Monday, January 09, 2006

ID untestable? Nonsense!

Jay Richards and Jonathan Witt have put up a piece here about whether ID claims are empirically testable or not. I am interested to know what those who have claimed in comments on this blog -that ID is not science - think of this piece.

(Also, I am interested in what they think of Behe and Snokes paper in Protein Science - "Simulating Evolution by Gene Duplication of Protein Features That Require Multiple Amino Acid Residues" )

10 comments:

Lifewish said...

OK, I'm trying to debunk the article, which I think is wrong on several points, but it's taking a while because each article is apparently based on half a dozen more, each of which I also have serious problems with. I'll get back to you on this when I find the end of this particular roll of sticky tape, but the main point I intend to make is that failure to discover the "continuously functional Darwinian pathway" they describe doesn't mean it doesn't exist, just that we haven't found it yet.

I also hope to make the subsidiary point that, as Ken Miller (correctly, as far as I can tell) points out, we're partway there already.

I'd also note the extreme broadness of the get-out-of-jail-free card that is the phrase "moreover, any detailed evolutionary pathway that is articulated will presuppose other irreducibly complex systems". Since they are the ones that describe whether something is irreducibly complex, and since any biological system rests on a hundred others, they could keep me busy for years hunting for and arguing over the thousands of explanations I'd need for myriad vaguely-related components that they could claim the irreducible complexity came from. This letout clause makes their test effectively untestable.

As an amusing side-note, I'd like to point out that Behe's favourite irreducible complexity example, the mousetrap, would be completely evolvable if mousetraps could breed (sadly for us they can't, or they might get even more efficient).

Anonymous said...

"Trying to debunk" arguments on intelligent design, hmmm, I always heard from evolutionists that ID was unfalsifiable?

Anonymous said...

Let's make this simple.

What is the falsifiable hypothesis for ID?

Lifewish said...

"Trying to debunk" arguments on intelligent design, hmmm, I always heard from evolutionists that ID was unfalsifiable?

The underlying hypothesis of Intelligent Design - that somewhere, somehow, there's something that can't be achieved through evolution - is scientifically unfalsifiable. However, the calculations they do to "prove" this conclusion are mathematically falsifiable, the facts they use to "prove" this are often scientifically falsifiable, and the logical structure they use to reach their conclusion may be logically falsifiable. Any or all of these processes of falsification may be referred to as "debunking". I was referring to a combination of the latter two.

I hope this clears things up.

Anonymous said...

Jeff said: "IDists refusing to say anything (not even the vaguest guess) about the designer (methods, properties etc.)"

I'm not sure what point you were attempting to make here. It is of no relevance that ID theorists cannot identify a specific designer, it is not necessary to do so in order to provide arguments for design.

Jeff said: "otherwise they're all too eager to say that they think that ID proves that the designer is God."

Again what relevance to testability is this statement? So what if religious groups use ID as support. This is not important. What is important is whether arguments for ID are testable, and I believe that argument's like Behe's are. If you read the article in this post they make this very clear.

You call ID arguments: "vague and unrefined"
I just read Behe's book, and his idea's are anything but vague, in fact they are very specific, articulate arguments. If you want to talk about vague and unrefined, click on lifewish's link above and follow that little mousetrap fairy tale. Sure its an interesting idea, but can you give me any real pathways for real, biological, irreducibly complex systems?

I say again, it is not necessary to identify a designer, or even prove anything about a designer in order to provide arguments for design.

Lifewish said: "The underlying hypothesis of Intelligent Design - that somewhere, somehow, there's something that can't be achieved through evolution - is scientifically unfalsifiable."

You are inferring that ID is a strictly negative argument, which I do not believe that it is.
From the article: "Behe argues that this tiny flagellar motor needs all of its parts to function—is “irreducibly complex.” Such systems in our experience are a hallmark of designed systems, because they require the foresight that is the exclusive jurisdiction of intelligent agents."
This is not a negative argument.

Lifewish said...

Sure its an interesting idea, but can you give me any real pathways for real, biological, irreducibly complex systems?

Here is a brief example of the state of the evolutionary art. For more information on the same component, I recommend either looking up Ken Miller's recent talk in Ohio (available for download) or getting an evolutionary biology textbook.

In particular, note how more detailed and testable this hypothesis of development is than "at some unspecified point in the past, someone did something. We don't know who and we don't know what, but it made stuff more complex".

Lifewish said...

You might find this analysis of the evolution of the eye informative too. Note the fact that eye evolution was actually reproduced using evolution of computer models.

Anonymous said...

Thanks for the article Lifewish, I read it, and overall impression of this "evolutionary art" is just that, art. Its a good story, and I am not really all that impressed. I have seen this kind of answer time and time again. If you read the article carefully, you note statements such as "converts to", "eventually forms", "cell fortuitously becomes associated with...". There is no explanation for why this happens, it just happens, and I call THAT vague.

Have you read Behe's book? If not I would highly recommend it. His arguments for design are really articulate and convincing.

Regarding the computer simulation of the eye, I have heard of these programs before, and I don't find them all that convincing, anyone can make a computer program do anything.
I want to see biological evidence.

Lifewish said...

Its a good story, and I am not really all that impressed.

It's more than just a good story, it's a falsifiable story. If someone discovers a reason why any one of the steps couldn't have worked, or discovers that one of the components actually bears no real resemblance to the thing it's supposed to have evolved from, that falsifies the story. Subject to peer-review, that would be game over for this particular story, and we'd have to come up with a more accurate one.

Have you read Behe's book? If not I would highly recommend it. His arguments for design are really articulate and convincing.

As a poor student there's a limit to how many random books I can buy without cutting into my beer budget :P There's a copyright library at my university that'll have a copy, I'll check it out there to see if it's worth it.

Regarding the computer simulation of the eye, I have heard of these programs before, and I don't find them all that convincing, anyone can make a computer program do anything.
I want to see biological evidence.


The point of developmental simulations and genetic algorithms and so on is that the computer program does things it hasn't been made to do. And we know it hasn't been made to do it because no human being could possibly come up with some of the solutions that genetic algorithms produce. For example, in the Tierra "digital environment", the guy who wrote it originally seeded it with 80-unit long "lifeforms" - the shortest length he could figure out how to produce. The system eventually "evolved" a 22-unit lifeform. That's almost four times as efficient as a professional programmer.

Regards biological evidence: which bit do you want biological evidence of?

Anonymous said...

The ID is/isn't falsifiable claim has been going around for a while now. ID proponents seem to get thoroughly confused when attacked on both fronts.

The root ID claim, that an unspecified "intelligence" has somehow, at some unknown time, and for some unknown purpose, intervened in the workings of the universe, is essentially unfalsifiable. The designer may choose to do his designing in a way that mimics evolution entirely. His/Her/Its reasons for doing this may be inscrutable to us humans. Science is not a sharp enough tool to dissect situations like this.

Leaving that root proposal aside, some other aspects of ID are possibly falsifiable, particularly when they make claims against evolution (which, lets face it, is most of ID's intellectial "content"). For example, IC has long been falsified from its original incarnation, Behe having backpedalled from the definition he presented in DBB. However, now all that has happened is the definition of ID has been refined in terms that are harder to falsify. Now us evos have to present a logical progression for some biochemical structure and somehow demonstrate that it's probable. Nice! Of course, ID proponents refuse to state what the probability of their designer doing it is ;) I imagine that eventually someone will come up with credible paths for quite a few biological structures, what's the betting that ID will simply move onto other ones?