Thursday, December 20, 2012

Paul Nelson, Computer Simulations and .... Gravity.

Evolutionists and Anti-Evolutionists continue to slug it out.

This video featuring a presentation by Homunculus IDer Paul Nelson has stirred up a response from the evolutionary establishment. The relevant links are:

Paul Nelson is not just a Homunculus IDer; he is also a Young Earth Creationist. But there are YECs and YECs; As far as I can tell  Nelson is not one of the hardened “heretic burning” religionists exemplified by the likes of Ken Ham and his ex-business partner John MacKay (In fact MacKay is probably lunatic fringe). Nelson, like most of those in the Humunculus ID community (or “HIDs”), seems to be a reasonable and intelligent man; somebody, in fact, that any other reasonable person could do business with even in the face of mutual disagreement; Nelson probably has a YEC background, but he is nevertheless prepared to work with non-YEC Christians.

The argument that Nelson has triggered seems be about the relative importance of natural selection over and against other mechanisms that may drive evolution natural history. I’m not going to comment on this matter because firstly I'm not a biologist and secondly it’s one of those grey areas about degree: Viz: Just how important does the evolutionary establishment claim natural selection to be as an agent of change? Have they overestimated it or underestimated it? Contrariwise just how unimportant is Nelson claiming natural selection to be?  Is he overestimating or underestimating its unimportance? The outcome of this kind of dispute is that it has a tendency to degenerate into a “he said, you said” farce.

Although Nelson is the kind of guy I can respect and moreover I think the evolutionary establishment needs the well-motivated and intelligent criticism (albeit largely negative and destructive) we get from people like Nelson, I still feel that there is something highly unsatisfactory about the underlying motivations and concepts behind Nelson’s attack on evolution. My sense of unease is bound up with the pervasive "God Intelligence did it vs. Natural Causes did it"  paradigm with which Nelson frames the issue. This framing is evidenced by the implicit reference Nelson makes to William Dembski’s explanatory filter.

The explanatory filter is an epistemic method whereby one works in sequence through the possible explanations of an observed configuration starting with "natural causes" founded in law and disorder. (i.e. what the HID community inappropriately refer to as “chance and necessity”). If law and disorder fail to explain the configuration this leaves us with, by default, intelligent causes. Certainly, within the confines of the cosmos the ideas behind the filter are robust:  Law, disorder and intelligence (perhaps even alien intelligence) cover all possible agents of change that we can conceive. But though this may be the case, in a theistic context the filter proves to be problematical. To see this let’s imagine for the sake of argument that establishment evolutionists “prove” their case beyond reasonable doubt. Using the explanatory filter it would then appear that the game is up for theism:  The explanation that “Intelligence (=God) did it” has been displaced in favour of “Natural causes did it”. To salvage their position, however, the HIDs will then tell us that our universe is fine tuned to favour the generation of living configurations and that this fine tuning has the effect of triggering the explanatory filter's default explanation of intelligent agency. But the trouble is that the explanatory filter cuts both ways: It now suggests that if one can find a law and disorder explanation for the fine tuning of the universe, then there is no need to posit divine intelligent “first causes”. But it turns out that law and disorder explanations have their own problems: To cut a long story short we find that attempts to arrive at a complete law and disorder explanation has the potential to lead to a classic turtles all the way down regress: Law and disorder explanations always leave a residue of brute facts that either have to be simply accepted as axiomatic, or as a trigger for another level of explanation accounting for why these particular givens have been specially selected for and not others. The inevitable problem here is acknowledged by Max Tegmark who gets round the conundrum of special selection with his mathematical universe, a weird and disturbing place where every conceivable mathematical object has been equally reified.

Naturally enough the HIDs want to stop this regression process from the outset by unequivocally showing that law and disorder are incapable of explaining living configurations; for if the situation is allowed to get into a multiverse regress it becomes too abstruse for a decisive victory to be claimed by either side. The consequence is that the HIDs have put down a very high stake in favour of intelligence in the “God Intelligence did it vs. evolution did it dichotomy, a dichotomy of thought encouraged by Dembki’s explanatory filter. Nelson and many other HIDs therefore have a great vested interest in discrediting evolutionary theory. They follow a negative approach which by a process of destructive elimination is intended to leave us with the default of God intelligence as the “cause” of life.  In the video I have linked to Nelson refers disparagingly to the wall surrounding law and disorder explanations, a wall which implicitly excludes thoughts of intelligent “causes”. Therefore for Nelson it’s a stark choice between naturalism and supernaturalism intelligence. William Lane Craig’s support of the Kalam cosmological argument comes out of the same stable: He is adverse to multiverse scenarios because they ostensibly attack his “divine first cause” argument by muddying the waters with a potential "natural causes" regress. Craig, like Nelson, thinks in terms of causes rather than patterns amenable to mathematial description. This causation concept of explanation readily leads to a way of thinking that views God’s involvement as the diametrically opposed alternative explanation to natural causes. (See endnote 3)

Nelson and Craig have categories of thinking which inclines them to put “natural causes” and “divine causes” on the same logical level thus pathing the way for God and nature to be seen as competing agents of explanation. This fails to do justice to the nature of intelligence, especially divine intelligence. I can best express this with a computing metaphor: It is possible to simulate a computer within a higher level host computer. In fact a simulated computer could be just one of many agents that act within the simulacrum of a virtual world rendered by the host computer. In such a simulacrum the simulated computer would stand alongside and have the same logical level as other categories of causation in the simulacrum. But in this scenario computation is actually the outer embracing context and therefore it is of higher logical level than the particular law and disorder regime rendered within the simulacrum.  Intelligence and computation are closely related; both have a dynamic that involves searching, rejecting and selecting, along with the interim production of partial results. So, if intelligence is anything like computation it too is likely to have an abstract, general all-embracing superset definition. It is conceivable, therefore, that our universe is running on (or "in") an intelligence/sentience rather than a "nuts and bolts" computer.  If a host intelligence is the embracing substrate on which our world is being rendered this means that the schism between intelligence and naturalism prompted by Dembki’s explanatory filter is very inappropriate; this schism is a result of placing law & disorder on the same logical level as intelligence. 
Like other HIDs Nelson defines evolution as a dumb process (See here where we find Richard Johns doing a similar thing). But even if the mechanisms of evolution are as the academic establishment would have it, then it is very wrong to call evolution dumb: If evolution is to work its efficacy is likely to be very dependent on the selection of the right physical regime. If evolution works then the aim at a distant functional target (As Nelson puts it) has taken place in the selection of the right physical regime. Intelligence has already been built into the process of evolution and evolution is, in fact, intelligence at work.

But having said that this is not to say that evolutionary theory is sown up.  I’m the first to concede that we may be far from understanding the full set of mechanisms driving evolution. Evolution is a present tense continuous process, but unlike the objects of spring extending and test tube precipitating science the changes it generates are smeared over very large tracts of time. We are like ants crawling over the surface of a huge evolutionary tree (or bush!) trying to reconstruct a highly complex shape.  Our attempts at reconstruction should be admixtured with a certain amount of epistemic humility. But in the polarized and passionate North American political and religious environment the sad fact is that a shift toward epistemic humility is not going to happen.

Other relevant links
Nick Bostrom’s computer simulation theory is relevant to this post. A flurry of links have appeared on the subject of a simulated universe. See:

I touched on the subject of a simulated cosmos here (with tongue in cheek!):

David Deutsch is good on the subject of Physics and Computation: He is developing a theory of physics that only makes recourse to the constraints on computation, an indication of the superset generality of computation. See here:
1) A computer needs to be instantiated by some kind of law and disorder regime, but because computation is a concept that is independent of the exact computing model instantiating it, then the precise nature of the instantiating regime of the host computer is not easily accessible to the inner simulacrum. This is because the computer model reifying the host computer is one amongst many possible models and the formal structure of the simulacrum is insensitive to that model. If, like computation, intelligence is a general abstraction this will mean that the nature of the medium reifying the host intelligence is likely to be all but unrecoverable from the inner simulacrum.

2) As with computation one might expect sentience to be reified on some kind of law and disorder medium. The question then arises: Which comes first: Law and disorder or intelligence?  Law and disorder regimes are too simple in structure to be self-explaining – they cannot have the property of aseity.  Since complexity has no apparent upper limit aseity may be hidden in the upper reaches of sentient complexity. It may be wrong  to think in terms of law & disorder causing intelligence, or, vice versa, intelligence causing law and disorder. Law & disorder and sentience  may not proceed one out of the other but of necessity go hand in hand: Viz: Sentience will explain itself in law and disorder terms, but for the elementals of law and disorder to have positivist/experimental meaning the a-priori complexities of conscious cognition are required to host these elementals and give them context.

3) Patterns vs. Causes: In the final analysis our scientific epistemology only ever reveals the pattern of things; that is, science provides us with what is essentially a means of describing the cosmic state of affairs. In fact we find that many of the patterns the cosmos presents us with are amenable to description using algorithms (what I refer to as “law”) and statistics (what refer to as “disorder”) - hence my use of the phrase “law and disorder”. The concept of “cause” is really a special case term that is usually applied when a pattern can be described with a “deterministic” algorithm that computes the pattern in relatively short time. (Although one could say that a quantum event has been “caused” by randomness - i.e. a disordered pattern - this usage would probably be considered as rather strained).  “Causation”, in my view, is very much a subset category within more general categories of pattern description. It is with the foregoing in mind that I am uneasy about William Lane Craig’s and the HID community’s stress on the causal role of God in creation. This view of God favours the assignment of God to the wrong logical category; that is, as an ancillary homunculus source of causation rather than as an embracing substrate.

4) How Paul Nelson is viewed at Sandwalk

No comments: