For a lot of people, I seem to have become the go-to blogger for
information theory stuff. I really don’t deserve it: Jeff Shallit at
Recursivity knows a whole lot more than I do. But I do my best.
Anyway, several people pointed out that over at the Disco Institute,
resident Legal Eagle Casey Luskin has started posting an eight-part
series on how the Kitzmiller case (the legal case concerning the teaching of
intelligent design in Dover PA) was decided wrong. In Kitzmiller, the
intelligent design folks didn’t just lose; they utterly humiliated themselves.
But Casey has taken it on himself to demonstrate why, not only did they
not make themselves look like a bunch of dumb-asses, but they
in fact should have won, had the judge not been horribly biased against them.
Casey’s basic argument is really pathetic. The idea behind it is that the
judge in the case allowed the ruling to be dictated by the National Center for
Science Education (NCSE), which filed an Amicus brief that was quoted
extensively in the decision.
Casey’s key piece of evidence?
In the NCSE brief and in testimony, the fact that mutations can add
information to genome was discussed. In both cases, the key citation was a
review article from a group led by Maunyuan Long of the University of Chicago,
which provides an overview of research that describes how new genes are
As a disco-creationist, Casey claims that random processes like mutation
cannot create information. If a random process can’t create
information, then the information in an organism’s genes must have some
non-random source – like the “intelligent designer”.
Given that, according to Casey and his fellow disco-dancers,
evolution can’t create information, there must be something
wrong with the work that was cited at the trial. And Casey,
true genius that he is, discovered the problem: words!
You see, in Long et al’s paper, they don’t use the word
“information”, much less the phrase “new genetic information”. Therefore, the
paper doesn’t discuss the origin of genetic information. Anyone who says
that it does is clearly lying!
Virtually all of those “publications” mentioned by Judge Jones
came from one single paper Miller discussed at trial, a review article,
co-authored by Manyuan Long of the University of Chicago. The article does
not even contain the word “information,” much less the phrase “new genetic
The catch, of course, is that the paper does discuss the origin
of genetic information. But the authors are primarily biologists, not
mathematicians. Instead of talking about “new genetic information”, they talk
about “new genes”. What’s a new gene? It’s a gene that’s different
from any pre-existing gene. So what’s the content of the new gene? All
together now: “information”!
But simple reasoning like that isn’t exactly Casey’s strong suit. Being a
scientific illiterate, he can’t read the paper to understand what it says. All
he can do is look at the words. But that’s enough for Casey. Clearly,
a paper titled “The Origin of New Genes” doesn’t discuss the origin of new
genetic information – after all, it doesn’t have the words “new genetic
information” in it!
It’s a ridiculous claim. It’s so damned stupid that it’s hard to believe
that even a disco dude like Casey could possibly say it without being
embarrassed. You’d think that after this long, I’d have reached the point
where I couldn’t be surprised by these shmoes any more, but they keep lowering
And it gets worse.
But are Judge Jones’s, Ken Miller’s, and the NCSE’s bold proclamations
supported? Does Long et al. actually reveal the origin of new biological
information? Is Explore Evolution wrong? A closer look shows that the NCSE is
equivocating over the meanings of the words “information” and “new,” and that
the NCSE’s citations are largely bluffs, revealing little about how new
genetic functional information could originate via unguided evolutionary
mechanisms. This bluff was accepted at face value by Judge Jones, who
incorporated it in his highly misguided legal ruling.
This argument is particularly silly coming from Disco. Many of disco’s
arguments go back to the stupidity of Dembski’s “complex specified
information”. In all of the years of arguments about CSI, Dembski has
never provided an unequivocal definition of CSI. While claiming that
it’s a specific, computable quantity, he has never defined it well
enough to allow anyone to compute the CSI in any system; and he has never
demonstrated a complete computation of CSI. In fact, he can’t possibly do
that, because to do it, he would need to stop fudging about what CSI actually
is, and the moment he did that, the whole CSI argument can be refuted. Dembski
constantly relies on the equivocation in the definition of CSI, so that
any time anyone tries to refute it, he can just wave his hands
and say “Oh no, you idiot, that isn’t CSI.”
For their arguments, the disco folks continually discuss CSI. They
constantly cite it as the primary reason that people should take
intelligent design seriously as science. And yet they’ve never managed to
actually compute it, or even to actually define it without hedging.
But it’s the other guys are trying to pull the wool over our eyes by
equivocating over the definition of information.