assumption that individuals of a species were genetically identical, necessitating a long time span before a gene locus could pass from one homozygous to a different homozygous condition. For them, evolution seemed to be “tandem evolution”: the succession of genetic events produces a succession of homogeneous populations or species. They did not consider the possibility of a simultaneous genetic variability of ten thousands of gene loci because it would be impossible to calculate its effects. When they spoke of the evolutionary improvement of “the eye.” they made the silent assumption that all eyes in a species are identical, although there is much reason to believe, as biologists have pointed out, that there may be a billion different eyes in a species with a billion individuals. Physical scientists find such variability quite inconceivable, and yet no one can truly understand the evolutionary process until he accepts the magnitude of genetic variation, Another issue at the conference which prompted a great deal of discussion was that of the “constants” or “parameters” of evolutionary change. If one wants to simulate the evolutionary process on the computer, one must assign a numerical value to the frequency of mutation, selection pressure, population size, fertility, gene flow, and all other biological factors of evolutionary significance. Here again there was a great difference of opinion between mathematicians and biologists. The mathematicians quite naturally wanted clear-cut, hard figures to be inserted into their equations. The biologists, on the other hand, demonstrated that most of these factors varied by several orders of magnitude in different species under different conditions. In other words, if one wants to make computer simulations one must repeat them for the entire range of possible values of the controlling factors. The great range of variation of another factor, the site of populations, must not be overlooked. That even the same population may greatly fluctuate in size was emphasized 50 years ago by Chetverikov and Timofreff-Ressovsky, who further stressed that the total population of a species very often consists of numerous more or less isolated small populations, each of which up to a point can go its own way.

Torbjörn Larsson, OM: Ulam was great AFAIK, and the bits and pieces of texts presented here doesn't show the whole argument, but doesn't he seem a bit naive of biology?

To my admittedly untrained eye, you and/or your anonymous colleague may have “misunderstood” the contents of the 1966 paper.

Would it be possible to get a choice set of creationist mine-quotes and how they have "evolved" as an example. This can be published here at PT or other sites for future references on how evolution applies to creationist propaganda. In addition is a good example of evolution in action.

"Dembski has been informed of the full context of Ulam’s work. Let’s see what happens. My thanks to ‘larrycranston’ for his observation that: ..."

And has Larry now been banned from posting at UD?

And has Larry now been banned from posting at UD?

Someone asked larrycranstone for the source of his quote. Sparc linked here. If larrycranston doesn't get banned he can look forward to davtard scolding him for looking deeper.

And the banned played on...

Here's a Scout salute of sincere thanks to PvM and those others here who have troubled to track down the actual texts in this interesting case, rescuing them from the creationists.

Do the Homework -- "that is the Law," to paraphrase the chanting demi-men in The Island of Dr. Moreau . . .

A couple of other obvious criticisms that haven't shown up here...

First of all, neither in 1955 nor even in 1967 were the molecular genetics of cellular reproduction very completely understood.

Second of all, population genetics, which had a lot of development at about the same time, but slightly predates the era, has no problem creating simple but in many cases adequate mathematical models of certain types of evolution, within the context of classical, pre-molecular genetics.

Granted, population genetics, at least when I studied it, treats the "allele" as a more or less invariant entity - it is primarily concerned with changes in the frequency of the alleles that now exist, so to speak, not with the emergence of new ones. Nevertheless, the state of it in 1967 would have made claims of non-existence of mathematical models of evolution a bit tenuous, even then.

http://en.wikipedia.org/wiki/Population_genetics

Could you provide a reference of that interesting quote? I would like to check it out since it seems to conflict with the general consensus reached by the Wistar attendees.

RBH: I note this comment in the UD thread:

Could you provide a reference of that interesting quote? I would like to check it out since it seems to conflict with the general consensus reached by the Wistar attendees.

How do you reckon JPCollado knows what the "general consensus" reached was?

mr_p: I still don't really get their (Dembski, et.al.) point. Even if the possibility for a mutation approaches zero, once the population become sufficiently large, those mutations will happen. I think I have heard the average person has about 100 billion cells in his body. How many single cell organisms could have first populated the Earth's oceans? If we have trillions (most likely many more) reproductions in a day, that becomes many mutations - every day. Give this process millions or billions of years to happen and things will evolve. Why can they not see this?

ONE biologist, just ONE, needs to step up and demonstrate a series of accumulating mutations leading to the construction of novel protein machinery with a useful, and therefor selectively advatageous, function.

Regarding the Wistar meeting, here's one more link that might be of interest:

http://www.infidels.org/library/modern/richard_carrier/addendaB.html#Wistar

(This is focused on refuting Murray Eden's paper.)

Do these people have no familiarity with evolutionary science? Or have they been misled by creationist literature.

— PvM

I think it's helpful when attempting to understand the mind set of IDers to think about people as robots whose decision making is based on a Bayesian analysis. I recall reading a paper a few years ago about attempts to use Bayesian analysis to create a diagnostic program for physicians. They would input probabilities of the various diseases... There was one huge persistent problem with it. It would consistently misdiagnose rare diseases and performed much worse than a person. The reason was that the Bayesian filter (prior probability) for those diseases was so low that regardless of the symptoms it would find a better match with a more common disease.
The analogy here is that the prior probability that IDers have in their minds about being wrong is so low that it's always easier to find some other explanation even if it is a worse fit for the data.

David Buller:

mr_p: I still don't really get their (Dembski, et.al.) point. Even if the possibility for a mutation approaches zero, once the population become sufficiently large, those mutations will happen. I think I have heard the average person has about 100 billion cells in his body. How many single cell organisms could have first populated the Earth's oceans? If we have trillions (most likely many more) reproductions in a day, that becomes many mutations - every day. Give this process millions or billions of years to happen and things will evolve. Why can they not see this?

Remember though that any mutations that occur in normal bodily cell division (mitosis) are irrelevant as they will not be passed on to future offspring. The ones that natural selection act on are those that occur in gamete formation (by meiosis, which produces more variety anyway). Nonetheless, your point stands.

PvM: The amount of ignorance on UcD is incredible

ONE biologist, just ONE, needs to step up and demonstrate a series of accumulating mutations leading to the construction of novel protein machinery with a useful, and therefor selectively advatageous, function.

Do these people have no familiarity with evolutionary science? Or have they been misled by creationist literature. Either way, is this what you want your children exposed to in public schools?

JPCollado: I think you are misreading the Panda’s Thumb entry. When PvM says “I do not have access to the original proceedings”, I believe he is referring to the 1966 Wistar monographs. Did you think PvM was referring to the source document for the quote I provided? larrycranston: Which part of “The discussion there, as reported in the proceedings of the meeting, was rather frequently misunderstood and the impression might have been left that the results somehow make it extremely improbable that the standard version of the survival-of-the-fittest mechanism leads to much too slow a progress.” (italics mine) is so difficult to understand? I submit to you again that Dr. Dembski and/or his colleague misunderstood the 1966 work and were apparently unaware of the later clarification.

Design theory has had considerable difficulty gaining a hearing in academic contexts, as evidenced most recently by the the Polanyi Center affair at Baylor University. One of the principle reasons for this resistance and controversy is not far to seek: design-theoretic research has been hijacked as part of a larger cultural and political movement. In particular, the theory has been prematurely drawn into discussions of public science education where it has no business making an appearance without broad recognition from the scientific community that it is making a worthwhile contribution to our understanding of the natural world.

Easily the biggest challenge facing the ID community is to develop a full-fledged theory of biological design. We don’t have such a theory right now, and that’s a problem. Without a theory, it’s very hard to know where to direct your research focus. Right now, we’ve got a bag of powerful intuitions, and a handful of notions such as ‘irreducible complexity’ and ‘specified complexity’-but, as yet, no general theory of biological design.

I also don’t think that there is really a theory of intelligent design at the present time to propose as a comparable alternative to the Darwinian theory, which is, whatever errors it might contain, a fully worked out scheme. There is no intelligent design theory that’s comparable. Working out a positive theory is the job of the scientific people that we have affiliated with the movement. Some of them are quite convinced that it’s doable, but that’s for them to prove…No product is ready for competition in the educational world.

It’s rather ironic that in a badly conceived and ill-advised “outing” of a Discovery Institute research symposium, all that Nick Matzke and the NCSE have succeeded in doing is outing themselves: they have private knowledge of solid ID research and are actively seeking to repress it.

I think we can be fairly sure that these people have no familiarity with the science. That is the way Dembski and friends need it to be. Otherwise, they’d have no market for their drivel. The ignorance is self-perpetuating: ignorant parents teach their children, who pay no attention in science classes, but who may later read creationist literature that reinforces what they were brainwashed into believing as children… It would take a strong-minded and naturally sceptical individual to break out of that cosy loop.

See Gordon on Brooks on the shy face of ID at Stranger Fruit

What confuses me is why the ID creationists do not spend the time and effort to correct these obvious flawed understandings of evolutionary theory? It makes Christianity look foolish and ID even more nonscientific than it already is. What do the have to gain by not addressing it? And should such ignorance be part of a science curriculum?

— PvM

What Ulam was doing here, as was Waddington in his paper celebrating the 100th anniversary of On the origins of species (Chicago, 1959, ed by Mayr, if my memory is correct), was trying to describe the difficulty in evolving new protein functions based on a step wise mutational model of protein evolution. It is important to keep in mind that at this time many thought that protein sequence differences that were observed were related to phenotypic differences. Relevant to this discussion is that the first protein sequences had only been collected in the late 1950's and the field of molecular evolution was in its infancy. Those who first tried to make sense of these sequences were truly overwhelmed with the difficulty of trying to understand how sequence diversity could explain phenotypic diversity. Both of these men were true scientist who were asking very difficult questions and they were right about one thing; these question could not be answered with what was known at the time.

It turns out that the questions they were asking could not be answered. There were four subsequent discoveries or realizations that put these questions in a different light that makes thinking about them today much simpler. These are {at least}:

1}The realization that most protein sequence differences are neutral.
2) Selection for new protein functions invariably result in chromosomal rearrangements, not in the creation of novel protein sequences.
3} Coding sequences of proteins are organized in exons and exon shuffling can create new functions {this is a specific case of point 2, but of sufficient importance that it deserves its own category}
4) Horizontal gene transfer is an important factor in species obtaining new phenotypes (ie a species does not need to reinvent the wheel every time it faces a new ecological challenge).

So today these questions that Ulam raised are not as difficult to consider than they were in his day. Does that mean we have answered them? The answer is no. His questions point directly at one of the most challenging questions that face those of us who think in molecular genetic terms: what is the relationship between genotype and phenotype?

Both of these men were true scientist who were asking very difficult questions and they were right about one thing; these question could not be answered with what was known at the time.

— syvanen

Frank J: My confusion is sometimes the opposite of PvM's (Comment 143,974). Sometimes I think, "why didn't activists in the 1960s-80s go straight to a design-free 'weaknesses in evolution' approach, and avoid calling attention to the weaknesses and contradictions in their own 'theories' as well as the obvious religious implications?" The only answer I can think of is that they'll try anything they can get away with. They can afford to. We can't.

As long as “no one knows” how a certain process happened, I’m free to comfortably believe what ever I want about it, without feeling like an idiot in the eyes of others. In this light, LACK of knowledge is actually more useful than knowledge. They know ignorance of a natural cause does not logically mean “design,” but that’s not the point. They already believe in design. Ignorance of a natural cause simply allows them to go on believing in it, confident that no one else can say otherwise…. like the “were you there?” taunt of so many who post here.

— Sylvilagus

Frank J: Sure, but the YEC and OEC alternatives also contain implicit "naturalistic" claims that are subject to "were you there" incredulity.

Granted, most adults, let alone students, never think to ask, and thus seem completely unaware of their double standard. But there's always a risk that a well-read student might call an unprepared teacher on it. Whatever anti-evolution activists thought they could get away with in the past, most of them know now that the best chance at success is the "don't ask, don't tell" approach that just promotes unreasonable doubt of evolution, and leaves it to the audience to infer design, a designers identity, and what the designer did, when, and how, in lieu of evolution.

I will expand on this in another article.

In other words, if one wants to make computer simulations one must repeat them for the entire range of possible values of the controlling factors.

First of all, neither in 1955 nor even in 1967 were the molecular genetics of cellular reproduction very completely understood.

It is important to keep in mind that at this time many thought that protein sequence differences that were observed were related to phenotypic differences.

Nevertheless, the state of it in 1967 would have made claims of non-existence of mathematical models of evolution a bit tenuous, even then.

Having painted themselves into a corner they seem to see little option other than continuing to charge ahead, or slink quietly away someplace into obscurity.

Sylvialgus,

Sounds like you talk to many people who are in the mid-ground between the "I don't come from no monkey" rank and file and the scam artists. The one who "converted" apparently recognized and overcame Morton's Demon, while the others are sill in it's grip. For the others, you might want to ask which of the mutually contradictory creationist accounts they find convincing (if any), and why all the others fail. If you can keep the discussion away from straying into a "design vs. no design" or "evolution vs. 'something else'" debate, it's possible that some of them might overcome the "demon."

Another thing you might mention to those most concerned with #2 is that, just because ID/"teach the controversy" is a religious idea inappropriate for public schools, does not mean that it brings God back in schools. In fact, as you probably know, many theologians criticize anti-evolution as bad theology, i.e. by pretending to catch God red handed, and confining Him to ever-shrinking gaps.

But, I believe that the comments of Professor Eden, in the first five minutes of his talk at least, refer to a random construction of such molecules and even those of us who are in the majority here, the non-mathematicians, realize that this is not the problem at all.

Did Dembski give an accurate representation? What do you think?

These miner-demons are presented with countless possibilities, but select only those that fit the required creationist environment.

We also need to note even when the mined gems don’t fit very well, they have very well trained specialists to make them fit. These master craftsmen use ellipses as a very powerful tool.

What confuses me is why the ID creationists do not spend the time and effort to correct these obvious flawed understandings of evolutionary theory?

What confuses me is why the ID creationists do not spend the time and effort to correct these obvious flawed understandings of evolutionary theory?

And again I ask, how could you be so certain about any kind of alleged misunderstanding when neither you or that panda guy has the source readily available for verification? Do you really believe that the 1966 meetings were actually that amenable to Darwin?

Ulam and Schrand’s subsequent paper confirms their Wistar paper regarding the extreme improbability asexual evolution. Their higher probability modeling on sexual reproduction does not negate this result. i.e., how evolution was supposed to have begun before sexual reproduction “evolved.”

Our first problem with the code name ADAM concerned asexual reproduction. We feel that the time scale to acquire a characteristic in an organism, such as the development of an eye, by a sequence of consecutive favorable mutations, is extremely long if one does not resort to something like sexual mating in the population. In the following rough and elementary estimate, the constants assumed are crude, but err toward "faster" evolution than what is to be expected. Definitions: Let T = time of existing life (~10⁹ yrs.); τ = time for one generation say (~ 3 days); G = the number of generations=T/τ=10¹¹; N= the existing population size (~10¹¹ ); k = the total number of "favorable mutations" necessary to produce the desired characteristic (~10⁶); α = the chance of a favorable mutation per individual per generation (~10^-10); γ = the "value" of a single favorable mutation expressed as a survival rate (~10^-6). That is, an individual having this mutation would have (in expectation) (k + γ) descendants, versus k descendants for an individual not having this mutation. Therefore, in the first generation, the expected value of the population that could have one mutation is Nα = 10. In 1/γ =10⁶ generations, a sizeable portion (approximately l/e) of the population would have this mutation, and in about 10⁷ generations, most of the population would have it. But the time to acquire all the mutations would be about k * 10⁷, or 10¹³ generations, which is like the age of the universe.

Dr. Mayr: This gamma you chose is really quite unreasonable. Fisher and Wright also took extremely low gammas when they started; but the recent work indicates that very often a mutant has a 30 percent advantage - in other words, up to 70 percent advantage. The average gamma, and particular the ones that do get incorporated, are really very much greater than you have here. Dr Ulam: No Doubt. On the other hand, a naive mathematician like myself would reason as follow: if you are going to produce a complicated objkect, say an eye, by many successive changes, any one of these gives only a tiny little bit of advantage for survival, if any at all. Dr Wald: It is very good. You have exaggerated the difficulties so your argument is going to be a minimum Dr Ulam: Oh, you will see that even so it doesn't come out so bad

PvM: UcD is still in denial

if there was a need for a clarification on the part of Ulam (four years later in 1970), then Dembski or any other subsequent Darwin dissenter should not be faulted for a so-called misreading that was perpetrated and caused by the same author four years earlier. Therefore, the onus fell on Ulam for failing to expound on the misleading details, for which he felt obliged to illuminate further in 1970, and not fellow darwin dissenters who had to wait four years for a clean-up PR explanation.

Perhaps PvM would do us the honor of reviewing Haldane’s Dilemma and commenting on the reasonableness of the parameters used by Haldane and then by Remine. That appears much more important than Ulam’s results. I would be very surprised to see evidence for thousands let alone millions of mutations sufficient to show Blind Watchmaker evolution.

Well if Dembski can't even read the paragraphs in the same paper immediately after the quoted one, he can hardly be faulted for not reading one published nearly 4 whole years later people. Maybe Ulam should have anticipated Dembski and inserted his contention that his arguments were gross first approximation over-simplifications in every single sentence.

What’s hilarious, tyharris, is that whereas initially Ulam was not qualified to issue his statement (because, of course, of its seemingly contradictive stance contra Darwin’s idea magnifica), now he is lauded as a purveyor of truth in light of a supposed recantation, notwithstanding the original critique of his having no qualification to speak of biological origins to begin with.

And they claim that Darwinism leads to amoral behavior.

Does anybody know why Sewall Wright, one of the constructors of neo-Darwinism, did not attend the conference?