Jason Pratt on ID

I got this note from Jason Pratt:

The question was recently asked, whether failures of Darwinism can be pointed to, similar to the crop failures resulting from trying to apply Lamarckism.In one sense, the obvious answer is that no such similar failure could_ever_ occur; because Lamarckism has a bit of leeway in its basis proposal which could allow the introduction of intentional development (e.g. a person _chooses_ to do this or that, and so passes on any physiological advantages thus gained genetically); whereas neo-Darwinian theory explicitly and exclusively proposes non-directed accidental effective mechanism. Consequently, any directed experiment (whether success orfailure) necessarily includes an element in its process and result which is supposed to be utterly foreign to the whole point of NDT process and result. Parts of the biological mechanism can be tested, and inferences can be drawn about how well the results fit into the basis theory; but the basis theory itself cannot (in this case) be validly tested.The results of testing parts of the mechanism can of course be combined with inferences grounded in systematic observations about actual real-world behavior; with these being compared to inferences grounded in what thebasis theory itself should be implying.So, for instance, if gradualistic NDT is correct, there should be asignificantly large number of 'link species' still in biological play.Mutant B from species A, isn't going to increase its numbers unless it canstill breed with species A. Ditto for Mutant C, D, etc. At some point (callit species X--the changes all being gradual, per gradualism), X shouldn'tbe able to breed with A anymore, and so an actually new species would be inexistence.(Taxonomically I suppose it wouldn't be called a species--a genus maybe? In popular pro-NDT arguments it's still called a new species, generalizing for the sake of simplicity--and I'm okay with that; the principles would be ineffect either way.) But even once X is in existence, there also ought to be a significant number of transitional species still around, able to interbreed with oneanother fairly successfully. (Even if, say, species F has gone extinct in the meanwhile, species K still ought to be able to breed fairlysuccessively with species D et al.) Humans, for instance, ought to be a rare exception--we can't breed with anything else, all our previous transitional species having gone extinct in the meanwhile (maybe because weintentionally killed them off?)In point of fact, however, we humans _aren't_ the rare exception. Cats, for instance, cannot breed with skunks, even though these are genetically very close to feline; just as we cannot breed with gorillas or chimpanzees. Thisis the very common norm. Whereas, even if X can no longer breed with A,there still should be x-number of genetically distinctive intermediary interbreeders between them, _all_ of which are capable of breeding with each other (except for X and A). It's also rather hard to see how multiple kinds of DNA configuration (different numbers of chromosomes, different amounts of DNA per chromosome,etc.) could effectively be transmitted once accidentally generated.A mutation to a gamete-producing cell, or a copy error from such a cell while producing a gamete, could theoretically generate a change of that sort (and it should be obvious that only cell mutations of this very limited sort would be in any position at all to generate mutants--heartcell mutations don't generate mutant offspring for instance). The question is whether the resulting mutant--assuming the mutation even succeeds in running the gamut of sheer luck against any particular gamete, up to thepoint of becoming a breedible entity that is actually performing nominal breeding operations (so to speak)--is carrying anything in his/her/its own gametes that could successfully pair up with non-mutated gametes (keeping in mind that only one pair of the mutant's chromosomes would be carrying the mutation in the first place at all--half the gametes being fired by the mutant will be altogether non-mutated, therefore not possibly sending along the mutation.) And this is only at the macrobiological level. The feasibility of the theory doesn't (currently) hold up very well at the microbiological level either.
Preliminary microbiological restrictions----------------------------------------
The key restriction that must be remembered, is this: **If a protein ismade from a random order of amino acids, chances are nearly certain the protein will do nothing at all.** Furthermore, in many (though not all) cases, something must already be in existence to be the _receiver_ of the protein behavior. Some proteins behave in ways that build such receiver-units, of course. The point is that even if a protein's order might do something, it usually needs a receptor already capable of receiving it to work. Otherwise it might as well be totally useless.Thus, (at least) five levels of protein functionality-composition: a.) the new protein is totally function less in any possible situation (within that organism); b.) the new protein would function if there was a receptor, but there isn't one; c.) a receptor exists which can receive the effect of the protein, but theuse is harmful or effectively neutral; d.) a receptor exists, which can _usefully_ use the protein;ande.) the protein builds a receptor. (It may be possible that this always results in a receptor which canusefully use a protein that does already exist; but I haven't yet heardthis for sure.) The complexity (and restrictions) intensify when we consider that sometimesa road (so to speak) must already exist to get the protein _to_ an alreadyexistant receptor.Aside from questions of original mutations _accidentally_ creating DNA sequences which generate proteins that build receptors and/or roads; anysubsequent random mutation in a DNA sequence becomes exponentially more improbable to generate useful (not even merely useable) proteins the more accidentally complex (so to speak) the organism _already_ is. Besides receptors, and roads (the chemical paths between receptors andDNA), the relevant reactions also require 'catalyst' chemicals. In many cases: no catalysts, no reactions at all. These protein catalysts are the'enzymes'. Enzymes are terribly important in another way, too. They not only helpcertain chemical reactions to work at all, they also help speed reactionsup to rates that are useful. All known enzymes speed up protein-receptor chemical reactions by a factor of at least a million. Factors of ten billion to a hundred trillion are not uncommon (i.e. a reaction that would normally take 3000 years happens in 1/1000th of a second.) One way or another (or sometimes both), enzymes are absolutely necessaryfor the other existant processes to work; thus adding yet another development-complexity factor--and adding yet another way in which an accidental copy error of the DNA in the chromosomes could (much morelikely) mess something up (rather than adding functionality). As I have been stressing, any addition to (existant) complexity not only adds development time constraints (especially when we consider that someextinct species _already_ had biological and genetic systems virtually as complex, at the microbiological level, as modern animals--subtractx-hundred million years for _their_ development time). It also adds increased liklihood that a random mutation will result in something that runs _against_ the existing system.Granted, a system could be complex enough, _in the correct ways_, to bemore tolerant to random copy errors. (Though mere toleration isn't the sameas positive increase of effective complexity.) But this level and (far moreimportantly) type of complexity cannot be legitimately posited to _already_exist to offset the problems which must be overcome _before_ such a system(on naturalistic evolutionary theories) _does_ exist.Also granted, any redundancy in DNA molecules might help overcome thedemerits of a bad copy error (mutation); but accounting for the developmentof those redundancies themselves introduces more complexity (and tightensthe time constrains for species development), even if it doesn't actually harm the organism. Which itself is somewhat questionable--a newly mutated creator-strand of protein 'x' will at least be taking up resources making that many more protein 'x's which could have been spent (with more overtprofit to the organism) elsewhere. Any excess of those proteins (i.e. produced by conveniently redundant protein generator sequences) might causeproblems elsewhere, too.In order to prevent runaway reactions of exactly this sort within a cell (or elsewhere outside the cell), a cell must be able to turn an enzyme group on or off. This adds _another_ level of complexity to the system. (A typical 'off' signal is generated by side-products of reactions, whichunder NDT must accidentally happen to inhibit further enzyme behavior.) More complex information can be carried by RNA strands than by simplerproteins (and their amino acids). RNA resembles DNA in structure: A,G and C nucleotides, plus a U which is a more unstable molecule than thymine butotherwise serves the same purpose, matching to any As. RNA functions a lot like portable DNA, but (because of the U) it isn't as stable. Also it doesn't have a double-helix structure like DNA. Evolutionists tend to think that RNA evolved first out of simpler proteins; and then DNA out of RNA--which again tightens the time constraints. A lot of the sequences in DNA strands don't seem to be contributing at allto the cellular reaction. Recently, biologists have proposed that this'junk DNA' may actually be equivalent to RNA, and perform much the samepurpose.The problem with this, is that the genetic sequences in question are _not_in fact RNA. It's attractive, but ultimately not very feasible (under thebasis NDT anyway), to propose that all such 'not-junk' sequences are the original RNA codes evolved into DNA upon which the rest of the DNA sequence then developed, perhaps melding various strands together. To whatever extent such sections of DNA do behave like RNA, though, more development-complexity with its inherent functionality-and-development-time restrictions must be allowed. This is why I suspect we'll be seeing a push back to 'junk DNA' eventually: it may require more complexity than 'mere'DNA, but requires _less_ complexity in development than RNA-equivalence strands. Though there's nothing saying that a DNA strand couldn't have someRNA-eq _and_ junk, of course. Ultimately, however, RNA-eq ability tends to lend weight against gradualistic neo-Darwin evolutionary theory, per se.These are some, by no means all, of the problems with the _basis_ theory(NDT). The problems are rising, because what we are learning doesn't square with the foundational theory being proposed. I think sooner or later, the basis theory itself will have to be significantly revised--which happened before, too: what we were discovering didn't mesh with the originalDarwinian theory (or its various alternatives, up to that point), so in1941 (notably at a meeting of the _Geological_ Society of America), a suggestion was made that geneticists should join hands with morphologists, taxonomists, zoologists and paleontologists to try to come up with a modernized version of Darwinian theory, rebuilding it from the most recent finds in their fields. (This mixture, happening where it did, gives evidence of all this being done in service to a strong and overarching ideological committment. Three guesses as to what _that_ might have been...) Together, over the next few years, these specialists created the modern synthetic theory of evolution, or as it came to be known' neo-Darwinian Theory' (or NDT). These new NDTheorists emphatically rejected the Lamarckian theory of acquired characteristics (e.g. a giraffe reaches for taller branches to eat, stretching his neck, and his descendents receive the advantages ofthis); and also rejected Darwin's suggestion of environmental induction ofheritable variations (which would work according to the same principles as Lamarckian theory but in reverse.) It seemed clear to them that neither environmental influences nor acquired characteristics could affect the germ cells (the gametes created by males and females, distinct from the somatic cells everywhere else); and heritable variation could only stem from changes in the germ cells.(Note: I don't know whether they recognized the existence of mutagenic environmental effects or not, but if they did they must have thought these were too few and too inherently harmful to contribute effectively to the number of positive mutations necessary for the theory to work.) Since they were unwilling to accept environmental influence as a cause of variation (probably because the natural evidence for it was lacking, but also possibly because it tasted too strongly of _directed_ evolution), they settled on sheer random copy-error as the source of the variations (a source Darwin had rejected).They agreed that the majority of these copy-errors (as experimentally demonstrated) would be detrimental to the organism's chances of surviving and/or breeding (which are two rather different processes--an improvement in one might be a detriment in the other); and much of the remaining possibilities would be effectively neutral. But, they figured that at leasta _few_ mutations might be beneficial in survival and/or breeding (without a sufficiently detrimental offsetting result in one compared to the other);and that the theory would work sufficiently well on these few positivemutations (plus perhaps the more common neutral mutations combined with eventual changes in environment, though only a fraction of a fraction ofthese would make those previously neutral changes positively beneficial), given the vast time frame available for the process to work.About a decade later, Watson and Crick discovered the DNA structure, and identified the heritable variation with random copy-errors in DNAreplication.This was, however, also the first in a series of dramatic scale-ups ofrecognized complexity in biological organisms. Notably, Crick decided that the vast complexity of DNA could not possibly have been created in the geological time-frame, given the process of random copy-errors plus naturalselection (which would almost always select _against_ the resulting mutant,if it did anything at all.) His solution was to propose that Terran DNA had been seeded on the planet millions of years ago by aliens. This was mocked by the establishment, partly because he was only putting the natural development of even _more-complex_ entities _further_ back in the cosmic time-scale (actually increasing the problem); and partly because it smacked too much of theism (i.e. the only reason Crick proposed naturally developed aliens was because he refused, as a dedicated atheist, to propose God.) Now, I have several problems with what the ID fellows are often doing. Butwhat they _are_ doing, isn't much different from what the NDT synthesisteam was doing back in the early 1940s. They're noting the basis theory isn't just failing to account cogently for operation in the real world, but actually runs _against_ the grain of significant parts of what we'refinding; so they're proposing revisions of the basis theory to try toaccount better for what we _are_ in fact discovering. It's a fact of history that the original NDT team members (along with many of their successors) weren't much (if any) less motivated than modern ID proponents, to be thereby attempting to defend and propogate an ideology (naturalistic atheism in their case, by and large); and it should be obvious that even if one of them (or someone supporting and agreeing with them) came up with crackpot theories (e.g. panspermia) to account for the data, this didn't obviate the other work he (or the others) had done.