lpetrich

It is true that this Dembski flub does not have the grotesque proportions of Gish's or Wells's flubs, because it looks like an offhand remark, but I do think it qualifies as a flub. To me, he was claiming that beavers intelligently design their dams -- even though there is evidence that beavers have a dam-building instinct that involves placing sticks and mud on whatever sounds like rushing water, even an underwater speaker playing the sound of such water. Nic Tamzek's observation of beavers building damless lodges in big rivers is consistent with that hypothesis -- no rushing water, no dam.

Beavers are a good example of a serious problem for the intelligent-design hypothesis -- and one that many ID critics have failed to appreciate. In fact, I seem to be the first ID critic that has realized this(!) If there are species other than Homo sapiens that can practice intelligent design, then there are additional intelligent designers right under our noses! And sometimes in a very literal sense!

Animal intelligent design is a common item of folklore, and many pet owners seem to believe that their pets are capable of intelligent design. And some animal structures do seem to be the result of intelligent design, like spiderwebs, honeycombs, and beaver dams. In the late 19th cy., the biologist George Romanes collected numerous seeming examples of human-scale animal intelligence, though his work was later widely ridiculed as a splendid example of how not to do animal-behavior research.

Such intelligent-design capability seldom reveals itself in controlled contexts, and there are plausible instinct-based hypothesis on how these animals can construct these structures. For example, spiderwebs have a stereotyped and species-specific overall architecture -- and one that each individual spider has no chance to learn how to create. And some months earlier, there was a thread here that featured an article on simulated spiderweb construction that used an algorithm that contained several subalgorithms with the form "if you feel this configuration of nearby threads, move here". Such complexities indicate that complexity alone is no argument against the instinct hypothesis.

And though much animal behavior is various combinations of instinct and learning, most forms of learning do not qualify as anything close to intelligent design, either. Here's a list:

Imprinting -- seeing something at one point and recognizing that sight as a part of later behavior. Thus, Konrad Lorenz convinced some geese that he was their mother by being near them when they hatched; they would follow him around despite his very ungooselike appearance. This experiment has been repeated many times since, sometimes with objects like white balls.

Habituation -- stopping response to stimuli that are associated with nothing significant. A bird may no longer be startled by nearby traffic that it hears continuously.

Latent learning -- exploring and picking up details without making an effort to act on what one has learned; such learning may get used later.

Classical or Pavlovian conditioning -- transfer of some response to an additional stimulus in addition to the original stimulus that produces that response. Thus, Pavlov's dogs learned to salivate at the sound of a bell -- a sound that the dogs heard when they were given food.

Operant or instrumental conditioning -- learning to do something to get some desired result, or to avoid some undesired result.

There is, however, one additional form of learning which suggests intelligent-design capability:

Insight learning -- pausing and then implementing a solution. This was first observed by Wolfgang Koehler in chimpanzees back in WWI; there are a few other species that show that capability, like ravens. But only a few species, and it would be very surprising if beavers could perform insight learning. Which is consistent with their building dams by instinct.

So I stand by my example of Dembski and the beaver.

Valentine Pontifex

Well copying the column on the chart in TBA and pasting to Google and hitting the button finds <a href="http://www.geocities.com/Heartland/Plains/7075/resource/creation.html#19" target="_blank">this</a>. It was debunked <a href="http://freespace.virgin.net/karl_and.gnome/donthavetoignore.htm" target="_blank">here</a>.

lpetrich

Although it is certainly true that butterbeans do not have blood, this might be a reference to some butterbean protein that can provoke an immune response in some animal subject. Imagine injecting a butterbean extract into a rabbit and then checking on what proteins that that rabbit has become sensitive to.

And this reference to butterbeans is an example of another flaw in creationist literature: reliance on "ordinary" names as opposed to Linnean ones. This is OK for familiar species, but for others, the "ordinary" names are often very imprecise, and many species are mainly known by their Linnean names. Does anyone ever call Escherichia coli anything but that?

As to butterbeans themselves, they are a variety of soybeans (Glycine max).

RufusAtticus

From one of the above links it says that Butterbeans do have a type of sugar which in high concentrations can cause antibodies to blood antigen-A to react. I suspect that antigen-A is a "sugared" protein, which is why a sugar can cause such an immune system reaction.

pz

Close! The A antigen is a glycolipid. It has a linking fucose molecule bound to a galactose, linked to an N-acetylyglucosamine. The A antigen has an N-acetylgalactosamine coupled to the galactose, the B antigen has another galactose added to it, while the O antigen has no terminal sugar added. It is not surprising that purely by chance, the arrangement of such a small number of sugars might correspond to another collection of sugars in a plant glycoprotein.

RufusAtticus

Thanx, pz.

theyeti

The real problem for Dembski and ID "theory" is how to define intelligence. Just what is it? Normally it wouldn't matter for most biological theories, but ID is predicated on being able to recognize the results of "intelligence" where and whenever it may manifest itself. You'd think they'd start out with a highly rigorous and unambiguous definition of intelligence, but AFAIK they haven't bothered. In fact, the ambiguous nature of the word "intelligence" makes for a good rhetorical device -- it's a moveable goal post. By using this term IDists can avoid referrences to the supernatural and other unpalatable concepts. For example, IDists talk about things in terms of "natural vs. intelligent" causation, when what they really mean is either "natural vs. artificial", or "natural vs. supernatural" (note that these last two are not the same thing at all, but by adopting the term "intelligent", IDists freely conflate the two when convienient.)

The only defintion of intelligence that IDists give as far as I know is "something capable of making choices". But of course that just begs the question of what it means to make a choice. Does a beaver choose to make a dam? Could it choose not to? The whole thing quickly degenerates into some sort of free-will vs. determinism debate. And while I don't keep up with those debates, the very fact that there is a debate just goes to show what shakey ground ID is on.

In particular, I would like to know where one finds "intelligence" and where one doesn't. Which of the following would be intelligent, and more importantly, why?

• human being
• chimp
• beaver
• snake
• trout
• octopus
• temite
• round worm
• paramecium
• E. coli

The facts of nature are such that there doesn't seem to be any clear cut dividing line between intelligent organisms and non-intelligent. Rather, intelligence appears to be an emergent property that's manifest from the nervous system. So while I wouldn't call E. coli intelligent, it's a much harder call with the round worm. But of course this is Dembski's problem.

I don't think it's fair to call him on a blunder the magnitude of the Bullfrog or mutant shrimp afairs. Dembski's real flub is in adopting an undefined term to hinge his entire argument upon.

theyeti

lpetrich

I think that a good definition of the intelligent-design ability would be the ability to create a mental model and then manipulate and use it.

Thus, if I wanted to build a dam, I would not compulsively collect sticks and mud and deposit it wherever I heard rushing water, as beavers do. I'd construct a schematic model of that dam in my mind and ask where it ought to be, how high to make it, what to make it from, etc. And when I found one that I consider satisfactory, I would proceed with construction.

The next question is how to look for evidence of such a process in other species, since only ours has full-scale language that is intelligible to us. A hint is to look for evidence of insight learning, which is pausing and then implementing a solution. This form of learning suggests the use of such a mental model; a chimp would pause to manipulate a mental model of stacked crates before going out and stacking real ones to reach an out-of-reach banana.

This ability is very rare, and outside of our species, is best-developed, or at least best-studied, in the chimp. Which is anatomically and genetically closest to our species -- another triumph for evolutionary biology.

And checking theyeti's list, our species and the chimp are thus clearly capable of intelligent design, but the others are not.

This suggests a possible way of recognizing intelligently-designed evolution; would it be possible to tell if some features had originally been mentally modeled before they were built?

lpetrich

A search for info on beaver dam building has revealed discussions of a <a href="http://dnr.metrokc.gov/wlr/Dss/beavers/beaver-deceiver.htm" target="_blank">"Beaver Deceiver"</a>, a trapezoidal fence placed at the entrace or exit of a culvert to keep beavers out. I've also found <a href="http://www.usroads.com/journals/rmej/9804/rm980401.htm" target="_blank">a more systematic discussion of beaver-control methods</a>, one that illustrates how to drain beaver dams without provoking the building of new ones.

And according to <a href="http://lubbockonline.com/news/090197/deceptio.htm" target="_blank">this article on the Beaver Deceiver</a>, its inventor, Skip Lisle, describes why its trapezoidal shape is so effective:
Which clearly indicates that beaver dam building is essentially instinctual; it is provoked by hearing and feeling rushing water.

Clutch

This is interesting stuff, because it highlights that the heart of Dembski's claims about intelligent design is an underlying view -- not even pseudo-argued for via the (giggle) Explanatory Filter -- about the nature of intelligence itself.

Assuming (per impossibile, it seems) that some natural system can be shown to be designed, there is simply the further question of what explains the intelligence of the designer: natural selection? another designer? The theistic payload of ID-ology is really carried by a completely a priori conviction that mind cannot just be naturally constituted or explained.

I saw a paper on BillyD's site a while back, the text of a talk, I think, where this was argued in roughly the same dismal manner that was once the trademark of Undercover Elephant here -- I can't think of how the mental could be physical; ergo, it's logically impossible.