Re: My rebuttal of the desin inference.

From: DNAunion@aol.com
Date: Mon Oct 23 2000 - 18:33:51 EDT

  • Next message: Ivar Ylvisaker: "Re: Turing"

    >>>Richard Wein: ...
    http://website.lineone.net/~rwein/skeptic/whatswrong.htm >>

    DNAunion: I spotted several gramatical errors or typos that need to be
    corrected.

    (1) Search for "mostly widely". It should be "most widely" (or is it
    "most-widely").

    (2) At the beginning of section 2.4 where you quote someone, you left off the
    ending quotes on, "... chance explanations.["]

    (3) At the end of section 2.5, you say that Dembski has to show that he is
    using the terms the same way as others "... before practicing such
    conflation". But if he shows he is using the terms the same, then he is not
    practicing conflation. This needs to be reworded.

    (4) In the very last sentence, you wrote "... to provided..." instead of
    "...to provide..."

    I would also like to point out something I found interesting, personally.

    Richard Wein: "However, given the uncertainties about what an application of
    the Design Inference entails, it is impossible to accept such claims coming
    from any source other than Dembski himself."

    That is one reason I take a "non-position" on the technical details of
    Dembski's EF.

    In fact, I have entered into debates with Richard Wein at ARN thinking I had
    a reasonable grasp on Dembski's material only to be shown otherwise. For
    example, I tried applying the law of small probability to the minimal genome,
    and found design (specified small probability). Richard pointed out problems
    with my argument (can;t remember exactly what they were off the top of my
    head) and I had to agree with him. I *personally* have found it impossible
    to apply Dembski's EF to biology (I personally think the only reasonable way
    is to go back to the ORIGIN of very first biological molecules, but that
    itself is problematic).

    In addition, yes, I did have a discussion about Dembski's EF with Elseberry
    before elsewhere, and he too showed that I was ill-equipped to defend the
    idea (I informed him towards the end that I had only received the book a
    couple weeks before getting into our discussion, and had only read the first
    5 or 6 chapters).

    So why won't I argue the details of Dembski's EF here? Because Richard and
    Wesley have shown that I am not qualified to defend it at that level. I
    don't focus on the EF; I don't know it inside out; and I agree with Richard
    that Dembski does cause confusion at points. For example, he equates
    Specified Complexity with CSI at some points (specified complexity is used by
    others, and myself, differently), he calls the information in a phone number
    CSI, and says that CSI must possess at least 500 bits of information (which a
    phone number does not). These misuses of terms are unfortunate, and could
    be, it appears, easily corrected by Dembski, improving the readability of his
    work.

    I too would like to see Dembski marshall all the *valid* objections to this
    proposal (he could ignore the petty ones) and write a new book addressing
    them fully. Dembski seems to have been "mokeying around" with the EF over
    the last several years, reworking it (I assume to make it more robust in
    response to objections). I hope we see him continue this reworking until he
    arrives at one that addresses all the current problems. And it WOULD be
    great to see if he could apply it to biology, for example, in relation to the
    bacterial flagellum (as RW quoting Dembski as already considering it to come
    out of his EF as designed).

    Finally, note that I do not speak for ID, but for myself, on this. I
    *personally* have problems applying Dembski's EF, but that could be because I
    don't fully understand it, which I don't. Other IDists may have a deeper
    understanding and the ability to use it correctly - I am not saying that
    Dembski's EF is necessarily invalid.



    This archive was generated by hypermail 2b29 : Mon Oct 23 2000 - 18:34:03 EDT