Field of Science

Taking Intelligent Design seriously

One Intelligent Design advocate commented on a previous post, asking me to show him/her how I would argue for Intelligent Design, because, in this person's view, that's the only way to understand something, trying to argue for it, as opposed to trying to disprove it. I wholeheartedly disagree with that view, but I don't think there is any danger in giving a theory (scientific or not) one's best shot, so to speak. So here goes my attempt at arguing for ID, in the most honest way that I am able, with a ending explaining why I am not an adherent (but add to that that ID obviously is not a scientific theory, because science is about invoking natural causes, not supernatural ones).

First, contrary to what many anti-ID debaters have argued, it doesn’t matter what the motivations for the origination and adherence to intelligent design are. Many other scientists have begun fruitful research motivated by a desire to elucidate God’s work, and yet no one disagrees that this doesn’t matter, as long as the results are testable and reproducible. I therefore do not think that ID can be dismissed simply because it is religiously motivated.

Second, in anthropology, criminology, and possibly other areas of science, detecting design is a valid and important undertaking. It is valid to ask the question whether a stone of a particular shape was designed (with the intent of function coming before the form of the object), as opposed to appearance by natural causes (fashioned by geological processes, or in the case of living organisms, evolved). Similarly, despite any preconceptions (and lots of evidence that organisms evolve), we can ask whether there is any evidence for prescient design in living organisms. Did some conscious designer (who need not be specified anymore than a crime lab investigator need suggest who the murderer is, when he concludes that death was not an accident) have a hand (and mind) in the origin of any part of what makes up living organisms?

Without initially considering how to detect intelligent design, I think it beneficial to ponder how such evidence would be dealt with statistically. Imagine that some hitherto unexamined system turns out to show evidence of design. This would be breaking news, of course, and suddenly there would be researchers trying to back up as well as demolish the claim of design. If it is found on second look that there is indeed a way that this system could have evolved by natural causes (a claim that would have to be independently tested for good measure), then the hypothesis that a designer was involved would have to be dropped. The two models (design vs. nature) aren’t equal in the sense that so far many biological systems have been shown to be caused by natural processes, whereas none have been shown to be designed, and the burden of proof is therefore on the hypothesis of design. Evidence that the system evolved therefore rules out the hypothesis of design, and best scientific practice is then to abandon the idea of design in this case. On the other hand, if continued research fails to explain the data by natural causes, then the hypothesis grows in strength, as it becomes more and more unlikely that it will ever be falsified. This is not special to claims of intelligent design. Rather, this is the general rule for all research. As is often stated, nothing in the natural sciences can be proven beyond any doubt, but only, as they say in the courtrooms, beyond any reasonable doubt. This doesn't mean, though, that by disproving evolutionary theory we have proved that a designer designed stuff. However, if evolutionary theory were to fall with all at once (that would take a whole lot of odd results from all over the field of biology, and possibly physics, geology, and astronomy as well), then no other scientific theory would be present to take over, and it would thus lend some credibility to Intelligent Design. However, this setback would emphatically not stop scientists from looking for new scientific explanations for the evidence.

So how about it, then? Do we see design in nature? The short, confounding, answer is yes, we see much design in living organisms. Wings really are designed for flying, the immune system is designed to thwart intruders, and Toxoplasma gondii really is designed to infect mammal eyes, hearts, brains, and livers. Further, the teleological inference is that since they appear designed, then they are designed with those functions in mind. It would appear true that the design is prescient, but that is something that would have to be established. The design that can be inferred so far is not that of preconceived plan, but of exceptional fit to function. Again, in order to make the leap from the latter to the former, it has to be shown more probable that it was consciously designed than designed by nature (evolved).

Then how would someone go about proving teleological design? William Dembski and Michael Behe are the only two people who have proposed any such tests. Dembski proposed specified complexity (basically, it’s too improbable to have evolved so it must be designed), and Behe proposed irreducible complexity (this could not have evolved because taking away any component of the system ruins its function). Specified complexity is widely refuted (as I recall, he even himself admitted something to the effect of it not working, but please look that up yourself), but irreducible complexity far less so (but see this post for a rebuttal). The appeal of irreducible complexity is that it is actually (wait for it…) testable. Data can be gathered, and it can be falsified. This is the signifying strength of scientific hypotheses. You can gather data, and that data can serve as evidence for or against the hypothesis. If I hypothesize that flagella could not have evolved, because taking away any of its components would destroy its function, then that’s directly testable. If trying this out molecularly, and it is found that it never works without all components, and that no other components have homologues elsewhere, and that no components have other functions, and if theory and experiments didn't predict that more than one change can in fact occur at the same time, then we can start taking the hypothesis seriously. That would be a problem for evolutionary theory. But as it turns out, flagella (and all other systems looked at) can be split into smaller components that do serve different purposes, and it is possible (even probable) that two changes could occur at the same time, so that alone shoots big holes in irreducible complexity. The same goes for other systems: blood-clotting cascade, eyes, immune system… Anything looked at so far.

And this is the problem for Intelligent Design at the moment (I’m still trying to argue for it here). Some major setbacks have occurred, but who knows, it might be that some clever(er) person comes around one day to show us that all those things in nature that look so designed by some unnamed very intelligent designer are indeed fashioned with those functions in mind, and not in nature. Until then, Intelligent Design, R.I.P.

Related posts:
10 minutes on Intelligent Design
Why teach Intelligent Design?
The red swan hypothesis
Intelligent Design is a failed scientific theory
The big Judge John E. Jones III interview
Khmer Rouge chemistry

11 comments:

  1. I will respond to this post in a number of separate posts.
    First I must say this is a decent attempt at being fair in regards to the case for ID. I would be pleased if the general response to ID ideas was as fair-minded.

    My first comment is about this:
    "Specified complexity is widely refuted (as I recall, he even himself admitted something to the effect of it not working, but please look that up yourself)..."

    You have mistaken criticism for refutation.
    Here is a link from Dembski entitled:
    "Obsessively Criticized but Scarcely Refuted:
    A Response to Richard Wein
    By William A. Dembski"

    http://www.designinference.com/documents/05.02.resp_to_wein.htm

    The point I am making is that a criticism of an argument is not a refutation.
    There is not a judge (at the Hague or elsewhere) who sits up on the bench declaring a criticism to be a valid refutation or not.

    This leads to the more general point that many people have a strong stake in evolutionary theory (eg. professors, public lecturers, book writers, members of NCSE etc, etc) and they feel compelled to present a criticism whether it has validity or not.
    This is just the way the world is. We just need to recognize that.
    So when it is said that something is "widely refuted", "widely discredited", etc we should immediately realize that there is an agenda behind these characterizations.
    As to your saying:
    "he even himself admitted something to the effect of it not working"
    If you can support this with a link please do so, otherwise it would be best to retract.

    ReplyDelete
  2. To continue ...
    You have said:
    "Without initially considering how to detect intelligent design, I think it beneficial to ponder how such evidence would be dealt with statistically."

    I recommend a careful reading of Michael Behe's book "The Edge of Evolution".
    It presents how such evidence can be dealt with statistically.

    ReplyDelete
  3. To continue...
    You have said:
    "If it is found on second look that there is indeed a way that this system could have evolved by natural causes (a claim that would have to be independently tested for good measure), then the hypothesis that a designer was involved would have to be dropped."

    The problem with this statement is that it relies on the words "could have evolved".
    Anyone can present an argument that something could have happened. The question how likely is it to have happened.
    Someone in my downtown could bump into a telephone pole and that telephone pole could have been really rotted and the pole fell into a store selling lamps and it happened to fall onto a lamp that was lit knocking over the lamp which started a small fire and the shop clerk happened to be in the back of the store checking stock and did not see the fire start and the shop became engulfed in flames and spread through the row of stores and the fire department people happened to be very short of staff that day and were also dealing with a fire just out of town so our fire went blazing through the town and spread to the neighborhood around and started my own house on fire and I happened to be away and my precious stamp collection was burnt.
    IT COULD HAPPEN.
    Right? - it could happen.
    But is it at all likely?
    The Darwinian arguments amount to the same kind of argument. It could happen. Some really smart person (with an agenda) comes up with how if this might have happened and this might have happened and this might have happened and this might have happened - then evolution could explain the result we see.
    But that means nothing. Any child learns how to make up a story.
    For example, ... honest Mom, I did not break the vase. A small earthquake hit our house and it knocked over the vase. It was a really small earthquake. IT COULD HAVE HAPPENED. Mom it could have happened.
    Or, Mom a huge bee came through the window and I tried to kill it before it but I bumped into the table and the vase got knocked over. I tried to save the vase but the bee buzzed around my face and the vase fell on the floor and broke.
    IT COULD HAVE HAPPENED.

    If you look carefully at any Darwinist argument you see this pattern.
    It could have happened.
    But the odds are so astronomically against it that no reasonable person would accept it.
    Here is the joke.
    Darwinists dismiss ID people because they say that it is "personal incredulity".
    But the Darwinists overlook their own credulity to believe any far fetched Darwinian explanation.
    "MOM - IT COULD HAVE HAPPENED".

    ReplyDelete
  4. First I must say this is a decent attempt at being fair in regards to the case for ID. I would be pleased if the general response to ID ideas was as fair-minded.

    Thanks. I take that as meaning that you accept that I understand ID. Again, I disagree that "Nothing can be understood when your goal is to disprove it," as you said, but now that point should be moot.

    You have mistaken criticism for refutation.

    No, the consensus is that it has been refuted:

    Rich Baldwin (2005). Information Theory and Creationism.
    Mark Perakh (2005). Dembski 'displaces Darwinism' mathematically -- or does he?
    Jason Rosenhouse (Fall 2001). How Anti-Evolutionists Abuse Mathematics.
    H. Allen Orr (2002). Book Review: No Free Lunch.

    "he even himself admitted something to the effect of it not working"
    If you can support this with a link please do so, otherwise it would be best to retract.


    I read this... somewhere, but cannot find it now. I will post the link here when I find it. I apologize for stating it without a reference (but stand by the fact that I read something to the effect).

    I recommend a careful reading of Michael Behe's book "The Edge of Evolution".

    Are you saying that what I'm saying about statistics needs to be replaced by what Behe says?

    The problem with this statement is that it relies on the words "could have evolved".
    Anyone can present an argument that something could have happened. The question how likely is it to have happened.


    Not in terms of IC. IC does not deal with probabilities, and rules out that something could have evolved.

    But the odds are so astronomically against it that no reasonable person would accept it.

    Let me see those calculations. (I'm not unaware that such have been made, but you'll have to present them in support of your statement that a proposed way something could evolve is too unlikely. Just please don't refer me to Hugh Ross, okay?)

    ReplyDelete
  5. To continue...
    You have said:
    "If I hypothesize that flagella could not have evolved, because taking away any of its components would destroy its function, then that’s directly testable. If trying this out molecularly, and it is found that it never works without all components, and that no other components have homologues elsewhere, and that no components have other functions, and if theory and experiments didn't predict that more than one change can in fact occur at the same time, then we can start taking the hypothesis seriously."
    This is a misrepresentation of the argument as I expect you must know.
    This misrepresentation has been shown to be a misrepresentation many times.
    Parts could be used for other purposes, not all parts may be necessary. These things do not refute the idea of "irreducible complexity".
    But my point is simpler.
    My point is that, you think you have honestly and fairly presented ID but in fact you have presented a misrepresentation of the idea.
    This is called a "strawman".
    So that is part of the problem. You do not understand the idea of ID but you think you understand it.
    That is one of the reasons I suggested that you can only understand an idea if you try to prove it right. If you had tried to prove it right you would have made the effort to understand what the idea actually is.

    ReplyDelete
  6. If people make criticisms of an argument that does not mean it is refuted. That is why I went to the trouble of explaining it. Which you then ignored.
    You list 3 examples of people who have made criticisms.
    They are criticisms. Nothing more. They express the author's opinions. Nothing more. They are not "refutations".
    I am not going to argue that point.
    The only reason I went into length to explain it was that you and many others make that same error. You think that if some people present their criticisms that this means that something has been "refuted". It does not mean that.
    You need to learn the difference.
    Especially if you are going to be a professor.

    ReplyDelete
  7. I said:
    The problem with this statement is that it relies on the words "could have evolved".
    Anyone can present an argument that something could have happened. The question how likely is it to have happened.

    You replied:
    Not in terms of IC. IC does not deal with probabilities, and rules out that something could have evolved.

    I am starting to see that either you do not understand English very well or that you have trouble following an argument or that you are just trying to avoid dealing with the real issues.
    My point is about how far fetched the evolution arguments are. But Darwinists still say "it could have happened".
    But you ignore everything I said and take a swipe at ID.
    This looks like it is a waste of my time.

    ReplyDelete
  8. You list 3 examples of people who have made criticisms.
    They are criticisms. Nothing more. They express the author's opinions. Nothing more. They are not "refutations".
    I am not going to argue that point.


    Of course, the refutation lies in the argument. I presented four (4) such, and among scientists consensus is that Dembski is wrong on this point. Don't argue if you don't wanna.

    I am starting to see that either you do not understand English very well or that you have trouble following an argument or that you are just trying to avoid dealing with the real issues.

    Let me know what you decide.

    But you ignore everything I said and take a swipe at ID.

    I certainly have not ignored everything you have said. You, on the contrary, ignore my request for references with calculations that "odds are so astronomically against [evolution] that no reasonable person would accept it."

    This misrepresentation has been shown to be a misrepresentation many times.

    Also, please yourself supply a reference that shows how I present IC is a misrepresentation.

    Parts could be used for other purposes, not all parts may be necessary. These things do not refute the idea of "irreducible complexity".

    Co-option/exaptation has been shown to enable systems to evolve what is by definition irreducibly complex systems. The bottom line is that flagellum, blood-clotting, immune system, etc. have been shown to be evolvable. There is no case left for IC.

    There is no case left for ID. Please explain how it could work if you should choose to comment here again.

    ReplyDelete
  9. Dembski admitted that the explanatory filter doesn't work, not CSI:

    I’ve pretty much dispensed with the EF. It suggests that chance, necessity, and design are mutually exclusive. They are not. Straight CSI is clearer as a criterion for design detection.

    Via The Austringer, Wesley Elsberry's blog. Elsberry referred me to this article where he argues that SC or CSI is incoherent, and not really in need of refutation, as such.

    ReplyDelete
  10. This is hilarious! Michael (a.k.a. "Anonymous"): You know perfectly well what a P-value is. Nobody writing in a scientific journal can get away with writing "it could have happened". You have to calculate, using the data available to you, what is the probability that "it could have happened by chance alone". That's the P-value. Science is brutal that way. They want numbers.

    ReplyDelete
  11. Michael? Chris, please enlighten.

    ReplyDelete

Markup Key:
- <b>bold</b> = bold
- <i>italic</i> = italic
- <a href="http://www.fieldofscience.com/">FoS</a> = FoS