Joined: Mar. 2003
Disbelief in the processes of mutation and natural selection as the sole cause of life is due to logical and scientific misunderstandings and misinformation. Evolution has been and will almost certainly continure to be the scientifically accepted view of biological origins for all of time. Further there is no scientific case for intelligent design.
In the following exchange I debate Michael Behe, a biochemist from Lehigh University, via email. A few very small changes have been made in the wording of a few sentences, grammar, spelling; but nothing significant to the debate in any way whatsoever.
The conversation exposes the complete fraud that is "intelligent design". I thank Behe for his time, and the discussion is, hopefully, due to continue at some point if he will indulge me in that.
My name is Clay Schentrup and I've completed the majority of my computer engineering degree at the Univsersity of Kansas. I've read quite a few excerpts of your work involving intelligent design, and in fact I've even seen Philip Johnson speak at KU a few years back. I also plan to see you speak this summer in Kansas City.
I understand you are busy so I'll make this as brief as possible. My question relates not so much to your research or the physical structures (like flagella) but to the reasoning behind your conclusions. I'll briefly illustrate with an example. Say I flip a coin 100 times and get the sequence "THHTHTTHHHHTT..." (you get the idea). Now once I have obtained that particular sequence of tails and heads, I could calculate that the odds off getting that particular combination would be 1/2^100. Now say I jump up and shout, "My goodness..out of all possible combinations that could have occured..the odds this one would occur were only one in several trillion trillion..therefore an intelligent being must have affected the trajectories of the coins to make them land in that particular pattern. This sounds quite strange, but it seems indistinguishable from your argument; that the genetic mutations involved with potentially creating a flagellum in an organism whose ancestors didn't have one, are so improbable that an intelligent entity must have had something to do with them.
I've been proven wrong about matters of science before, and I truly seek the best knowledge possible. But regardless of any level of detail you go into regarding these so called "irreducibly complex" systems, the fundamental premise is the same. "This event was so unlikely to have happened by chance that an intelligent being who wanted the event to happen must have caused it."
Perhaps you can explain my apparent confusion.
Thanks very much,
Hi, Clay, nice to meet you. Yours is a very good question, and goes to the heart of detecting design. It turns out that improbability is not enough to infer design, for the reason you apprehended. Any event is improbable. The exact structure of a pile of rubble is improbable, and the faces on Mt. Rushmore are improbable. So why do we feel compelled to conclude design in the latter case but not the former? The key is the notion of "specification". Not only is Mt Rushmore very improbable, but it also matches the faces of four presidents. Not only is the arrangement of the amino acids in the proteins of the flagellum improbable, but they also code for a functional machine, unlike virtually any other arrangement of amino acids. So the key is not just improbability, but improbability plus specification.
You might enjoy reading more about this. A good book for a computer fellow like yourself would be William Dembski's The Design Inference, published a few years back by Cambridge University Press.
First let me say that it is an honor and a pleasure to hear from you. I know that you are a busy person. I appreciate your literary suggestions, but I am already familiar with their underlying "design inference" argument, and that is specifically the topic with which I take issue. This is why I wrote to you personally. I am very patient, so if you could reply at your leisure I would very much appreciate it. Time is no issue. But I'll understand if you can't. I'll go ahead and reply to your letter, and see what comes of it.
> The exact structure of a pile of rubble is improbable, and
> the faces on Mt. Rushmore are improbable. So why do we feel
> compelled to conclude design in the latter case but not the
> The key is the notion of "specification". Not only is Mt.
> Rushmore very improbable, but it also matches the faces of four
Exactly. The monument "matches" known patterns of behavior of a specific known and observed process: humans. Humans are observed, and they are known to create visual replicas of other people, animals, trees, etc. So if we weigh the probability of human creation of those monuments against the probability of erosion and other non-human processes creating them, we can deduce humans. And we know that humans are "intelligent", therefore we can additionally deduce that the monument was intelligently created.
Flagella, on the other hand, don't match any known intelligent processes. I don't know what you mean when you say they "code for a functional machine". "Code", "functional", and "machine" all suggest intelligence to begin with, which is what you are trying to prove. I would state the matter more objectively. That particular sequence of amino acids chemically reacts forming a particular arrangement of atoms that we have dubbed a "flagellum". But how does this particular arrangement of atoms match any known intelligent process?
And now to get back to the pile of rubble. Say you were to visit an architect's house and snoop through his computer while he was in another room. And say you turned up well diagrammed plans for an arrangement of steel and other materials that precisely matched the "pile of rubble" you saw earlier. And say he came into the room and said, "Oh that..yeah I designed and built it. I'm into abstract art." Now suddenly you would find that the pile actually _had_ been intelligently designed. Now if you found another virtually identical pile elsewhere the distinct possiblity would still exist that it had been laid down with no intelligent intent and just had happened to fall that way. But because you would then be familiar with a known "process" with a greater-than-random liklihood of causing that particular (arbitrary) arrangement, you would be better to deduce that he had something to do with it.
I could go into analogy after analogy, but the point is glaringly obvious. "Specificity" is just "familiarity". You have cited numerous examples of cellular components that you believe to have been intelligently created. But the fact remains that you have not observed that God, or any other "process", has a greater-than-random propensity for causing these particular arrangements of DNA. In order for you to deduce that they were intelligently designed you have to show some process that is both intelligent and could has a propensity for designing such things.
For my final analogy back to your flagellum contention: Say I drop a bag of rocks on a gym floor and then drop a basketball on them and watch it roll, because of their inconsistencies and positions to one of the thousands of places it could potentially end up, and then mark off that spot. I could make the exact same argument you made. Not only is that location improbable, but virtually all other ways those rocks could have ended up on the floor would have caused the ball to land in a location other than that one. Therefore God, or some other intelligence, altered the seemingly random way in which they fell on the floor. It's the same argument you are using, so far as I can tell.
Thanks so much for your time,
Hi, Clay. You're right, like everyone else I'm rather busy. To compound matters, my wife had a new baby three weeks ago, so things are pretty hectic for us these days. So I'll be able to give you just a few more comments, and then sign off.
You are right that the "design inference" can give false negatives. I have pointed out in my book Darwin's Black Box and Bill Dembski has emphasized in The Design Inference that one can never rule out design, because intelligent agents can imitate random processes. However, the problem of false negatives is not the critical issue. The crucial issue is whether the design inference gives false positives. That is, does one ever come across something like Mount Rushmore and discover that it was the result of random processes? In your basketball example, suppose I had painted red all the rocks the basketball would bounce on, and marked with an "X" the spot on the floor where it would come to rest. If, when I dropped it, the ball did indeed hit all those marks, most people would easily conclude the ball was aimed, it wasn't bouncing randomly. The critical question is not whether a pile of rubble was designed, it's whether a system that looks very specific and functional was not designed.
You write that ""Code", "functional", and "machine" all suggest intelligence to begin with, which is what you are trying to prove." Perhaps those words do suggest intelligence, but they are not my words. They are widely used by the entire biology community. Biologists routinely speak of "molecular machines" that are "functional" and are "coded" for by an organism's genome. If you think it is incorrect to use such language, then your argument is with the biology community as a whole, not just me.
Congrats on your new baby. As to this discussion I don't expect a future response, but I would at least appreciate that you read my response in its entirety at your leisure. I do believe it goes into exceptional depth about the nature of the invalidity of your contention for the inference of design.
>..one can never rule out design, because intelligent agents can
>imitate random processes. However, the problem of false
>negatives is not the critical issue.
The only point I was making by bringing up the pile of rubble (intelligent design that one wouldn't initially think was designed)is that the only difference between it and Mt. Rushmore is that we know of a phenomenon with a substantial propensity for creating visual replicas of humans--humans! Were we to find a phenomenon with a propensity for making a particular type of pile of rubble, we would then be able to do the same thing with the pile of rubble, and if we ever saw that particular type of rubble pile in the future we would certainly deduce that it had been made by our intelligent architect friend than by being randomly dropped with no care as to its placement. There is nothing inherently different between the pile of rubble and the monument. Both have an overwhealmingly low probability of being in their particular arbitrary arrangement as opposed to all others. The one and only reason we deduce that intelligence made Mount Rushmore is that we know humans to be adept and likely to create such things, and we know that humans are intelligent. On the same token, were an alien race to visit our planet in the distant future, and have no familiarity with humans they would have no valid reason to deduce that Rushmore was intelligently designed. It would just be another "random" rock formation. It's just one of a zillion different ways those rock particles could be arranged.
> [the issue is] whether a system that looks very specific and
> functional was not designed.
Well for starters it would be nice to know what objective measurable property you are referring to when you say "specific". Can I analyze an object and state how "specific" it is with any sort of standard units? And if I can, and I agree that a flagellum, for example, is extremely "specific" (whatever that's supposed to mean) how is this significant to your argument? Are there any verifable observations of any intelligent phenomena which have a propensity for producing things which are highly "specific"? The only intelligent things I know of are humans. Do they tend to produce things which are "specific"? I don't know, but it really doesn't matter because we can already rule them out as the potential designers/builders of flagella, for obvious reasons.
And then there's this term "functional". As you pointed out scientists frequently speak of cellular components in terms of performing "functions". Yes, it is handy to use these terms because flagella remind us of things like motorboat propellers and other propulsion mechanisms. But the reason I took issue with your use of the term is that you go beyond just using it out of convenience, and instead truly believe that it _is_ performing a purposeful intelligently designed function.
In the case of flagella they whip microorganisms along through liquids. This is their "function". In the case of the water molecules a puddle in the road, they (the water molecules) perform the function of rapidly moving, bouncing around off one another and exerting minute gravitational and electromagnetic forces upon each other. Both the flagellum and the puddle of water are examples of arbitrary arrangements of molecules. But I don't hear you claiming that some intelligent entity purposefully and intelligently put those trillions upon trillions of water molecules in that specific arrangement.
The intelligent design argument, in a nutshell is as follows:
I. This structure (say flagellum) reminds me of something made by humans.
II. Humans are intelligent.
III. Humans couldn't have made this structure (obviously).
IV. Therefore some other phenomenon with algorithmically similar behavior to humans must exist and have designed/built this structure.
>The crucial issue is whether the design inference gives false
>positives. That is, does one ever come across something like
>Mount Rushmore and discover that it was the result of random
Put into more clear objective terms your question is, "What is the probability that an object which can be reasonably deduced to have been intelligently created is in fact not intelligently created?" The most obvious problem here is that you are specifically talking about objects which we have already deduced _are_ intelligently designed. To deduce that something is intelligently designed in the first place you would have already shown that there is a known intelligent process which has a significantly greater liklihood for producing the given object than any known unintelligent process. In the case of Mount Rushmore the reason we deduce in the first place that it was designed by humans, an intelligent process, is that the odds humans would make a large visual replica like that is far greater than the odds of its being created by erosion. And we know this only because we are familiar with human activities and thus we know very well that humans are prone to making sculptures, especially of other humans. And no other known processes are very likely to do so. So your question is like asking, "How often do you find eggs for sale in a sports store?"
The bigger issue though is that this has nothing to do with anything in biology. We _do_ know of an intelligent process (a human being) which has a tendency to create sculptures like Mt. Rushmore. We _don't_ know of any intelligent process with a tendency to create flagella, or any other cellular components. So you have no justification for comparing Mt. Rushmore to cellular components. The entire analogy is invalid.
If your conclusion is that the flagellum was intelligently designed, you must have some verifiable observations of a "phenomenon X" such that:
A) Phenomenon X has a tendency to alter DNA in such a way as to produce flagella (or any other cellular components you suspect as being intelligently designed) that is more likely than all other known processes (carcinogens, radiation, etc.)
B) Phenomenon X is intelligent.
C) Phenomenon X is older than any cellular components it may potentially be alleged to have produced.
I have yet to see any observations of any such phenomenon from yourself or any other ID scientist. And if you claim that "God" was the intelligent designer behind the "flagella alleles", then you would have to additionally satisfy a third requirement:
D) Phenomenon X has the necessary characteristics to fit the definition of "God".
The only known phenomenon that satisfies B is a human being, but it most certainly doesn't satisfy any of the other three requirements. Simply put, DNA, flagella, the blodd clotting cascade (activator, zymogen, substrate coagulogen), etc. are not consistent with any known intelligent phenomena, period. The design inference is scientifically empty. It stands purely because of the psychological habit of anthropomorphism.
And to thoroughly address the basketball analogy you made: In order to determine whether the basketball had landed on the X by chance, we would need two values. First the probability that the basketball _would_ land there by chance, and second the probability that the person dropping the ball would make it end up at it's destination intentionally.
Now the former we could calculate with reasonable precision by using data like floor surface area, the ball's size, and perhaps even some trial drops. Let's say we get some value, p, around one in a million, one in a trillion..whatever reasonably low number you prefer. Now as for the liklihood that the person dropping the ball (we'll call him Ed) would have intentionally altered it's course to land at its specific destination; that would be product of two other probabilities, we'll call k and m: k = the probability that Ed would _want_ the ball to end up in that particular spot, and m = the probability that a method exists by which this could be done.
Now most people are aware, mainly because of televised "magicians" like David Copperfield, that there is a dazzling array of techniques that various illusionists have employed to perform feats that have no immediate logical explanation. Typically though, when the explanations are revealed we are striken with how plausible they are; we simply wouldn't have thought of them on our own . And aside from all this, the extent of human technology is so vast that there would most certainly be a plausible manner by which Ed could make this happen if he so chose, which makes the theory of intentional placement at least slightly plausible. So m is greater than zero.
As to k, the probability that Ed would _want_ the ball to end up in its particular destination, there is a drastic difference between the ball's ending up on the X compared to its ending up on any other spot. If it did not land on the X we would have no reason to suspect he had intentionally caused it to land in that particular spot. There is simply no motive, because it wouldn't impress anyone. We have no reason to suspect that Ed would have gone to some great lengths if the ball landed on any particular unmarked spot on the floor. So the odds he would have wanted it to land in any such spot would be some extremely low value, which we could essentially deem insignificant or effectively zero. And even if we considered the chance that he did want the ball to land on that particular unmarked spot, the odds that he would have chosen that spot out of all potential spots is already essentially the same as the odds that the ball would have landed in that particular spot at random, by definition.
If the ball lands on a spot aside from the X, then k can be assumed to be extremely low. So mk is most certainly lower than the odds that the ball would end up in the destination spot purely by chance. On the other hand if the ball _did_ land on the the X, the probability that Ed had some intent for it to land there would be dramatically higher, simply because we know that this would be dramatically more impressive to the observers. And we know that people in general have a much higher tendency, in these types of situations, to do things that impress people as compared to things that don't. That is, we don't see people practicing for hours every week learning to shoot a basketball and make it pass directly through an imaginary hoop precisely 1 meter to the left of the rim, so if we were to see someone shoot a ball in that spot we wouldn't assume it was intentional.
So were the ball to land on the X we would have to consider the probability that randomness would have landed the ball there compared to the probability, mk, that Ed would have intentionally caused it to land there by some unknown technique. Because the human brain automatically tends to search for explanations to uncommon events which seem to fit a pattern consistent with known intelligent behaviors, we would initially feel compelled to see significance in the case that the ball landed on the X. That is we would search for some explanation which dealt with intelligence. Were we to fully investigate the matter with a large body comprised of physicists, illusionists, and detectives, and find some fairly plausible mechanism by which Ed could have directed the ball to that spot, we could possibly conclude that he had indeed caused it to happen intentionally and that it was not simply random, depending on how mk compared to p. This would especially be true if we found some evidence of such a technique, such as an electromagnetic system within the floor and some magnetic material hidden within the ball.
However if this body was to exhaustively search for such a technique and not be able to find one, we would have to conclude an extremely low value for m, and thus mk. If mk was less than p then we could conclude it was just a random event. And there's nothing impressive about that.
Now getting back to cellular biology with our flagellum example. Were we to know of any intelligent process with a probability of wanting to make flagella genes _and_ a probability of being able to, which, when multiplied together, were greater than the liklihood of flagellum DNA having been formed by unintelligent mutative processes [*that is, the probability that unintelligent mutative processes would create flagellum DNA], then we could reasonably conclude that this intelligent process was behind it. But as I've pointed out again and again you haven't shown any such process. And that's the critical hole in the intelligent design argument.
I very much hope to hear back. We'll see whether the ID movement has anything more substantial in their arsenal. Judging by this I'd wager not.