I was interested to read Larry Moran’s post about the “Top Ten Darwin and Design Science News Stories for 2011”. As
an over a zealous evolutionist Larry is rather contemptuous about the whole Intelligent Design (ID) thing. What particularly piqued my interest, however, was that Richard Johns’ paper entitled “Self Organisation in Dynamical Systems: a Limiting Result” came in at number 10. The reason for my interest is that I looked at a preprint of this paper a few months back and remarked on it in the following blog posts:
Richard Johns himself very kindly got back to me on my comments and his response can be seen in the above posts. As Richard didn’t respond to my response to his response I was left feeling that there was unfinished business here.
I very much regret the polarized quarrelling between the anti-evolution community (who appear to have picked up Johns’ work) and scientists like Larry Moran. I myself respect professionals like Johns and Moran and I don’t really want to fall out with either. So, whenever I approach a competent technical paper like that of Richard Johns' I don’t particularly feel that I’m on the side of anyone and therefore under no partisan obligation to either defend them or attack them. I’m a disinterested party whose main interest stems from a curiosity about the riddles and mysteries our world confronts us with, although it’s true that I’m having a serious romance with theism.
Although I don’t share his anti-evolutionism I do have one thing I have in common with an “ID” leader like William Dembski; namely, the understanding that in terms of absolute probabilities high improbabilities cannot be banished from our science: We may be able to arrive at high conditional probabilities such as Prob(Life|R0) where R0 is the class of life generating physical regimes, but this still leaves us with the enigma of the selection of a member from R0, a selection which seems to mysteriously appear out of a huge range of mathematical possibilities thus imbuing it with an improbability beyond measure. R0 has a rarity in platonic space which means its existential weighting is inexplicably skewed far in excess of its statistical weighting. For Dembski this fact leads to the conclusion (via his assumption of equal a-priori probabilities and his explanatory filter) that intelligence has selected our physical regime from R0. Ironically the problem with the existence of a member taken from R0 is best appreciated in the context of Multiverse theory: Multiverse theory attempts to place R0 within a greater physical regime, call it R1, whereby it is hoped that Prob(R0|R1) is high. But if pressed this quickly leads to a “Turtles all the way down” regressing series of the form of Prob(R1|R2),…...., Prob(Rn-1|Rn), ultimately leaving us with an enigmatic Rn whose probability, if it is not meaningless, appears to emerge from an incalculable number of possibilities.
But although I believe workers like Johns and Dembski have lessons for us I am not wholly uncritical of the anti-evolutionism that has so thoroughly got hold of the community who identify themselves under the rubric “Intelligent Design”; their anti-evolutionism, due to an impassioned and angry polarization, is now all but irrevocable as they cannot backtrack without loss of face before their evolutionist critics (but the vice versa is also true – polarization cuts both ways). This combative atmosphere is not conducive to self-criticism.
Richard Johns’ paper is a case in point. I could see no fault with the technicalities of his argument; the problems were more to do with the interpretation and use of his results and this interpretation and use is encouraged by the combative pressures the “ID” community is under. Richard uses a cellular automata system to draw his conclusions about limits on the dynamical generation of complex configurations (what he calls “irregular” patterns). He refers to a particular cellular automata program as a “dynamic system” which in turn I identify as the computational equivalent of a physical system taken from a subclass of physical regimes, R. In summary my criticisms were as follows:
a) The concept of complexity (or “irregularity”) used by Richard lumps together, biological complexity, fractal complexity, and the output of random number generators.
b) Richard’s limitative theorem states that a large maximally irregular object cannot appear by self organization in any dynamical system whose laws are local and invariant. But we know that there are such things as algorithms (which could be written in cellular automata form) able to generate fractal and disordered patterns in realistic times. These algorithmically generated configurations classify, on Richard’s’ definition, as large configurations approaching maximal irregularity. Therefore at first sight these latter facts appear to contradict Richard’s limitative theorem – or at least in the way he has stated it.
c) When one reads Richard’s paper one realizes that his limitative theorem is not as strong as it looks because one finds that what he really means is that a blindly selected dynamic system is, in terms of probability, very unlikely to be one that generates a specified subclass of complex configurations in realistic time. This conclusion is, I believe, correct; but if an outcome is improbable that is not to say that it is impossible.
d) If couched in terms of probability Richard’s limitative theorem is correct, but because the theorem is, in fact, referring to improbable outcomes we are lead to overlook the possibility that there may be members of a class R (although it may be a very small class) that have a realistic probability of generating a specified class of irregular configurations in a realistic time.
e) My conclusion is this: In drawing our attention to self organization as an improbable possibility the limitative theorem fails to rule out OOL and evolution as a mathematical possibility.
Richard has effectively hamstrung the possibility of self organization by demanding that the physical regime be chosen at random, “blind watchmaker” style (Thus making it very unlikely it be from the life generating class R0). Hard-line anti-ID evolutionists like Richard "Blind Watchmaker" Dawkins would perhaps be inclined to concede this practice, thus, ironically, conniving with the anti-evolutionist community that OOL/evolution, by definition, is not resourced by the improbable pre-condition R0 . It is surely an irony that Richard’s over interpretation of his conclusion actually falls over if one assumes a creative intelligent agency is available; for in an ID context improbable and/or mathematically rare classes of conditions are the name of the game! If a small class of R exists which promotes the self-organisation of living structures then an intelligence with sufficient processing power and motive could presumably find that class...in other words the physical regime is not chosen at random!
In his response to my original post Richard commented as follows on the possibility of self organisation being an outcome of a carefully selected physical regime:
…..in that case, self-organisation theories of evolution will be in a difficult position. For they will then be committed to the claim that living organisms are algorithmically (and dynamically) simple. In other words, living organisms are like Pi, merely *appearing* to be complex, while in fact being generated by a very short program. (Vastly shorter than their genomes, for example.)
My reply was:
…..it is wrong to conclude that life must be algorithmically simple for this reason: The space of all possible (short text) algorithms, though a lot smaller than the space of all possible configurational objects, is still a very very large space as far as we humans are concerned. I suspect (and this is only a hunch) that not any old algorithm has the right Self Organising properties required to generate living things - in which case selecting the right algorithm is then a computationally complex task; that is, life is not algorithmically simple in absolute terms.
One more thing: Imagine that you were given the problem of PI in reverse; that is you were given the pattern of digits and yet had no clue as to what, if any, simple algorithm generated it. The hard problem then is to guess the algorithm – generating PI after you have found the algorithm is the easy problem. So to me life remains algorithmically complex even if it’s a product of SO.
That’s where the discussion ended. Once again the same lose end is left dangling; namely, the fact that in order to contrive a physical regime taken from the class R0 favouring the self organization of life requires an extraordinary level of computational complexity to sift through the possibilities and secure a member from R0. If OOL and evolution have occurred then it entails quite a “miracle”! Trouble is, the expression Prob(Life|R0) seems to have the effect of obscuring that “miracle”. As a consequence many evolutionists and anti-evolutionists continue to inappropriately caricature the development of life as a putatively “blind” process; hence humunclus ID is an abomination to zealous atheists but the darling of anti-evolutionists.