Why Darwinism's quest for a free lunch re:information is doomed.

Probabilistic Programming and Conservation of Information

Evolution News & Views May 14, 2015 3:53 AM 

MIT scientists have improved on a new kind of computer language that enables "probabilistic programing," Larry Hardesty writes for MIT News. The new algorithms allow a programmer to write in fifty lines of code what used to take thousands. Hardesty emphasizes that this advance came at a high cost. And interestingly that is just what William Dembski shows in the chapter on "Conservation of Information" in his newest book,Being as Communion.

Let's say you want to write facial recognition software. This task, which humans perform so effortlessly, has long been a challenge for artificial intelligence. Traditional attempts have been deterministic: trying to specify all the variables. Probabilistic programming, by contrast, iterates attempts at a match till an acceptable level of success is achieved. "It means canvassing lots of rival possibilities and selecting the one that seems most likely," Hardesty explains. An MIT-developed language called Picture uses this method:

In a probabilistic programming language, the heavy lifting is done by the inference algorithm -- the algorithm that continuously readjusts probabilities on the basis of new pieces of training data. In that respect, Kulkarni and his colleagues had the advantage of decades of machine-learning research. Built into Picture are several different inference algorithms that have fared well on computer-vision tasks. Time permitting, it can try all of them out on any given problem, to see which works best. [Emphasis added.]

The problem fits exactly what Dembski describes in his discussion of "search for a search" (S4S). In his previous technical book, No Free Lunch, Dembski proved from the "No Free Lunch" (NFL) theorems that no evolutionary algorithm is superior to blind search, once all the costs of obtaining additional information are factored in. For instance, the cost of a search for a treasure map meets or exceeds the cost of digging around an island at random. The treasure hunter first has to pay the cost of searching for the right treasure map -- a search for a search. The cost of S4S has to be cashed out in information from some other source or by trial and error; thus, information is conserved.

In the current case, the MIT computer scientists are also conducting a search for a search: finding a method for recognizing faces out of all possible algorithms that attempt the task. Notice that the researchers "had the advantage of decades of machine-learning research." This had reduced their task to "several different inference algorithms that have fared well on computer-vision tasks." In other words, they didn't find Picture's algorithm by blind search. They first had to pay the cost of all that prior information, just like the treasure hunter had to either receive or test information to find a reliable treasure map.

The article might seem to suggest that you can still find a successful target for free by letting the machine learn the information:

Moreover, Kulkarni says, Picture is designed so that its inference algorithms can themselves benefit from machine learning,modifying themselves as they go to emphasize strategies that seem to lead to good results. "Using learning to improve inference will be task-specific, but probabilistic programmingmay alleviate re-writing code across different problems," he says. "The code can be generic if the learning machinery is powerful enough to learn different strategies for different tasks."

But do you see where the cost of information snuck in? The machine will never "learn" anything by blind search without the intelligent programmer providing it with the right information on what constitutes a success. Information must also be supplied in both the hardware and software from outside the machine. Conservation of Information is not violated.

How does this principle intersect with evolutionary theory? In both books, Dembski shows that no evolutionary algorithm improves on blind search, once the accounting is done. Darwinian biologists mistakenly think that natural selection is an algorithm that drives organisms to higher fitness. The mathematical models that attempt to show this, however, all violate Conservation of Information by inserting intelligently designed information into the model.

The well-known analogy to evolution from Richard Dawkins's book The Blind Watchmaker is a case in point. Dawkins got a computer algorithm to write the Shakespearean line "methinks it is like a weasel" in a finite number of iterations, starting from a random sequence. What he failed to factor in was the information he himself provided as to what the target was, and what constituted a success. The same oversight plagues all evolutionary algorithms and models, such as Avida. Accounting for the extra information proves that no evolutionary algorithm is superior to blind search.

Dembski writes:

The point to realize is that those whose job it is to find the right optimization procedure (e.g., operations research people) must, for a given search, try to choose one search strategy that works well to the exclusion of other search strategies that work less well. Such a choice of strategy identifies one possibility to the exclusion of others within a matrix of possibility (the matrix, in this instance, consisting of search strategies, or classes of different types of searches). It follows that such a choice of search strategy entails an input of information into the search in the exact actualize-exclude informational sense emphasized throughout this book. (Being as Communion, pp. 149-150)

What about biology? Don't organisms succeed at their searches for fitness when they adapt to changing environments? Of course they do, but only because of the information programmed into them. It's analogous to the cost of the hardware, software, and algorithms in a computer. Such information is explainable by a mind, but is exceedingly improbable by blind search.

MIT's advance in probabilistic programming is commendable, but it illustrates the tremendous informational investment that went into being able to reduce programs to fifty lines of efficient code: decades of research, testing, and insight. It even cost a lot of money:

To make machine-learning applications easier to build, computer scientists have begun developing so-called probabilistic programming languages, which let researchers mix and match machine-learning techniques that have worked well in other contexts. In 2013, the U.S. Defense Advanced Research Projects Agency, an incubator of cutting-edge technology, launched afour-year program to fund probabilistic-programming research.

When hearing reports about the success of evolutionary algorithms that appear to explain the spontaneous emergence of complex organisms, be sure to look for the information snuck in the side door. Information always comes at a cost.

Being as Communion is the third of Dembski's books that, after The Design Inference and No Free Lunch, present the theory of intelligent design with rigorous logic and math. It's also the shortest and most accessible to the layperson.

The fundamental stuff of the universe is not matter and energy, Dembski argues, but information. If you're interested in learning more about Conservation of Information, search, probability, and related concepts, check out this engaging book that not only explains the principles clearly and cogently, but defends a startling new way of looking at reality.

Argument by asertion and insult?

Contradictions, Irony, and Appeals to Authority Permeate The Language of Science and Faith

Casey Luskin May 31, 2011 12:38 PM

The title of Chapter 1 in Karl Giberson and Francis Collins' book The Language of Science and Faith is "Do I Have to Believe in Evolution?" The very title of the chapter itself implies that affirmative belief in evolution is an indisputable matter for Christians. If you doubt that they are so adamant, bear in mind that it was Giberson himself who recently wrote that "Jesus would believe in evolution and so should you." They're welcome to believe in evolution, but should they present their case in such a non-inclusive way? But before delving further into the rhetorical strategies of The Language of Science and Faith, let's revisit some of the main scientific evidence and arguments they cite in this chapter for why you "have to believe in evolution":

repeated appeals to a singular pseudogene which they call "broken DNA," which is shared by humans and other primates (this turned out to be a weak argument since we're continually finding new function for pseudogenes);

feathers evolved from scales (an argument that has been abandoned by evolutionary biologists);

eyes evolved from light-sensitive pigments (they gave no evidence to help the reader accept this claim, but the evidence strongly challenges their assertion)

humans share common ancestry with Neanderthals (many consider humans and Neanderthals to be so similar that they are -sub-races of the same species, so this doesn't show common ancestry with some non-human type); and finally...

macroevolution is simply repeated rounds of microevolution (they provided no empirical evidence for this claim other than bald assertions and misguided appeals to intelligently designed technological evolution).

Giberson and Collins want the reader to feel that the answer to the question posed in the chapter title is an unequivocal 'yes.' But as we've seen, the evidence cited in this chapter was extremely unimpressive. Their arguments were outdated and in some cases consisted of little more than bald assertions that is contradicted by the evidence. But again, they are welcome to 'believe in' neo-Darwinian evolution if that's what they want. We can agree to disagree and I have no problems with the making a case for their viewpoint. They have their right to believe what they wish; likewise there's nothing wrong with logically critiquing their argument in a civil manner to show why I find it unpersuasive. 

But there's another quality to The Language of Science and Faith that is much more disconcerting. The troubling part about The Language of Science and Faith is the level of rhetoric that that Giberson and Collins use while trying to lead readers to conclude that they "have to believe in evolution." The book is full of appeals to authority and attacks upon the character and competence of Darwin-doubting scientists. It is into this most unfortunate territory that we now cautiously step. 

Appeals to Authority
Theistic evolutionists seem to love making appeals to authority. Last year we saw Biologos president Darrell Falk rely heavily upon appeals to authority in his book Coming to Peace with Science. It seems that Giberson and Collins have followed Falk's rhetorical strategy. Thus, in chapter 1 of The Language of Science and Faith alone, we find comments like:

"almost all Christian biologists accept evolution." (p. 30)

"in most large gatherings of scientists you would not find even one person who rejects the theory of evolution." (p. 30)

"The scientists at the BioLogos Foundation are unaware of anybiologists who have abandoned evolution in the past few years. Not one." (p. 31)

"we are equally unfamiliar with any premier scientists who reject evolution." (p. 31)

"Christians should take no comfort in the misplaced hope that the scientific community is gradually abandoning the theory of evolution." (p. 34)

"the validity of scientific ideas is best addressed by the leading experts." (p. 33)

And just in case you didn't get the message, they even go so far as to claim that "[v]irtually all geneticists consider that the evidence proves common ancestry with a level of certainty comparable to the evidence that the earth goes around the sun." (p. 49) (Note: this is from the paragraph where they again cite a singular pseudogene as a "broken" gene.") 

Get it? The not-so-subtle message is that if you doubt universal common ancestry, then you're no better than a geocentrist. I get the sense that Giberson and Collins don't want people to think for themselves on topics like evolution, but to simply capitulate to those whom they deem "the leading experts." 

I wasn't sure if the book actually intended such a message until I read Giberson's recent response to William Dembski where Giberson makes it very clear that he doesn't want people thinking for themselves on topics like evolution. He writes:

To suggest that this "data" can be handed over to non-specialists so they can make up their own minds is to profoundly misunderstand the nature of science.

Stop and think for a moment about what Professor Giberson just said. 

Dr. Giberson doesn't think that the average person should be allowed to "make up their own minds" on evolution. 

Unfortunately, this mindset is becoming more and more typical of the Darwin lobby: they want people to stop thinking for themselves. Berkman and Plutzer's survey in Science from earlier this year went so far as to criticize a teacher who felt that "[s]tudents should make up their own minds" on evolution "based on their own beliefs and research." Such a mindset seems profoundly dangerous and wrong in many ways. In fact it seems, dare I say it, anti-intellectual because it demands intellectual conformity from all. 

Self-Contradictory Argument
What's most ironic, however, is that Giberson and Collins later defeat their own appeals to authority. They write that "scientific truth is not decided by the number of names on a list, or who wins the debate or convinces the most people. It is based on the evidence." (p. 33) Here I completely agree! But if they really believe that statement, and the evidence is all that matters, then why do they feel the need to persistently rely on appeals to authority? 

I think that Giberson and Collins want to have their cake and eat it too. On the one hand, they want readers to imbibe a not-so-subtle message that they ought to capitulate to experts who 'believe in" evolution because those experts are in the majority. Realizing that's a fallacious and dangerous argument, on the other hand they want to be able to say that they still made the correct argument that that scientific truth is "based on the evidence," not "the number of names" who support a theory. They're completely right about this latter point, but they seem to be sending contradictory messages, not giving "straight answers" to these questions. 

But there are other contradictions that are even more troubling yet to be found in The Language of Science and Faith. 

Attacking Darwin-Doubting Scientists
What's most unfortunate about the The Language of Science and Faith is not the book's questionable and self-contradictory appeals to authority. Nor is it the book's apparent attempt to dissuade people from thinking for themselves on evolution. Rather, it's how Giberson and Collins try to win the hearts and minds of their readers by using the rhetorical strategy of attacking the competence and character of Darwin-doubting scientists. 

Let's start with their attacks on the competence of Ph.D. scientists who have bravely signed a list dissenting from neo-Darwinian evolution. In one instance Giberson and Collins write:

"The 'Dissent from Darwin' list includes philosophers, physicists, mathematicians and academics from other fields. Many of them never took even a single course in biology beyond high school." (p. 32)

This is a bald assertion--and there's no evidence at all that it ought to be taken seriously. And how do they know that is true? Their intended point, of course, is that these Ph.D. scientists are supposedly not qualified to express real dissent from Darwinian evolution. In fact, they make this argument explicit, stating: "Many of the scientists listed are not trained in biology and so are not in a position to evaluate the central theory of the field." (p. 32) 

What's ironic (and contradictory) about their argument is that the book's first author, Karl Giberson, is not trained in biology and even admits that he "took his last biology course in 1975." (p. 32) Now I'm not raising this point to attack Giberson's knowledge or credentials, and in fact I don't think that Dr. Giberson is unqualified to comment on evolution simply because he is not a trained biologist. In fact Giberson--who is the first author of The Language of Science and Faith--must feel the same way since he just wrote this book which spends many pages evaluating evolution. 

Giberson's admission that he lacks the same qualifications of those whose qualifications he attacks does not exactly make for a compelling argument. Given that he's writing extensively about evolution, his attack on the competence of Darwin-doubting scientists seems downright hypocritical. 

Giberson and Collins have also misconstrued the purpose of the Dissent from Darwinism list. As noted, they write that "scientific truth is not decided by the number of names on a list, or who wins the debate or convinces the most people. It is based on the evidence." Again, I could not agree more. But the "Dissent from Darwin" list was never intended to demonstrate that neo-Darwinian evolution is false simply because x number of scientists disagree with neo-Darwinism. What the list does demonstrate is that one cannot dismiss scientific dissent from neo-Darwinism simply by appeals to authority. The list shows that there is a critical mass of credible scientific dissent from neo-Darwinism. 

By overstating the purpose of the Dissent from Darwinism list, Giberson and Collins hope they can convince the reader to dismiss it. But when used properly, the list defeats their own quite explicit argument, that one ought to ignore scientific dissent from Darwinism because of appeals to the authority of the majority viewpoint. 

But their most unpleasant jab at Darwin-doubting scientists is yet to come. They write:

"The evangelical literature is so filled with misrepresentations and outdated information about evolution that even a lot of research might not lead an honest seeker to the truth." (p. 34)

To be sure, there have been some Christians who challenged evolution with bad arguments. But by making broad-brushed attacks upon the entire body of Christians who would argue against neo-Darwinism, Giberson and Collins apparently want to paint all Christians who doubt neo-Darwinism in a negative light. 

That faux pas aside, again, we see irony in their statement: If their claim is true, then Giberson and Collins' book defeats its own argument, since The Language of Science and Faith uses outdated arguments that feathers evolved from scales. 

The 'Seeds of Doubt' Strategy Fails
The rhetorical strategy of Giberson and Collins is now becoming clear: they want to plant seeds of doubt in the minds of readers not through discussing the evidence, but by attacking the trustworthiness and competence of skeptics of neo-Darwinian evolution. Aside from the fact that this implied argument of commits the genetic fallacy, Giberson and Collins haven't done a very good job of making this argument, because every attempt to plant doubt in the reader's mind is refuted by their own book:

Giberson and Collins make repeated appeals to authority but then admit that all that matters is the scientific evidence;

Giberson and Collins attack the qualifications of non-biologist scientists who doubt evolution, but Giberson himself is a non-biologist who lacks training in biology and evolution.

Giberson and Collins attack the "misrepresentations and outdated information" among writings that challenges evolution, but their own book is far from mistake-free, and uses some unambiguously outdated and now-abandoned arguments about evolution.

I raise these points not because I want you to distrust Giberson and Collins. They're both very smart, qualified, and informed scientists, and they make arguments worth considering. Read their books. Evaluate their arguments. Read the responses. Decide for yourself if they make a persuasive case. 

But also realize that there are a lot of smart, qualified, and informed scientists who doubt neo-Darwinian evolution. The fact that some of those scientific critics may happen to be "evangelicals" does not mean that therefore those responses are "filled with misrepresentations and outdated information about evolution." Giberson and Collins are making arguments designed to stifle your own self-investigation, not encourage it. 

Whenever Giberson and Collins stop talking about the evidence, and start appealing to authority, attacking Darwin-skeptics personally, and asking you to stop thinking for yourself, it's time to become skeptical: If the evidence is on their side, why do they feel the need to do this? 

On Darwinism's sexual issues.

Sex, the Queen of Problems in Evolutionary Biology

The origin of sexually reproducing organisms from asexually reproducing ancestors is a profound mystery which has baffled many an evolutionary biologist. The origin and subsequent maintenance of sex and recombination is a phenomenon not easily explained by Darwinian evolution. Indeed, there are several substantive, well-known reasons why the origin of sex presents a serious problem for conventional evolutionary explanations. Graham Bell described the dilemma in his book, The Masterpiece of Nature: The Evolution of Genetics and Sexuality:

Sex is the queen of problems in evolutionary biology. Perhaps no other natural phenomenon has aroused so much interest; certainly none has sowed as much confusion. The insights of Darwin and Mendel, which have illuminated so many mysteries, have so far failed to shed more than a dim and wavering light on the central mystery of sexuality, emphasizing its obscurity by its very isolation.

What Are The Problems?
There are several reasons why the origin of sex presents a problem. For starters, there is the waste of resources in producing males. Assuming a sexually-reproducing female gives birth to an equal number of male and female offspring, only half of the progeny will be able to go on to have more offspring (in contrast to the asexually reproducing species, all the offspring of which can subsequently reproduce). Thus, it is to be expected that the asexual female will proliferate, on average, at twice the rate of the sexual species. Given the disadvantage thereby confronting the sexually-reproducing species, one would expect them to be quickly outcompeted by the asexual species. Moreover, it must be borne in mind that, in contrast to the asexual species, the females of the sexually-reproducing species perpetuate only half of their successful genotype. To transition, therefore, from a state of asexuality to sexual reproduction is, in effect, to gamble with 50% of one's successful genotype. Given that the whole purpose of natural selection is the preservation of those organisms which pass on their successful genes, this strikes at the heart of evolutionary rationale.

These dilemmas are nothing new. In his book, Evolution: The Triumph of an Idea, Carl Zimmer acknowledges these problems:

Sex is not only unnecessary, but it ought to be a recipe for evolutionary disaster. For one thing, it is an inefficient way to reproduce...And sex carries other costs as well...By all rights, any group of animals that evolves sexual reproduction should be promptly outcompeted by nonsexual ones. And yet sex reigns... Why is sex a success, despite all its disadvantages?"

The problems extend even deeper than this. For there is, of course, the additional conundrum related to the fact that gametes (i.e. sex cells) undergo a fundamentally different type of cell division (i.e. meiosis rather than mitosis). Meiosis entails the copying of only half of the chromosomal material. In similar fashion to mitosis (which occurs in somatic cells), each chromosome is duplicated to yield two chromatids. In contrast to mitosis, however, the homologous chromosomes are also associated. So, at the start of meiosis, each visible 'chromosome' possesses four chromatids. At the first division, these homologous chromosomes are separated such that each daughter nucleus has exactly half the chromosome number. At this stage, each is present as two copies (chromatids). These chromatids are hence separated at the second division such that each new nucleus only has a single copy. In order for sexual reproduction to work, it is essential that the process of meiosis evolve to halve the chromosome number. And this ability must also only occur in the gametes and not in the somatic cells. This difficulty is accentuated by the multitude of novel elements which are found in meiosis, rendering it unlikely to be explicable in terms of single mutational steps.

And then there is the added problem of male and female complementarity -- a seemingly remarkable incidence of co-evolution.

Attempted Solutions
On the same day last week, two articles were published in the popular science media offering two different (but completely contradicting) accounts of the evolutionary origin of sexual reproduction.

One of these articles bears the headline, "Sex evolved to prevent parasite infections, say scientists". The article opens by acknowledging the problems which I outlined above:

Sexual reproduction that involves two partners is far less efficient than self-fertilization -- at least, from the perspective of evolution. So why did creatures like humans ever start having sex with each other? According to a new study, we did it to fight parasites. We spoke with the researchers to discover what this says about sex.

The article goes on to explain,

Biologists at the University of Indiana found some of the most convincing evidence yet that the evolutionary driver of sexual reproduction is a need to avoid death by parasites. The basic logic is that, if an organism reproduces asexually, then the genetic variation of its species as a whole will slowly grind to a halt, and it becomes increasingly likely that a parasite that can kill one member of the species can wreak havoc on the entire population. (For proof of that, just look at bananas.)

Sexual reproduction, then, serves as a way to keep introducing genetic variety, a process that has to constantly be repeated in order to continue staving off attacks the latest and deadliest parasites. This is known as the "Red Queen Hypothesis", taking its name from a line in Lewis Carroll's Through the Looking Glass in which, "It takes all the running you can do, to keep in the same place."

And thus, it is argued that the apparent disadvantages of sex, outlined above, are compensated by the advantage which comes from the immense genetic flexibility which arises from the shuffling of genes through the process of recombination and fertilisation. This entails that the population is in a better position to enable it to adapt more readily to environmental hazards such as parasites. This would thereby give the sexually-reproducing species a significant advantage, while rendering the asexual species more susceptible to extinction.

But here's the thing: the fact of this obvious advantage does NOT explain how sexual reproduction arose in the first place. Indeed, such immense genetic flexibility is of benefit only to future generations, and not to the present population. But natural selection, being devoid of foresight, is not able to retain biological phenomena for their potential future utility. The problems which I outlined above are potent short-term disadvantages which should have mitigated against sexual reproduction evolving in the first place! Moreover, sexual reproduction is a phenomenon of such complexity sufficient to render it extremely unlikely to evolve by mutation on a frequent enough basis that we might expect it to become fixed by virtue of a few organisms somehow surviving these obvious disadvantages.

Moreover, the efficacy of the Red Queen Hypothesis has been increasingly challenged of late. As noted in one BBC article in 2004, "for this theory to work, there have to be an awful lot of parasites about, and they have to have very dramatic effects."

As John Maynard Smith stated in chapter 12 of his book, The Theory of Evolution,

We are therefore driven to the conclusion that the early stages in the evolution of the sexual process took place under the influence of selective forces quite different from those which are responsible for the maintenance and spread of sexual processes once they were erected.

The second article, published on Science Daily, bore the heading, "Sex Is Not About Promoting Genetic Variation, Researchers Argue". This article acknowledges the problem with the hypothesis in the previous article, and reports,

Heng and fellow researcher Root Gorelick, Ph.D., associate professor at Carleton University in Canada, propose that although diversity may result from a combination of genes, the primary function of sex is not about promoting diversity. Rather, it's about keeping the genome context -- an organism's complete collection of genes arranged by chromosome composition and topology -- as unchanged as possible, thereby maintaining a species' identity. This surprising analysis has been published as a cover article in a recent issue of the journal Evolution.

"If sex was merely for increasing genetic diversity, it would not have evolved in the first place," said Heng. This is because asexual reproduction -- in which only one parent is needed to procreate -- leads to higher rates of genetic diversity than sex.

The article goes on to ask,

In fact, two billion years ago in Earth's biosphere, life relied exclusively on asexual reproduction, and every organism was capable of bearing young without costly competition to mate. With asexual species' faster and more efficient mode of reproduction, the origin and maintenance of sex -- not exactly the fittest means of reproduction -- puzzles scientists, who for decades have been asking, Why has sex evolved and survived?

And what solution are we offered this time, you might be wondering?

According to Heng, the hidden advantage sex has over asexual reproduction is that it constrains macroevolution -- evolution at the genome level -- to allow a species' identity to survive. In other words, it prevents "Species A" from morphing into "Species B." Meanwhile, it also allows for microevolution -- evolution at the gene level -- to allow members of the species to adapt to the environment.

That's right. Sex evolved in order to limit evolvability. Apart from the fact that this still doesn't provide anything like the type of causal explanation which we require (nor does it really resolve any of the problems I outlined previously), the thesis that sex evolved in order to prevent macroevolution renders even more implausible the grander claims of evolution that all extant species are the product of descent-with-modification as a result of random variation acted upon by selection.

Sex continues to reign as the ultimate conundrum, the queen of problems, in evolutionary biology. Accounting for its origin in terms of evolutionary rationale is inevitably doomed to failure. As David Tyler recently remarked, "Time and time again, Darwinists fill the gaps in knowledge with their theoretical models, but sooner or later, the next generation of scholars will realise that Darwinists have constructed a virtual world that does not match the real world revealed by research."