> I think all of this misses the point. The point is that CSI *is*
> positive evidence of mind's intervention.
Tim replied:
>Dembski is pretty specific where we wants to go with CSI. He is
>trying to develop a means of demonstrating that CSI cannot arise
>via natural mechanisms.
This is not how I would use it, thus I'll skip ahead.
Me:
> It is a very common way that mind leaves it traces behind (as shown,
> for example, by the archives of this listing). That some other dynamic
> "might" generate CSI does not remove CSI from the category of evidence
> of mind's intervention. After all, below you cite similarities as
> positive evidence of evolution, yet a common designer "might" likewise
> have generated these similarities.
Tim:
>Explain to me exactly what pattern a common (or uncommon) designer
>couldn't produce. Then compare that to the pattern that an IPU (Invisible
Pink
>Unicorn) might generate. I'd like to see how to differentiate models.
>Divine design explanations have the potential advantage in that they are
>not limited by need to explain mechanisms, but they are also hindered by
>their incredible flexibility, which makes differentiation difficult.
>Hence the need for auxilliary or amending hypotheses to make a
>nebulous idea more concrete and applicable (more below...)
This is getting away from my point. I was simply noting that the
similarities you attribute to evolution "might" be due to design. But this
doesn't keep you from attributing such similarities to evolution.
Well, in the same way, that CSI "might" be due to non-intelligent
causation doesn't keep me from attributing CSI to design.
Now, as for your point. I agree that invoking divine design entails
all sorts of methodological headaches as far as flexibility is
concerned. But I suspect that the problems only seem larger than
they are and if one were to analyze a particular aspect of biological
reality, he who invokes design and he who denies it would find
specific interpretations to differ about. It's kind of like emptying
that box full of pieces and parts to put together an entertainment
center. It looks intimidating, but you just take it one step at a time
and proceed gradually. The rest takes care of itself.
Furthermore, don't forget I am a relativist on these matters.
The approaches that exclude intelligent design seem likewise
to suffer from flexibility problems. Natural selection has
a history of being too "just-so"-like (we've just seen on this
listing how natural selection seemed to be such a good
explanation in accounting for no hollow bones among land animals,
yet the reasoning conflicted with the evidence). Similarities
need not be explained in terms of divergent evolution. One
can always appeal to convergent evolution, horizontal
transfer, and simple coincidence. And an approach which
excludes design can (and does) always appeal to chance at
some point or rely on "we don't know (yet)."
Me:
> Positive evidence for X does not mean that X is the one and only
> way to explain the data deemed positive evidence.
Tim:
>Of course not. You are certainly free to explore the possibility
>of other explanations such as special intervention. And when you're
>done, report back on how you were able justify the auxilliary hypotheses
>about a designer which you must invoke to get a testable or at least,
>a positively descriptive model of an intelligent designer (see the
>Sober reference, please).
As I explained in my last message, one can simply employ
the same basic explanatory strategy you employed to argue
myoglobin was the product of mutation, drift, and natural
selection.
>In my last letter, I suggested some formulations or auxilliary
>assumptions about a designer that might positively identify its
>interaction. I didn't provide those particular models of possible
>designer methods to defend them; they were included to demonstrate
>that positive models *could be* formulated.
But those were more than "positive models." Those were
essentially *proofs* of design. Like I said before, I
don't see why a design inference must be a certain proof.
One can make positive models just as I explained in
my previous message.
This need for proof is very interesting. Many deny
outright that design is behind any biological feature, thus
what is needed (I suppose) is some form of certain proof of design.
Isn't there plenty of room between a complete denial of X
and certain proof of X? If one is going to eventually adopt
a belief about X, doesn't it *begin* with suspicions, then
evolves into a growing conviction in light of a pattern of
evidence? I've never seen anyone jump from complete
denial to certain proof.
>I agree that those examples could seem ridiculous or unusual (or not
>palatable to Christian beliefs), but nonetheless they do provide positive
>models and produce distinguishable differences from natural explanations.
>But again, I'm not about to defend such models; I'm asking
>"design theorists" to present theirs.
So in what way is the standard evolutionary account a positive model
that produces distinguishable differences from an explanation that
employs intelligent design? Why do you want the ID crowd to
come up with something that cannot possibly be explained without
ID when science does not find things that cannot possibly be
explained without evolution? Why can't a design theorist do
exactly what the non-design crowd does?
I'm starting to get the feeling that a double standard is
in play here. You get to infer evolution behind myoglobin
because you find evidence you would expect if evolution
is true, you have a mechanism to extrapolate, and you
think the case for design is no good. You don't have
to come up with data that only evolution can explain.
Yet the only thing you will consider as a positive model
for design is the identification of things that only design
can explain. The way I see it, to be balanced, design theorists
should be allowed to adopt your "less-than-certain" approach
or you should start by finding things only evolution can
explain (to rule out the possibility that design "might"
be behind it).
>It's not sufficient for a good model to merely match the output of
>another, or to rely on negative evidence; it must hope to provide
>a differentiatible example or one that positively explains why one
>possible result is favored over another that a competing theory
>cannot provide.
So what has science shown that a design explanation cannot
provide?
>Sure, a designer can make something very complex
>biologically. However, there is a competing explanation which suggests
>that there can be "non-intelligent" routes to biological complexity.
So? I think we are not communicating because you are still
looking for proofs of design (things that cannot possibly be
explained by non-intelligent routes). I just think that's
not a very helpful way of approaching this topic.
>So how do I choose between possible explanations?
But you apparently have. You do think that myoglobin
is the product of evolution, do you not? What was
the basis for your choice?
In the end, I suspect it all boils down to a judgment call, as
this may be one place in epistemology where our inability to arrive
at certainty is not helped much by science. Thus, at the
most we might have two parallel explanations and our
perception of their relative merits might be a function of
other background beliefs.
*** *** *** *** ***
>Now in regard to globin's origin; I don't think this will be a
>productive subject to investigate in this area. The reason is that
>globin's origin is ancient and apparently shared between all the
>kingdoms.
I don't agree. Yes, the methodological problems pose a headache.
But by postulating design near life's origin, one need take only
a modest step away from Van Till's attractive approach. Recall
that I am curious about a hybrid of Van Till's and Dembski's
approach.
>Thus it is likely that the last common ancestor of all
>life already contained it and looking back to see what was there
>*before* this common ancestor will be very difficult if not
>impossible to do. It is better, I think, to look instead at more
>recently acquired components. That's because we're more likely to
>detect the presence or absence of evolvable intermediates; assuming
>related organisms which lack the feature can be identified. So if
>one is going to test whether a gift suddenly appeared in a lineage,
>it would be better to choose a more recently acquired "gift" in a
>system where noise has not had a chance to degrade the signal.
Okay, then let's consider important molecules apparently
not present in the last common ancestor of all life, namely,
ubiquitin and actin. These proteins play central house-keeping
roles among all eukaryotes and demonstrate extreme sequence
conservation over very long periods of time. Now you have
the entire bacterial and archaebacterial domains to survey
in order to find "evolvable intermediates."
Mike