Language - Single-Gene Epilepsy and More

Ah, I see _Nature_ v. 432, 11 Nov '04 has a couple excellent articles for us. I think if I were an evolutionist, they'd make me very uncomfortable indeed. But first: a couple notes about articles in other journals in the family, as presented on the "view" page (_Nature_ v. 432, 11 Nov '04, p. xv).

"Language: early learning" describes the article "Early language acquisition: cracking the speech code" by P.K. Kuhl in _Nature reviews Neuroscience_ 5, 831; November 2004. It notes that "The acquisition of language is child's play for human infants, but theorists remain perplexed as to how they crack the code of speech." Progress is being made, but "the strategies that infants use...prove different from any previously predicted." Heh, heh, heh. It seems they use several "statistical learning and social skills" that are "unique to humans." While scientists are busy researching under the assumption that humans are evolved apes, they keep running across ways in which we are distinct from all other animals. I'm sure it would be very obvious if all the unique traits were compiled, but as long as most scientists are looking in the other direction, they're never going to see it.

"Single-gene epilepsy" touches on a couple articles in _Nature Genetics_ 36, noting that "Around 40% of casess of epilepsy have a genetic basis, but only a small fraction of these are caused by defects in a single gene..." This seems to be another example of the overwhelming tendency for mutations to be bad or neutral, and also indicates that several "neutral" mutations may eventually add up to produce a serious defect. Another way to look at it: it may indicate our genome is designed to resist the harmful effects of certain single mutations, so that a problem develops only when several mutations accumulate. At any rate, there's no help for the theory that mutations can accumulate to produce greater complexity here. The item, "Cancer genetics: multiple choice" (based on an article in _Nature Reviews Cancer_ 4, 850; Nov. 2004) points out that, likewise, "inheritance of most common cancers is likely to be polygenic."

In the "news in brief" section, "Physicists see fraud as colleagues forge ahead" should be a tonic for those who worship Science and scientists. Keep in mind that physics is perhaps the field of science that includes the best examples of "hard" science, usually loaded with direct observations and repeatable controlled experiments. "Young physicists are quite likely to have come across fraud in the course of their work, according to a survey..." that found "that 39% ... had seen their peers behave in an unethical way..." such as "including an undeserving author on a paper, or omitting a deserving one. But the group also reported incidents of plagiarism, data falsification and 'less than truthful reports'."

The news feature, "A mole in hand..." presents another case of "convergent evolution" or things which share a number of similarities even though they didn't evolve from a common ancestor. The tiny itjaritjari ("no more than 14 centimetres long") is also known as a marsupial mole, and like at least one other burrowing marsupial, has a "backwards-facing pouch." But isn't it a bit of a strain on credulity to propose that the moles we are familiar with, marsupial moles, and the golden moles of Africa all evolved their highly specialized burrowing traits separately?

Showing again the fallibility of scientists, researchers discovered that this critter "produces stunning labyrinths of tunnels. This immediately dispelled one myth about the creatures: that their behavior is like that of the desert golden moles (_Chrysochloridae_) of southern Africa, which they strongly resemble. Desert golden moles are known as sand-swimmers because they produce no tunnels."

Also, in comparison/contrast with some of the claims for various results from ancient DNA of tens of thousands of years: "DNA studies could shed more light on their diversity, but researchers have had to rely on preserved museum specimens, the DNA from which has usually degraded." Good grief, even the oldest specimens couldn't be from any farther back than "the late 1800s," and if it's degraded so badly already, how is it that some researchers claim they have DNA from Neandertals and Cro-magnons that's in good enough condition to tell that the Neandertals didn't contribute to our modern gene pool?

The letters page is entirely devoted to responses to an article (which the authors describe in a reply as a "light-hearted piece") suggesting that women Olympic sprinters may be faster than the men by the middle of next century. It's interesting to see how something as apparently straightforward as an analysis of statistics can generate so much opposition and so many opposing views. (_Nature_ v. 432, 11 Nov '04 p. 147).

One writer points out that extrapolating the analysis (as evolutionary uniformitarians have done with so many things), would suggest "a far more interesting race should occur in about 2636, when times of less than zero seconds will be recorded."

A highschool Advanced Placement Biology Class wrote they believed "the projection ... is riddled with flaws," pointing out "one cannot draw these conclusions based on generalization by extrapolation. Tatem _et al._ used a domain of 104 years to extrapolate to a domain of 252 years. It is not logical..." to extrapolate so far. Don't you wish _Nature_ would print letters from creationary students, pointing out how ridiculously extreme it is to extrapolate into the billions of years from data gathered in a few hundred or fewer? The students here point out that "Using similar reasoning in 1992, Whipp and Ward suggested that women would run the marathon faster than men by 1998."

Another writer mildly points out that "Trend extrapolation can be an inexact science..." and cites a study with a different result: "sports physiologist Stephen Seiler has analysed Olympic and world championship running results and found that the mean performance gender gap in the world records has actually increased" slightly.

I find it interesting that the authors' reply equally speaks volumes about the subjective and uncertain nature of scientific reports that might seem objective and even mathematically absolute if not questioned: "We were, of course, careful to caveat predictions with 'if current trends continue.' But if we were to follow all the advice we have received, we would be both correct and incorrect to fit a linear model. If we were incorrect, we should instead have fitted a two-part spline, a lowess curve, not a lowess curve, a rational function of polynomials, a quadratic model (predicting times regressing back to 1900 levels by 2100), a cubic model and an exponential curve. We should have both removed and added points, and were both correct and incorrect in our use of confidence intervals!"

Doesn't this suggest how easy it must be for scientists to get things like data points for dating methods (including isochrons) to point to a result they can be happy with?

Until Next Time,

David Bump
Philippians 3: 13 Brethren, I
count not myself to have
apprehended: but [this] one thing
[I do], forgetting those things
which are behind, and reaching
forth unto those things which are
before, 14 I press toward the
mark for the prize of the high
calling of God in Christ Jesus.

http://home.att.net/~david.bump