Author: theyeti
Date: Nov. 26 2002,17:20

Hello all.  What follows is a brief essay (or rant if you prefer) about the claim that ID is an appeal to the best explanation, specifically within the framework of ID's supposed explanatory power.  This is posed as an answer to the question, "what's the logical fallacy here?"  


It's not so much a logical fallacy.  It's just that they're plain wrong as far as explanatory power is concerned.  To actually explain something, you not only need to give an account of why something happened, but why it should have happened and why it didn't happen differently.  In other words, your theory should predict the observed outcome, or at least predict a set of possible outcomes (the smaller the better) that overlap what's observed.  It's pretty safe to say that evolutionary biology predicts a much smaller set of possible outcomes than does ID, which itself predicts an infinite set of outcomes (which is to say that it predicts nothing at all).  So the plain fact of the matter is that ID is not only fails to be the best explanation, but it fails to be any kind of explanation at all.  Evolutionary biology on the other hand, while not only giving us a general explanation for the diversity and unity of living things, also gives us a research paradigm for explaining the exact genesis of specific structures and functions within living things.  Note that this is something that ID doesn't even aspire to.

The hypocritical IDist retort to this is to claim that evolutionary biology predicts an infinite set of outcomes, and is thus not testable itself.  Of course if this were really true, then the IC and SC arguments against evolution would not be logically tenable.  These arguments fail because they don't match up to the facts (and in the case of SC, because it's an excercise in question begging).  But they really can't have it both ways.  They can't claim that IC or SC falsify evolution while simultaneously claiming that evolution isn't falsifiable.

The mere fact that they try to make these arguments shows that evolutionary biology is constrained by what it can predict, and therefore can explain not just why things are as we see them, but why they're not somehow wildy different.  Consider for example if every species on Earth were morphologically and biochemically distinct.  Darwinian evolution could certainly not explain that.  Or what if there were only one species and there had always been only one species?  Again, Darwinian evolution would be untenable.  I often think that people are so used to the facts of nature as they are that they don't stop to think about situations that might make non-evolutionary accounts far superior.  (I also think this is why biologists, who are more aware of the facts of nature than your average joe, have a particularly hard time not accepting evolution.)  But in all of these cases, whether whether nature is like it is now or whether it's completely different, ID accounts for it either way by invoking the same uninformative explanation: The Designer wished it such.

Now it's not impossible to get explanatory power out of this.  But in order to do so you have to ask why the designer wished it such.  This is not an unreasonable demand, as IDists often claim it is.  When we see some sort of putative designed human artifact, the first thing we want to know is what it's for.  And implicit in this demand is knowing how and why a human would have built such a thing, because of course we don't expect humans to be able or willing to build just anything.  In other words, the set of possible outcomes is limited when we assume human construction.  If we can say what object X is for and why humans living in location Y would have built such a thing, then we've gone a long way towards explaining the genesis of object X.  But what's really senseless is that the IDists not only don't want to engage in this sort of discourse, they even claim that it's unscientific!

Naturally, they have their reasons.  Trying to bring the designer and its attributes into the discussion would force a few things.  1) They'd have to admit that it's all speculation.  There's nothing wrong with that per se, because every scientific hypothesis starts out as being speculative, but it would raise the issue of the acutal testing of ID hypotheses, and the lack of data on the Designer would make this difficult.  Furthermore, real world data can eliminate several popular Designer hypotheses if we insist on taking the scientific approach to ID.  2) The Big Tent philosophy, who's purpose is to allow any and every ID hypothesis (except maybe the Raelins) equal access, thus swelling the ranks.  This is just a political strategy.  3) They'd have to start comparing ID to Darwinian evolution.  As it is now, it's advantageous to be completely devoid of any theoretical basis, because it lets them sit back and take pot shots at Darwinian evolution without having to account for the so-called mysteries that they invoke with a model of their own.  And surely there are other reasons, but it's not necessary to figure them all out.  The important point is that they've chosen to uncouple the ID argument from the only thing that would actually give it some explanatory power, which is some actual considerations about what it is that's doing the design, how it's being done, why it's being done, when it was being done, etc., etc., etc...

The uncoupling of the ID argument with any considerations of the desiger is exactly what makes ID devoid of any exaplanatory power.  In the absence of a real design hypothesis, ID cannot ever be a scientific theory.  Instead it's just an argument, and the only point to the argument is to prove a designer, not actually to explain anything.  And it's not even a very good argument.  AFAICT, the ID argument has in remained essentially unchanged since Paley.  It was logically unsound to begin with, as shown by Hume, and then Darwin came along and demonstrated why it's unreliable.  Can anyone tell me what new and interesting explanations for the properties of living things that Paley or his followers were able to come up with?

theyeti

Author: RBH
Date: Dec. 02 2002,21:29

Last month I gave a colloquium on ID for high school biology teachers, and as a followup, I've been sending them resources.  In aid of that, I put together a short essay, something like theyeti's.  I plan to distribute it to them in the not-to-far distant future.  Comments/critiques welcome.

***************
In his post on ISCID John Wilkins wrote
Quote
Let us begin by asking how science tells us anything at all. In my view, science is the recognition of patterns in data, and the generation of models that are adequate to delivering those patterns as explanation. The information in science, the "signal" from the physical world, is the information of measurement - Fisher Information, AKA the Cramer-Rao Bound (which is, roughly, where the second derivative of the estimate of the accuracy of a measurement is zero). In my view, science is induction from data, and the models retain the information content of the measurements just to the extent they are accurate. (Note: induction may not be a justification of models, but it sure as #### is the way we gather our data together so we can make reliable inferences; still, let's not open that can of undergraduate Humean worms.) The information content of a scientific explanation is just the preserved accuracy of the data in the model.

Anything that we know through science we know from empirical data. So a design inference has to be not only consonant with data, but licensed by the patterns that exist in the data. To be achievable, we need to understand (that is, have a model of) design and designers.  (emphasis added)


And consider this from my OP in the Multiple Designers Theory thread on ISCID
Quote
D. There is a finite and limited number of multiple designers.  This premise is more difficult to support by empirical evidence than the others, but it is logically necessary to prevent the MDT enterprise from degenerating into a mere list of designed phenomena, a cosmic oddity shop of designs. Scientific theories condense (superficially) disparate phenomena into similarity classes and explain the behavior of instances of the classes by invoking general principles and laws that refer to those classes rather than to individual instances. If the number of designers is unlimited then in the limit each class would have just one member, and (since in that case no multi-member classes exist) no general laws are possible and therefore there is no science. It is logically possible that there is an infinite number of designers, but in that case no scientific study of design is possible. It is therefore a scientifically sterile speculation. (emphasis added)


As I read them, IDists argue that the Explanatory Filter (nowadays pretty much reduced to the assertion that IC structures or processes cannot be produced by evolution) detects a property (improbability) of some objects or processes in the world, and therefore (since that property is asserted to be shared by a set of objects and/or processes), there is a class of natural-world phenomena, a class defined solely by improbability, that evolutionary theory can't explain because it is not a similarity class in the terms of models based in evolutionary theory.  Their only common property, improbability, does not define a class that enters the theoretical laws and models of evolutionary biology.  Therefore the class is asserted to require some other kind of explanation, an intelligent design explanation.

There have been two general kinds of counter-arguments offered in the various critiques of current approaches to ID.  One focuses on the probability estimates.  An important component of this critique is the argument that "improbability" is not a property of an object or process but rather is a characterization of an object or process with reference to some probability density function (PDF), and the alleged improbability of some biological structures and processes is in large part due to an inappropriate choice of PDF for the estimates.  Thus the 'class' formed by highly improbable objects like bacterial flagella and blood clotting biochemical cascades is an artifact of the (idiosyncratic, unjustified) choice of PDF rather than being some intrinsic property of the phenomena.  "Probability" is not an inherent property of an instance; it is a description of a relation between an instance and a PDF.

The second general sort of counter-argument is to the effect that allegedly unevolvable structures and processes (whether the probabilities are correctly estimated or not) can in fact be accommodated in classes appropriate to causal models from evolutionary biology, and are therefore explained by those models.  Thus it is argued that while "direct" incremental evolution of some biological structure may not be possible, indirect routes (cooption, scaffolding, etc.) can account for the naturalistic evolution of the objects and processes.  In addition, "direct" incremental evolution may be actually be possible given that different evolutionary operators induce fitness landscapes with very different topographies, so what appears to require saltational cliff-climbing on one fitness landscape might be simple one-step-at-a-time incremental evolution on a gentle slope of another landscape.  On those arguments the probability estimates do not define a class of phenomena that must be explained in some way that evolutionary theory doesn't provide.

In addition, ID has serious explanatory problems.  By explicitly disavowing conjectures about the number and nature of purported intelligent designing agents and by avoiding hypotheses about the means by which abstract designs are transmitted to or implemented in matter and energy, IDists deliberately eviscerate their ability to provide a scientific explanatory model.  With no hypotheses about designers or mechanisms of design implementation, nothing holds the class of improbable structures and processes together except the (inappropriate) probability estimates.  The examples that have been offered have no properties in common but the probability estimates.  They constitute a mere oddity shop of disparate phenomena bound together by nothing but purported improbability.

The class of allegedly improbable structures and processes offered in the ID literature floats alone in empty conceptual space, unconnected to any causal or correlational explanatory model.  The class of improbable phenomena is not part of a relational structure of classes to which natural (or non-natural) laws and generalities apply.  No laws or generalities have been offered by IDists that go beyond a mere claim of the existence of a class of improbable instances.  IDists offer no testable hypotheses about relationships between the class of purportedly improbable instances and anything else in or out of the world of physical matter and energy, and so there is no scientific explanatory power in ID.

****************
btw, I note with fascination that this board censored John's use of H***!

RBH