Thinking litigation?
That’s the price you pay for trial and error with no foresight. But I really was addressing the question of information. And natural selection is an amazingly efficient mechanism for adding information to gene pool, whereas intelligence is an exceedingly wasteful and costly method.
It’s not that simple. Entropy and information are closely connected, and figuring out how so much information, so specific to all sorts of complex adpatations, got together is no mean feat.
Can you elaborate on this? I’m not familiar with what you are talking about.
In terms of information efficiency, this process is way out of control. And remember: the more effective your simulators (and for creating natural creatures with parts that are small enough to even operate on the quantum level, they’d have to be so effective that they basically recreate reality!) the more demanding the collection and manipulation of information becomes. Natural selection never needs to bother with ANY of that!
Right: natural selection gives us what we see in the fossil record: gradual changes leading in directions that were ultimately never anticipated when the direction first began. Designers, for instance, if they wanted flying creatures would create all sorts of prototypes of birds and wings that might not work very well. But why would a designer create running, then hopping, then short gliding, then controled gliding, then indefinate and propulsive flying? Why not get to the point?
And where would those parts have come from? Would the individual parts not be irreducibly complex, but the sum of the parts would be? Perhaps I misread, but that sounds similar to Behe’s claim about the bacterial flagellum.
A single mutation could, in principle, cause a novel assemblage of some number of other proteins if the mutation had a profound enough effect on the structure of the protein encoded by the mutated gene in quesiton. It would be highly unlikely (though far from impossible, given the time scales and numbers of permutations we’re dealing with on the species scale), but such a novel complex could confer a selective advantage. And if a sequence coding a functional motif from one protein is swapped through random recombination in frame into another gene, a completely new, naturally-arising fusion protein can be generated that can link disparate functions and associations together into one entity. Subsequent mutation, coupled with drift and/or selection, can scramble the evidence over many generations sufficiently to make deconvoluting the evolutionary history of a particular protein quite challenging. It may appear, at first glance, to have come almost from out of nowhere. It is this gap of uncertainty into which ID typically rushes, but as genomic data builds, and computers get faster, those gaps are rapidly closing.
To briefly summarize, what exactly does it take before I know something is sufficiently complex to rule out natural selection? All sorts of things fall under the heading of “presently unexplained”, but they don’t provide proof by themselves of ID. If incomplete knowledge always places natural selection in serious doubt, then it’s reasonable to conclude that only complete knowledge can restore confidence in it. But this is an absurd requirement to place on any scientific theory. Simply saying “It’s too complex” doesn’t show anything; you have to demonstrate this somehow. By what means? It’s the sort of argument creationists make when they claim evolution is disproven by gaps in the archeological record, the supposed lack of “missing links”. Well, what is proven by such a gap, that we lack complete knowledge, or that we have knowledge of volitive and instant complexity? The fact that natural selection can survive despite myriad such gaps is there’s so much other supportive evidence in other analogous examples, in addition to evidence provided by other disciplines. So far as I can tell, ID has no supportive “evidence” except the gaps, and as any theist will be happy to remind us, absense of evidence is not evidence of absense.
As it is, ID explains nothing, and provides no way to predict anything such that we can test it. If we synthesize living orgamisms ourselves, we’ll effectively have complete knowledge about them; but only God has the kind of knowledge to justify a verdict of ID in its current form. God is assumed by the ID “theorists”, so at present it seems their position is hopelessly circular.
What is your cost function? If you are measuring direct designer involvement, you’re right. If you are measuring time until you get to a result, time is cheaper. Genetic methods take a long time to converge, but they are simple. Design methods take less time, but are more complex.
That is another argument against ID. Since the environment changes, you’d either have to go back and patch your design, or let evolution change it to match the new environment. So, evolution happens anyway. Information in this sense is really the number of “bits” of info in the genome. Assuming a sufficiently advanced technology, assembling a gene should be simple. Designing one would be harder, but you’d assume they’d have a cookbook of genes and pathways and the other, more complicated, stuff the human genome project is revealing. Information is independent of how it was assembled. The amount of Shannon information in a message is independent of you tossing it off in five minutes or a team of 20 wordsmithed it for a week.
Let’s say you are generating a test for a bunch of logic in a circuit. Programs exist to parse the circuit, and use algorithms to create a set of ones and zeros at the inputs to detect all the faults inside. Genetic algorithms have been written that start with a random pattern of 1s and 0s, determine its coverage, and then mutate it in several ways. Each of the mutant children are graded, the best kept and the worst discarded. Given enough cycles, you eventually get to a good test without understanding the design. These things work, and are simpler than the algorithmic approach, but usually don’t get as good coverage and take longer to run.
Only if they were after an optimal result. If they were trading off time or effort for quality, they can make do with simpler simulations and heuristics. Considering the inefficiency of the result, I don’t think you can say they took the time to do an optimal job. That pretty much rules out the God hypothesis, since God could do an optimal job with no effort.
Now that I agree with. A designer wouldn’t have taken so long to go through the intermediate steps, nor would there be any reason to release them in the wild assuming they were even necessary. Since the designer would know how to design for flight, why develop a glider? This is another good argument against ID.
To re-iterate, my point was that if we have evidence of mutational saltation: huge and improbable all-at-once multi-dimensional jump in functionality, then that certainly wouldn’t be evolution at work. While Behe goes wrong when he argues that any final product we find that appears to require a whole host of complex parts must be intelligently designed, I’d certainly be willing to concede some sort of intervention if we had evidence that all the parts of such a system simply jumped into being in a single generation. That’s all I’m saying.
BTW, regarding quantum mechanics and life, I’ve yet to see a convincing argument for the need for quantum anything, except that that’s ultimately how atoms and molecules stick together. Magic velcro might work just as well. You don’t need to perfeclty simulate molecular orbitals and hydrogen bonds to model biochemistry. Largely classical approximations work very well for computational purposes; and as it is these classical models get intractably complex very quickly. Even relatively small molecules like glucose display negligible quantum weirdness in the temperatures and densities in which we find life, so there’s little justification at this point to posit a need for such weirdness to explain biological systems. Stickyness and complexity do the job pretty well without biologists getting physics envy.
Perhaps I lack imagination, but I can’t imagine what such a beast would be, or what reliable test I could use to conclusively demonstrate design, lacking complete knowledge. For example, if I came across a complex-looking fossil in some stratum of rock that I am asurdly confident predates any known instance of multicellular life, what do I now know? Well, maybe it was designed; or maybe complex life is older than I thought; or maybe aliens put it there; or maybe its ancestors rode a chunk of ejecta lofted from an impact on Mars three billion years ago. In any event, perhaps this line was outcompeted by simpler Terran organsims, and hence its a chance one-off. If I don’t find any other such fossils on Earth or Mars, and no aliens either, I still don’t have any proof of a designer. All I know is I’ve got something I can’t yet explain.
If you could give me a specific example of something you think I might be forced to conclude is a product of ID, I then might be able to evaluate that. All I’m aware of, or can think of, are the extant examples cited in claims made by ID theorists, which all have been quite conclusively debunked, so far as I know.
Mutations can involve quantum “weirdness:” most commonly by tunneling (protons go from one site to another, changing the chemical bond). And it may be involved in a whole host cellular processes like enzymes.
I actually emailed Behe a few weeks ago, I’ll post some of the relevant bits below (I’m not posting the whole email):
Behe wrote:
I had also written:
To which Behe also had responded:
:dubious:
FTR-The exchange was actually quite civil and nice actually, although I don’t agree with him.
For those who are interested: Here’s the entire exchange, it’s on my ‘myspace’ blog.
Hm…You know what, I’m actually wondering if posting a private email on the next (either via myspace or here) is a good thing to do…I hadn’t really thought about it until literally right now.
I can see why it could be a bad thing to do, but the communication that I’ve had with Behe (which is very very little) doesn’t reflect badly on him, I don’t think…
Can someone give me some input on whether or not posting this sort of thing is a bad move?
Thanks,
Ask Behe’s permission. Be sure to tell him that you’ll post the exchange unedited, and be ready to follow through on that promise.
That’s a good idea, I’m going to take the exchange down from myspace until then.
Mods-can you remove my first post to this thread? The one with the exchange.
Thanks,
You can’t get mutations through various processes of deamination and alkylation? Ionizing radiation? Reactive oxygen species? It has to be proton tunneling? Keto-enol tautomerization is a more pedestrian process by which the nucleotide base can spontaneously rearrange itself and become mismatched, I think, and does indeed involve a reconfiguration of electrons that can only be explained comprehensively by quantum mechanics. Actually, everything I discussed above inovles chemistry, so it ultimately involves the reconfiguration of molecular orbitals, and of course that’s a process only completely described by quantum mechanics. What I’m saying is that’s fundamental. Nothing is “explained” about life except that the stuff it’s composed of sticks together this way. It’s not clear to me that you couldn’t get life if things stuck together differently, nor is it clear to me that you need a comprehensive description of every electron and nucleus in a molecule to model it well enough. Computational chemistry and biochemistry would be pretty much impossible if that were true, because accurately simulating even simple atomic orbitals in atoms much bigger than hydrogen is an intractable computational problem at present.
Isn’t this questionable? The mutation of a control gene such as that discovered by Walter Gehrig’s group at the University of Basel for the eye, has a drastic effect on how eyes function.
I agree this isn’t the creation of a new species in one jump (saltation) but it certainly is an “all at once change” in a major component.
Wouldn’t a big difference in the DNA of two species that are closely related otherwise be such an indicator? My understanding is that if you compute something like the Cartesian distance between the DNA of species, it varies proportionately to the distance between these species on the evolutionary bush. If you found a species that was an outlier - that is seemed closely related in the fossil record and in many characteristics, but far apart genetically, you might have a candidate. Of course you’d have to confirm that the apparent relationship is real, but ID doesn’t seem to include a complete redesign.
You might have a section of the bush distinguishable from the rest, which would indicate that the common ancestor of that session got ID’ed.
No such beasties that I know of, but that is the kind of thing IDers should be looking for.
Sounds like you are describing convergent evolution, which is easily accounted for, and in fact quite elegantly explained, by natural selection.
Like I said, you have to confirm the relationship. You can genetically identify the true relatives of a convergent species in this case, if there was ID you would not be able to.
It isn’t a question of “can’t” but that quantum weirdness is one of the things that contributes to mutation (as well as, perhaps, the operation of a number of subcellular structures) and our understanding of mutation would be incomplete without it.
Actually, KE tautomerization is just a fancy case of quantum tunneling, I believe.
True, but again, there may well be subtlties that we have yet to appreciate, and I wouldn’t want to hang my hat on the idea that none exist with any relevance to evolutionary history.
LOL! Did you note how the editor of the piece must have added the [ness] to the bit about “eyeless[ness].” The biologist was referring to the gene "eyeless’ in fruit flies, but the editor must not have realized it and thought it was a gramatical error that needed correction! Pretty amusing.
To be fair, I don’t think this is an example of what I was talking about at all. Control genes don’t actually involve a single mutational leap that creates all sorts of complex interworking structures.
I was thinking more along the lines of something like mammalian blood clotting. Behe thinks it sufficient to show that such a system is such that removing a part breaks the system: it is a remarkable Rube Goldberg setup where all the parts are necessary. Unfortunately for Behe, there are actually plenty of plausible ways that the clotting system could have evolved gradually. My point, instead, was that of Behe could show that something like the system of blood clotting wasn’t just complex and interlocking, but rather that it arose all at once rather than gradually, THEN he might have something.