PBS Offers Intelligent Design Documentary

According to Illustra Media, the Public Broadcasting System uploaded the film Unlocking the Mystery of Life to its satellite this past Sunday. For the next three years, it will be available for member stations to download and broadcast. In addition, PBS is offering the film on their Shop PBS website under Science/Biology videos (page 4).

The film, released a little over a year ago, has been called a definitive presentation of the Intelligent Design movement. With interviews and evidences from eight PhD scientists, it presents strictly scientific (not religious) arguments that challenge Darwinian evolution, and show instead that intelligent design is a superior explanation for the complexity of life, particularly of DNA and molecular machines. The film has been well received not only across America but in Russia and other countries. Many public school teachers are using the material in science classrooms without fear of controversies over creationism or religion in the science classroom, because the material is scientific, not religious, in all its arguments and evidences, and presents reputable scientists who are well qualified in their fields: Dean Kenyon, Michael Behe, Jonathan Wells, Steven Meyer, William Dembski, Scott Minnich, Jed Macosko, and Paul Nelson, with a couple of brief appearances by Phillip E. Johnson, the "founder" of the Intelligent Design movement.

Check with your local PBS Station to find out when they plan to air it. If it is not on their schedule, call or write and encourage them to show the film. Why should television partly supported by public tax funds present only a one-sided view on this subject, so foundational to all people believe and think? We applaud PBS's move, but it is only partial penance for the Evolution series and decades of biased reporting on evolution.

This is a wonderful film, beautifully edited and shot on many locations, including the Galápagos Islands, and scored to original music by Mark Lewis. People are not only buying it for themselves, but buying extra copies to show to friends and co-workers. Unlocking the Mystery of Life available here on our Products page in VHS and DVD formats. The film is about an hour long and includes vivid computer graphics of DNA in action. The DVD version includes an extra half-hour of bonus features, including answers to 14 frequently-asked questions about intelligent design, answered by the scientists who appear in the film.

Intelligent Design Gets a Powerful New Media Boost 03/09/2002
Exclusive Over 600 guests gave a standing ovation Saturday March 9 at the premiere of a new film by Illustra Media, Unlocking the Mystery of Life. This 67-minute documentary is in many ways a definitive portrayal of the Intelligent Design movement that is sweeping the country. Intelligent Design is a non-religious, non-sectarian, strictly scientific view of origins with both negative and positive arguments: negative, that Darwinism is insufficient to explain the complexity of life, and positive, that intelligent design, or information, is a fundamental entity that must be taken into consideration in explanations of the origin of complex, specified structures like DNA. The film features interviews with a Who's Who of the Intelligent Design movement: Phillip Johnson, Michael Behe, Jonathan Wells, Paul Nelson, Stephen Meyer, Dean Kenyon, William Dembski, and others, who explain the issues and arguments for intelligent design as the key to unlocking the mystery of life. The film also features nearly 20 minutes of award-quality computer animation of molecular machines, manufacturing plants, and storage libraries of elaborate information - DNA and proteins at work in the cell, climaxing with a dazzling view of DNA transcription and translation.
In his keynote address, Dr. Paul Nelson (who appears in the film), gave reasons for optimism. He said that Time Magazine, usually solidly Darwinian, admitted just last week that these Intelligent Design scientists may be onto something. U.S. News and World Report is also coming out with a piece on I.D. And Stephen Meyer, who also appears in the film, could not be at the premiere because he was on his way to Ohio (see next headline), armed with copies of the film to give to the school board members. Nelson said that scientists should not arbitrarily rule design off the table. "Keeping science from discovering something that might be true is like having a pair of spectacles that distorts your vision," he said. "It does profound harm to science." He described how Ronald Numbers, evolutionist, once told him that design might be true, but science is a game, with the rule that scientists cannot even consider the possibility of design; "that's just the way it is," he said. (See this quote by Richard Lewontin for comparison.) Yet design is already commonly considered in archaeology, cryptography, forensics, and SETI, so why not in biology? Apparently this arbitrary rule has become a national controversy. Intelligent Design, says Nelson, is finally removing a "rule of the game" that is hindering science. If the reaction of the crowd at the premiere luncheon was any indication, Unlocking the Mystery of Life has launched a well-aimed smart weapon at the citadels of Darwinism.

Wow. Okay, I take back what I said...it was not "sorry". I did not read all posts, but rather just saw your response, which I thought was a little condescending. My apologies.

I'd really like to commend you for that -- I know how hard it is to publicly retract something, and most people's first reaction is to just quietly drop the issue or keep trying to support their initial statement. I give you big points for being able to change your mind and clearly say so. And I really appreciate it.

Now could you do me a favor (little quick to ask for a favor?) if you have time?

I have been short on time, so I apologize for the slow response, but I'm here now, and better late than never. :-)

I had posted a link to the programming code for the weasel program (the Java is downloadable from the site). Not sure if you have ever checked it out, but the program is the WORST representation of what I would consider simulating blind chance.

Well, part of the point is that evolution actually *isn't* just "blind chance". Because there is reproduction and selection involved, it can actually accomplish things that "blind chance" alone couldn't do in a billion years (literally).

It literraly took me about 5 minutes to see that the "blind" portion of the program does not exist.

Again, evolution itself isn't "blind" -- the selection process "looks" with a sharp eye on various candidates and (often harshly) selects among them.

In biological evolution, the selection process is due to what Darwin called the "struggle of life": living creatures use the results of their genetic inheritance (if any -- fatal mutations usually don't even get beyond the embryonic stage) in an attempt to survive long enough to reproduce successfully (and repeatedly, if possible). And generally, the ones which are better equipped (again, genetically) tend to be more successful at doing so than the ones which are less well adapted (or genuinely handicapped) to eat, live, and reproduce.

Mother nature is hardly "blind" -- unfit creatures die or fail to reproduce, more fit (i.e. faster, smarter, more efficient, more able to exploit an unused resource, etc. etc.) creatures will more often jump the hurdles and pitfalls of nature (which includes competition with or dangers from other creatures) and pass on their own genetic code.

In short, nature isn't "blind", it actually casts a very cold, sharp eye on all who live in it, and the cost of not measuring up is often death.

It uses simple random characters empiracally to get the end result. It actually goes through and checks individual characters, not the entire sequence, which would be more a little bit more realistic. I have taken simulation courses, as have you if you have a CS degree, and I gotta tell you, this is a poor, poor simulation.

It's not as poor as you think, but it is indeed *simplistic*. It was intended to be an educational tool, purposely simple so as to make a few points, not a grand simulation of biological evlolution in all its more subtle aspects.

One thing to keep in mind was that Dawkins' program was a *response* to certain creationist arguments. There's a broad range of creationist alleged "disproofs" of evolution which fall into the category best summed up in Hoyle's term, the "tornado in a junkyard".

These are the arguments which try to claim that it would take some insanely long time for [pick some biological structure/process] to come together "by chance". Hoyle's technique, for example, was to picture a tornado going through a junkyard, randomly rearranging the junk there over and over again, and then ask the reader to ponder how long it would take the tornado to randomly assemble a 747. Admittedly, it would most likely never happen at all.

The creationists then claim to have shown how ridiculous it is to believe that evolution could "randomly" produce anything complex, either.

The problem with such analogies, however, is that this is *not* an accurate picture of how evolution works. Yes, evolution does include a component of random change. But that's not the *only* process at work. Random variation (and not-so-random variation) is only one factor at work -- the other two are reproduction and selection. Without all three present, evolution goes nowhere (see below). *With* all three present, evolution can take off and do some very surprising things.

Some of the creationist "tornado in a junkyard" arguments involved the difficulty of randomly producing a particular string of text. This is the "monkeys on typewriters" variation, named after the famous old saying that if you had enough monkeys randomly smacking away on typewriters long enough, you could produce all the works of Shakespeare. True enough, but as the creationists rightly point out, for even a relatively short phrase the amount of time you'd have to wait for the monkeys to get lucky would likely exceed the expected lifespan of the universe itself.

And again, creationists use this to argue that evolutionary formation of something even more complex, say, a housecat is more unlikely than infinity squared.

But again, evolution is *not* just one random rearrangement after another, as are the tornado and the monkey examples.

So Dawkins decided to use an old creationist standby, the "monkeys on typewriters" example, as a starting point to put together a counterdemonstration that showed that using not just randomness, but all three of the processes at work in evolution, you could actually produce a desired text string not in mega-trillions of years, but in a few seconds! His point was to demonstrate that the three processs of evolution make for *enormously* more efficient results than the "pure chance" methods the creationists had been using in their arguments.

Dawkins' choice of phrase was indeed from Shakespeare:

Ham. Do you see yonder cloud that ’s almost in shape of a camel?
Pol. By the mass, and ’t is like a camel, indeed.
Ham. Methinks it is like a weasel.
Pol. It is backed like a weasel.
Ham. Or like a whale?
Pol. Very like a whale.
-- Hamlet. Act iii. Scene 2.

The Dawkins program to produce the string "Methinks it is like a weasel" involves three processes:

1. Random variation -- on each "generation", 1/8th of the character strings in the "population" (size selected by user) have one of their text characters completely randomized to some other character.

2. Selection -- the character string which has the most "correct" characters (or if more than one such string exists, the most recent such) is flagged, and a) will be "bred", and b) won't itself be mutated or replaced by one of its own "offspring".

3. Reproduction -- the current "most fit" character string undergoes "sexual reproduction' with randomly chosen other strings, and the resulting offspring replace the "mates". (This is actually more akin to biological lateral gene transfer.)

So all three of the processes necessary for evolution to take place are in the Dawkins program. And, as predicted by "evolutionists", the results are swift and sure -- the mutating, reproducing, subject-to-selection population very quickly (within seconds) produces a Shakespeare text string which the creationist "pure random" methods would not have produced before the Earth permanently froze over.

Things to note: While the selection function "tells" the population which character string among the population is "closest" to the target, it gives no hint whatsoever as to *which* characters in that string are right and which are wrong, nor how close the wrong ones might be, nor what/how to change the string to get closer or farther from the target. The population itself doesn't even "know" what the target string is (or technically, that there even *is* a target). All it "knows" is that certain members among its population are more favored (i.e. fit) for reproduction than others. But the population has no way to "know" or predict what changes might increase or decrease the reproductive fitness of any member of its population. Indeed, the result of each new mating and/or mutation is more likely to *decrease* the amount of match than to increase it.

This is why it's not incorrect to call the process "blind evolution" -- the population itself, and the "individuals" in it, as well as the mutation process and the reproduction process, have no "goal" in sight, have no "preferred" direction, have no knowledge of the selection function's existence nor what type of change would increase/decrease fitness. They just keep on mixing and matching and mutating, living or dying, while being totally blind to any aspect of the force (selection) at work shaping things. The individuals in the population (nor the population as a whole) is not "trying" to reach any particular result. Mutation in the model is purely blindly random (and actually is much more likely to take an individual farther from the desired result than closer to it). The mating "genetic" crossover is similarly blindly random and more likely to produce an offspring which matches less than its best parent. The only non-random component is that which lets the most "fit" individual in the population do more of the mating, but even then that individual is blind to the reasons why it was more successful than its kin.

It's also instructive to note what happens if you remove any of the three required processes:

1. Without a selection process which gives the "more fit" individuals greater reproductive success than the less fit, the population will simply randomly mate and recombine and mutate forever, always consisting of just randomly shuffled gobbledegook.

2. Without reproduction which multiplies the "good" genes, the population just stagnates in place and randomly mutates forever, in a result almost identical to the original "monkeys on typewriters" scenario. (Actually, in the Java program under discussion, the selection process causes the current "best" match to cease mutating, until some other individual gets astronomically lucky and just happens to roll the dice into a better match, at which point *it* stops mutating and the previous "best" begins to just degrade into randomness again -- end result is pretty much the same as the "pure monkeys" case).

3. Without mutation, the few letter sequences in the initial population will just get reshuffled endless (by reproduction), shaped by the selection process. But without any new "information" available, it may well be that the goal result can never be reached. For example, if no character string in the initial population has an "M" in the first character slot, no amount of reshuffling by sexual reproduction will ever cause one to be there, and the final string will always remain impossible. Worse, genetic drift and reproductive amplification of "partially fit" individuals with many "wrong genes" will quickly remove the initial random variation from the population, and cause many "right" letters to be shuffled out of the gene pool entirely, further increasing the odds of reaching an impossible situation where certain letters are completely unavailable at certain letter positions.

All in all, the Dawkins program is a very simple, but very instructive, evolutionary mini-experiment. It's meant to show the creationists what's wrong with their "monkeys on typewriters" analogies, show them what's much more powerful about evolutionary processes, and to give a quick introduction to what those processes are and how they work.

However, there are a few ways that the "weasel" program is somewhat misleading.

The primary way is that in the program, "genotype" and "phenotype" are identical. In biological evolution, and in most useful genetic algorithms used for solving real-world problems, they are not.

Definitions: "Genotype" is the particular "coding sequence" which defines an individual in organisms or genetic problem-solving algorithms -- for living creatures the genotype is its actual DNA sequence. "Phenotype" is the actual unique individual which results from the genotype, i.e. the actual plant or animal in biology, or the specific problem solution defined by the mathematical encoding used as genetic material in genetic algorithms.

The Dawkins program trivializes this distinction by making the "DNA" of the individuals the actual character string itself. But in biological systems, for example, there's not a one-to-one correspondence between genotype and phenotype. For example, there may be more than one gene sequence which happen to produce blue eyes. Furthermore, there's a richness and unpredictability in the way that small genotype changes may produce large results in the phenotype.

Another way that the Dawkins program can be misleading is that uses a distinct, specified "goal" as its fitness function. Biological evolution is much more complex in its workings because there's no single "match" that is "best". In fact, in biological evolution, the way that different living things compete against each other (not just within their own species, but predator/prey dynamics, etc.) causes an ever-shifting "arms race" which means that what might be "best" (i.e. "most fit") this generation may not be what's most likely to ensure survival 20 generations from now (not to mention changing climates, landscapes, etc.) Biological evolution is a constantly shifting game of "whatever works today, in the current place and situation".

Additionally, in real-world applications of genetic algorithms for problem solving, you don't know in advance what "solution" you want to pop out of the "evolution box". This is another way the Dawkins program is misleading -- it makes it look as if you have to pre-program the solution you want (in which case, why bother, right?) Actually, though, what's usually done is that you only know *what* you want solved (e.g., you want an electronic circuit that does cubic function generation better than human-designed circuits), but you don't know *how* to best solve it (i.e., what circuit would actually do the job well). So you set up computerized evolution such that a "match" is defined only by how well each computer-evolved "individual" (i.e., electronic circuit) performs on the "fitness test" (in this case, how well it produces cubic functions as output), without *any* predefined specification of how the circuit should actually be configured internally -- *that's* what you let evolution discover *for* you. This works because you're asking only for a particular *property* of the evolved result, not the entire specification of the result as was the (purposely simplistic) case in the Dawkins program. Evolution does the rest, coming up with often surprising results which (because of the fitness function) do the thing you wanted it to do without having to know how to do it yourself in advance.

And similarly, this is how nature can produce biological creatures which "solve" the problem of how to survive and reproduce ever more effectively, without nature having to "know" how to "design" a wing, or a placenta, or a clotting mechanism. Evolution "blindly" stumbles upon those solutions as mutation/variation randomly explore the almost infinite variations possible with DNA, selection filters the results to keep those which work better and discard those which don't, and reproduction amplifies the best ones to repeat the process in the next generation, millions of times over and over again.

My point with this program was that...well honestly, you would have to look at the posts. Evolutionary and genetic programming were brought up. I would love to hear your thoughts on my comments about the programming issues...no one else seemed to be interested or did not have a tech background which would include programming.

If the above doesn't address all your questions, let me know.

Now, I haven't read Dawkin's book, so I don't know if what is represented is modeled accurately.

Actually, neither have I, although I've read parts. I don't know whether he addresses some or all of the points I made above, but it's likely he did. That makes it all the more unforgiveable when his critics misrepresent him (for example, see below).

But I did find links later from a Biochemist, who also has a CS degree, who basically called it "nonsense". If you are interested, I can run down all the links.

I found the link you posted. The "Answers in Genesis" rebuttal gets a number of things flat wrong, like when it writes, "The experiment is repeated for only the positions where a match did not occur". This is very, very wrong, as a quick glance at the Java program will verify. What's interesting is that as I mentioned above, the "population" of character strings *doesn't know* which character positions have stumbled upon the "right" answer and which positions are wrong. Furthermore, both mating recombination and mutation have a greater chance of *messing up* correct letters than they have of "correcting" wrong letters. And yet, *the system still converges on the answer* nonetheless, because of the way that mating selection tends to multiply the percentage of "right" characters in the gene pool.

The rest of the lengthy "analysis" on that page is based on the same flawed misunderstanding of how Dawkin's process works and is therefore invalid as a critique of Dawkin's process (and thus evolution itself).

Furthermore, the section "Exposing the implied assumptions with another analogy" has problems of its own, apparently designed to make the notion of evolution seem as absurd as possible -- by among other things removing a necessary part of the evolutionary process, reproduction. The "another analogy" the author sets up, earthquakes in a pile of house pieces (hmm, "tornado in a junkyard", anyone?) is set up *without* replication of partial results, which both makes the process non-evolutionary (and impossible to proceed as evolution does), and ludicrously unlikely, as the author correctly points out. Now the only question is, did he *intend* to make a straw-man invalid analogy in order to falsely ridicule the notion of evolution, or does he actually misunderstand evolution badly enough that he *thinks* this is a good comparison? Neither option inspires confidence.

Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.