Mud-to-Mozart Atheology (Or, Who are the real skeptics?)

_{Gil Dodgen

October 22, 2011

Intelligent Design}

Share: Facebook; Twitter; LinkedIn; Flipboard; Print; Email

I find the “skeptic” claim on the part of Darwinian materialists very interesting and equally illuminating. Darwinists exhibit no skepticism whatsoever about the thesis that physical stuff turned into Mozart by chance. (Don’t try to deny this, Darwinists, that is the essence of your claim. You can try to obfuscate with legion “peer-reviewed scientific papers,” but you’re not going to fool me and many others about what you are actually promoting and advocating.)

I choose Mozart not just because I am a classical concert pianist, but because his existence epitomizes everything that Darwinian theory is totally powerless to explain.

Darwinists, claiming to be skeptics, actually exhibit the antithesis of skepticism — making transparently ludicrous claims and providing a never-ending stream of unsupported extrapolations, based only on wildly imaginative speculation with no empirical support.

How is it that Darwinian atheists are the only ones who get to declare themselves legitimate skeptics? Is mud-to-Mozart-by-chance philosophy the only worldview immune to skeptical inquiry?

Comments

Elizabeth: I have often given an explicit definition of dFSCI, and a personal threshold for biololgical dFSCI (150 bits). Most protein families analyzed by Dirston are wll beyond that threshold. Nor certainly Lenski or Avida.gpuccio_{October 27, 2011
October
10
Oct
27
27
2011
03:59 AM
3
03
59
AM
PDT}

Elizabeth: I am sorry, but I cannot follow you in that kind of "reasoning" (and believe me, I am being extemely generous here). You repeat "reproductive success" as a prayer or a mantra, and seem to believe that words will support your non arguments. “Neutral drift”, “RV” and “NS” are explicit concepts in explicit models. Dawkins knows that well. I know that well. You seem not to understand it. The Lenski and Avida things have nothing to do with dFSCI.gpuccio_{October 27, 2011
October
10
Oct
27
27
2011
03:56 AM
3
03
56
AM
PDT}

I don't know whether it did or not, gpuccio, because I don't know that anyone has attempted to measure it. But it seems to me that it produced functional information in normal usage (an algorithm emerged that could perform a function), and it was fairly complex (the genome of the function would have a very low probability of being generated by any process that did not incorporate differential reproductive success). And if you are going to distinguish between "microevolution" and "macroevolution" can you give the definitions you are using for those terms?Elizabeth Liddle_{October 27, 2011
October
10
Oct
27
27
2011
03:55 AM
3
03
55
AM
PDT}

gpuccio, it seems to me you are attempting to carve nature other than at its joints! "Neutral drift", "RV" and "NS" are not separable in the way you are trying to separate them, and "NS" does not have a "role" in any normal usage of the term. As so often, the metaphors are standing in the way of meaning. Both the generation of novel genetic variants and the effects those variants have on the reproductive success of the phenotype in the current environment are highly stochastic, but both are also biased in favour of what works. Most novel genetic variants will have little or no effect on the phenotype, and whether they do affect the phenotype, the effect may make absolutely no difference to the probability that the organism will reproduce successfully, in the current environment. This means, as you say, that the probability that a novel genetic variant will propagate through the population is a random walk, although the walk may be slightly biased. Imagine the classic drunk at a lamppost who staggers north and south down the street, the probability of going north or south at any one step being 50:50. Now imagine that there is a big cliff immediately south of the lamppost. Most drunks eventually will fall off the cliff, but quite a few of them will get quite a way northward before falling off, and some will even reach the temperance hostel at the far northern end, and be saved from the perils of drunkenness. If a variant is slightly beneficial, all you have to do is adjust your mental picture so that there is a gentle slope downwards to the north, so biasing the chances of the drunk ending up at the temperance hostel as opposed to off the cliff, and also increasing the number of drunks staggering about in the street. If it's slightly deleterious, the reverse will be true - the street will slope slightly upwards to the North, and the cliff edge will claim more victims, and the temperance hostel fewer successes. But the really important point is that at any given time, there are lots of drunks in the street, so that if an environmental change comes along, so that the street changes slope, or even new streets appear at right angles to Main street, with slopes down them, there's more chance that a drunk or two will find themselves somewhere interesting. And this is perfectly illustrated by the Lenski et al AVIDA paper, in which necessary steps to the beneficial steps were neutral, or even deleterious, yet they all evolved. They could not have done so unless neutral drift is reality, and the experiment showed (as have many others, as well as actual biological experiments) that it is indeed a reality. Indeed it is the reason why genetic diversity is so important to the survival of a population, and why small populations are in real danger even if there are substantial numbers of successfully breeding pairs. Variance is crucial to robustness, in many contexts. So let's get back to basics, and lose these high-level terms like "NS" and "probabilistic resources" and "RV" and "drift" and figure out what evolutionary models actually propose. And for goodness' sake, let's stop using the word "random" without specifying exactly what we mean! We are dealing with probability distributions here, and those distributions are not flat!Elizabeth Liddle_{October 27, 2011
October
10
Oct
27
27
2011
03:51 AM
3
03
51
AM
PDT}

Petrushka: Did Lenski's experiment produce complex functional information? Or are you again passing microevolution for macroevoloution? What about Thornton? Reference, please.gpuccio_{October 27, 2011
October
10
Oct
27
27
2011
03:46 AM
3
03
46
AM
PDT}

Petrushka: Briefly: a) In SCOP there are at present 110800 sequenced domains, grouped into 3902 families. In no case have we any evidence of bridges between those groupings. Do you really doubt that, even if that is certainly not the whole proteome, it is a significant sample of it? Have you any idea of how a statistical sampling is done? Are you saying that with 110800 sequenced domains we still have no idea pf the structure of the true proteome? b) Domain sequences do change in time through neutral mutations, with or without drift. But in most cases, they do retain hihly significant homology. How do you think that 1800 of those basci domains have been attributed to LUCA? Because, after almost 4 billion years, we still see homology between the proteins, say, in bacteria and humans. One example? If you blast the sequence of argynil trna synthetase in E. Coli and humans, a very old protein dating back to LUCA, you still get 232 identities out of 599 (39%), and 57% positives. The evalue is 1e-114. Well beyond any suspicion of randomness, I would say. And those proteins have diverged a lot of time ago. c) There were certanly extinctions. Who says they did not happen? Just explain why all the functional intermediaries between basic domains became extincted.gpuccio_{October 27, 2011
October
10
Oct
27
27
2011
03:43 AM
3
03
43
AM
PDT}

Elizabeth, 23.3.1.1.7 You completely missed the point of my post 23.3.1.1.6 (if you were replying to me). I specifically said in the last sentence:
Note: I am not saying that this is ID, to just look at biology as see that it is obviously designed. I’m just pointing out the fallacy that is your made-up design-detection requirements of archaeologists.
GinoB posted a fallacious representation of archaeological design detection, and I pointed out the obvious flaws in it. Of course if we find alien technology, we will immediately know its designed. Biological design is much tougher to scientifically decipher, because there IS a mechanism that can lead to adaptation. The problem there is that few outside of the ID camp seriously question the adequacy of neo-Darwinism to produce the "designs" at which we marvel in organisms. It was "100% adequate" around the turn of the 20th century when it was first adopted, and it has remained "100% adequate" every second since then through all of the challenges that should have been considered as we have observed more and more of the complexity. Every time we find a new level of complexity, the reaction is never "hmm, I wonder if the neo-Darwinian mechanism could have accomplished this", it's "Wow! This mechanism is even better than we thought! This genius design is highly selectable."uoflcard_{October 27, 2011
October
10
Oct
27
27
2011
03:41 AM
3
03
41
AM
PDT}

But how likely is that scenario?
Apparently it is fairly likely, because it occurred in Lenski's experiment, and Thornton seems to have found a historical example.Petrushka_{October 27, 2011
October
10
Oct
27
27
2011
03:35 AM
3
03
35
AM
PDT}

What's the big deal about fixation? Loads of highly prevalent alleles are not fixed. What matters is that there are loads of them.Elizabeth Liddle_{October 27, 2011
October
10
Oct
27
27
2011
03:30 AM
3
03
30
AM
PDT}

It is this impotence of neutral drift that makes the neo-darwinian approach the only even remotely plausible explanation for the rise of protein domains. This is why gpuccio is using that model. I don't see what's so hard to understand here.PaV_{October 27, 2011
October
10
Oct
27
27
2011
03:29 AM
3
03
29
AM
PDT}

I don't know what you call it, PaV, and I don't care very much; the important thing is that it's a key part of modern evolutionary theory. Darwin didn't even know what the source of variance in populations was, but what he noted was that if you have heritable variation in reproductive success, then the successful variants will become more prevalent. What we now know is that the vast majority of new variants are near-neutral, which means that any given population at any given time represents a pool of phenotypic variants of which some variants may prove more successful than others in surviving when the enviroment changes. This means that when we set up Darwinian computer models, we find, to our surprise, that complex functions evolve more readily than perhaps Darwin might have anticipated because necessary but neutral precursor variants are already likely to be there in the population. But let's not trap ourselves into yet another categorical simplification: there aren't three kinds of variant, deleterious, beneficial, and neutral. What there is is heritable variance in reproductive success. This means that (at least in sexually reproducing species) each generation will be a random sampling of the previous generation's genetic makeup that is nonetheless biased in favour of those variants that tend to promote reproductive success in the current environment. You can call that whatever you like, but it's fundamental to current evolutionary theory, and incorporates Darwin's insight that he called "natural selection", so it seems he earns his name on the tag. But what it does mean is that the old cliche that mutations are random and natural selection isn't, is pretty well meaningless (and I wish people, especially Dawkins, would stop saying it!) Mutations and differential reproduction are stochastic process that are strongly biased in favour of what works. In that sense, they aren't even "blind" or "undirected". But I'm not going to argue about labels, because what matters is what the theory actually is, not the descriptive words we use to describe it. Evolutionary theory does not propose that evolution is "directed" in the sense that an ID proponent would use the term, i.e. planned by something with a distal goal that takes selected actions to achieve that goal. But it does propose that it is "directed" in the sense that a river is "directed" towards the sea, even though we cannot predict the course it will take, and the course it does take may be influenced by "random" events such geological changes to the landscape, and even factors caused by the river itself (hence meanders).Elizabeth Liddle_{October 27, 2011
October
10
Oct
27
27
2011
03:29 AM
3
03
29
AM
PDT}

The "Father" of the Neutral Theory is none other than Motoo Kimura. Let me quote from his seminal work: "The Neutral Theory of Molecular Evolution": p. 101 In globins, as we have seen already, amino acid substitutions occur at the rate of about 10^-9 per amino acid site per year. This means that, roughly speaking, in the alpha-chain consisting of 141 amino acids, one substitution occurs on the average every 7 million years. Consider a mammalian species with an average generation span (g) of two years and comprising half a million individuals each generation (its effective population size may be much smaller). Let us suppose that the mutation rate, nu [Greek letter], per generation of alpha-globin is 10^-6. We restrict our consideration only to those mutations that cause amino acid changes. During 7 million years, the total number of alpha-globin mutation which occur in the species is then 2Nv x 7 x 10^6/g = 2 x (0.5 x 10 ^6) x 10^-6 x7 x 10^6/2 = 3.5 x 10^6. Thus, 3.5 million mutations occur, among which only one becomes fixed. Not all of them are unique. 7 million years, 3.5 million mutations, and only ONE is fixed. Is this supposed to explain evolution?PaV_{October 27, 2011
October
10
Oct
27
27
2011
03:25 AM
3
03
25
AM
PDT}

Elizabeth: Please, look at my post #50. It is meant for you too. And I would leave alone the problem of IC for the moment, and stick to the problem of origin of basci protein domains. The reason? Just to keep the discussion as focused as possible. We can discuss IC after, if you like.gpuccio_{October 27, 2011
October
10
Oct
27
27
2011
03:19 AM
3
03
19
AM
PDT}

DrBot: Why are you so chaotic in your reasoning? Let's try some order. Look at my post 38.1.1: "And it is well obvious, even if the dumb logic of darwinists will never recognize it, that neutral mutations in no way change the probabilities of a random walk, and neither does genetic drift, because it is a purely random event, and therefore does not change anything. The term “RV” includes all forms of possible random variations, including neutral mutations and genetic drift. And all RV is accounted for and evaluated in the ID model." and my post 47 (in response to you): "IOWs, the point is: a) We have a domain that apperas for the first time (B). b) To explain that by neodarwinian mechanisms, we need some precursor (A), previously existing, that by RV and NS gave the result of the appearance of B. The only alternative would be that the transition from A to B was merely by RV, that is neutral mutations and drift and similar. But, as both A and B are unrelated long molecules, that is practically impossible: the probabilities are completely against that, and any good neodarwinist knows that." And so on. You say: you are using a model that does not apply to Biology Why? I am using the standard neodarwinian model. RV + NS. The RV part includes all forms of random variaiton, including neutral muattions and genetic drift. The RV part has the probabilistic power of any random system. With the search spaces we are discussing, it is almost completely powerless, if it has to work alone. That's why we are discussing the NS part, that shouls "help" RV in its impossible tasl, by multiplying probabilistic resources when a selected tract expands. And a transition "helped" by NS implies functional selected intermediaries. Is that clear? Some more reflections on the reason why drift is non relevant in the computation of probabilistic resources. Let's say that a functional mutation is selected, and expands, for example, in a population, from one original individual to a population of 10^9, say, because of that positive selection. There is no doubt that, if (and only if) that mutation is really a step towards some future complex functional result, then the probabilities of the final result are multiplied by a factor of 10^9. Which is something, even in the huge search spaces we are discussing. That is the supposed role of NS. Now, let's say that in the same population we have a number of possible neutral mutations in a certain time. And let's say that in that time one drift event will expand one of those neutral mutations. Obviously, if one of the neutral mutations is "favorable" to the final result, and it is expanded by drift, in that case the probabilities of the final result will increase. But how likely is that scenario? Exactly as likely as a similar scenario for each of the other neutral mutations, that in no way contribute to the final result, or even make it impossible. IOWs. if we kust consider the pobabilities of a random walk from A to B (unrelated), all the states have similar probabilities of being reached, both with or without drift. That's why I always say that drift is irrelevant to the computation of probabilities, and favours no specific configuration, least of all a functional one. Is that clear? Please, if you answer, be specific in your objections, so that I may respoind in a pertinent way.gpuccio_{October 27, 2011
October
10
Oct
27
27
2011
03:14 AM
3
03
14
AM
PDT}

DrBot and Elizabeth: Let's talk about neutral drift. Can neutral drift in any way be called Darwinism, or even, neo-Darwinism?PaV_{October 27, 2011
October
10
Oct
27
27
2011
03:08 AM
3
03
08
AM
PDT}

GinoB: "Exactly as I said, no one in the scientific community labels theories as facts. That only happens with ignorant laymen, or intellectually dishonest scoundrels with an agenda to push. Which one are you?" ==== More importantly, which one is Dr Gerald Joyce ??? "Evolution is not a theory for us chemcists" Dr Gerald Joyce said, "Evolution has now been proven a fact, it’s not just a theory anymore."Eocene_{October 27, 2011
October
10
Oct
27
27
2011
02:59 AM
2
02
59
AM
PDT}

Gino B: 43.1.1.1.2:
No one in the scientific community labels theories as facts.
43.1.1.1.3: gpuccio, in response:
The National Academy of Science (U.S.) makes a similar point: Scientists most often use the word “fact” to describe an observation. But scientists can also use fact to mean something that has been tested or observed so many times that there is no longer a compelling reason to keep testing or looking for examples. The occurrence of evolution in this sense is fact. Scientists no longer question whether descent with modification occurred because the evidence is so strong.[19]“
Gino B @ 43.1.1.1.7:
Exactly as I said, no one in the scientific community labels theories as facts.
Can you read English?PaV_{October 27, 2011
October
10
Oct
27
27
2011
02:50 AM
2
02
50
AM
PDT}

For a neutral mutation to become fixed, 1/4N generations are needed. With N=100,000, that means 400,000 years. So, for 4 million years of divergence, that's 10 mutations. Now what about the other 59 million, nine-hundred ninety-nine thousand, nine hundred and 90 other needed mutations? And, BTW, there's a new study indicating that it is NC DNA---transposons---that are responsible for chimp/human divergence.PaV_{October 27, 2011
October
10
Oct
27
27
2011
02:29 AM
2
02
29
AM
PDT}

Which is why neutral drift is so important, of course. Without neutral drift, the Irreducible Complexity argument might have some force. As it is, it doesn't.Elizabeth Liddle_{October 27, 2011
October
10
Oct
27
27
2011
01:58 AM
1
01
58
AM
PDT}

And there is not trace in the present proteime of such intermediaries (A and B are separated unrelated domains, and no sequence is known that is a bridge between them).
You are making several unwarranted assumptions. One false assumption is the we have sequenced a significant percentage of existing genomes and can reliably judge whether domains are truly isolated. Another assumption is that domain sequences do not change over time due to drift. Another assumption is that there never were links and that no linking sequences have become extinct. We of course know that extinction doesn't happen. (Question: what would the status of Australian marsupials be without fossil evidence in South America?) But you find it reasonable to believe that invisible fairies poofed domain sequences into existence, fully formed, perhaps from the head of Zeus. Now that's rational. God of the gaps seems to be the only continuing theme in ID.Petrushka_{October 27, 2011
October
10
Oct
27
27
2011
12:40 AM
12
12
40
AM
PDT}

If ypou go on the SCOP (Structural Classification of Proteins) site, you can access a tool called ASTRAL. By that tool you can ask for the identifiers of domains sharing less than n% identity. 10% is the lowest value you can set.
One problem there is it is only comparing protein coding sequences. To get a full picture of the degree of isolation between domains you need to look at non-coding sequences as well because these can potentially provide routes between domains.DrBot_{October 27, 2011
October
10
Oct
27
27
2011
12:33 AM
12
12
33
AM
PDT}

Why is it false? Remember that we are discussing a model where NS intervenes, not a model of pure RV (including neutral mutations and drift). In a model based on RV + NS, the selected triat has to expand, to gain real advantage versus a purely random model. As I have explained, it’s expansion that multiplies the probabilstic resources. Without expansion, you are still in the random situation.
Umm, OK, so you are using a model that does not apply to Biology. Near neutral 'traits' can accumulate in a population - the population can drift. Look at it this way, if a neutral mutation or other genetic alteration occurs in an organism that is otherwise reasonably fit (produces offspring) then that neutral change can get passed to the offspring, and so on - those neutral changes can spread through a population.DrBot_{October 27, 2011
October
10
Oct
27
27
2011
12:17 AM
12
12
17
AM
PDT}

DrBot: You are forgetting about neutral drift. But... have you at least read my post 47.1.1, to which you are apparently responding? I quote myself: "Why is it false? Remember that we are discussing a model where NS intervenes, not a model of pure RV (including neutral mutations and drift). In a model based on RV + NS, the selected triat has to expand, to gain real advantage versus a purely random model. As I have explained, it’s expansion that multiplies the probabilstic resources. Without expansion, you are still in the random situation. So, what is your point?"gpuccio_{October 26, 2011
October
10
Oct
26
26
2011
11:44 PM
11
11
44
PM
PDT}

No, the intermediates whose traces we should find are those that were positivley selected and expanded at some time. Theresore, those functional. Indeed, more functional then their precursors. ... There is some strange law of nature that erases systematically all molecular intermediaries between basic protein domains, although they were functional enough to be naturally selected and expanded at the time they served as intermediate steps to the transitions.
You are forgetting about neutral drift. You are also falsely claiming that the precursors need to be functional. They don't.DrBot_{October 26, 2011
October
10
Oct
26
26
2011
11:38 PM
11
11
38
PM
PDT}

DrREC: Do we? I know of no such example. Please elaborate. You have to follow the whole reasoning to understand the statement. The reasoning was: We have domain B emerging at some definite point in natural history. We are trying to explain it not as a result of pure RV, but as a result of a transition form some already existin, unrelated domain, through selectable intermediate steps. So, it is already in the assumption that A and B are domains we can observe in the proteome. A can be whatever domain we hypothesize as a precursor. A and B are unrelated, because all basic domains are unrelated, and B originates at a specific time, and there is no trace of related sequences in beings that appeared preciously. So, the basci points are that we need to hypothesize molecular intermediaries, functional and selectable, if we want to explain B with a darwinian transition from any existing domain A. And there is not trace in the present proteime of such intermediaries (A and B are separated unrelated domains, and no sequence is known that is a bridge between them). e) (the first e, you have 2) is also false, but whatever. Why? Please, elucidate. The statement is: "Therefore, A1 had to be functional. Indeed, more functional than A, to be expanded." Why is it false? Remember that we are discussing a model where NS intervenes, not a model of pure RV (including neutral mutations and drift). In a model based on RV + NS, the selected triat has to expand, to gain real advantage versus a purely random model. As I have explained, it's expansion that multiplies the probabilstic resources. Without expansion, you are still in the random situation. So, what is your point?gpuccio_{October 26, 2011
October
10
Oct
26
26
2011
10:58 PM
10
10
58
PM
PDT}

e) Well, the problem is, in the existing proteome, we have A, we have B, but no trace of A1. Do we? I know of no such example. Please elaborate. e) (the first e, you have 2) is also false, but whatever.DrREC_{October 26, 2011
October
10
Oct
26
26
2011
09:57 PM
9
09
57
PM
PDT}

DrBot: Why do they have to be functional? All that really matters is that they are not too detrimental. No, the intermediates whose traces we should find are those that were positivley selected and expanded at some time. Theresore, those functional. Indeed, more functional then their precursors. IOWs, the point is: a) We have a domain that apperas for the first time (B). b) To explain that by neodarwinian mechanisms, we need some precursor (A), previously existing, that by RV and NS gave the result of the appearance of B. The only alternative would be that the transition from A to B was merely by RV, that is neutral mutations and drift and similar. But, as both A and B are unrelated long molecules, that is practically impossible: the probabilities are completely against that, and any good neodarwinist knows that. c) Therefore, we need some intermediates (for simplicity, let's assume one and call it A1; indeed, there should be a lot of them for each transition). d) The idea is that A1 was naturally selected and positively expanded. Otherwise, no help would be gained against the probabilistic wall of a purely random transition. e) Rgerefore, A1 had to be functional. Indeed, more functional than A, to be expanded. e) Well, the problem is, in the existing proteome, we have A, we have B, but no trace of A1. Now, I have tried to express this problem, that I would call the lack of molecular intermediaries between existing basci domains, to serious darwinist friends. Their answers are speculative, vague and non credible (not their fault, the problem is what it is, and IMO non solvable in darwinian terms). The usual explanation if thatthose transitions happened in the past, and that for some strange reason all the intermediaries were cancelled afterwards. Now, I could admit that in some cases (although I find it a very bizarre explanation anyway), but certainly not in all cases. So, observing the existing proteome, and if we want to explain it forcibly in neo darwinian terms, the only viable hypotheses would be: 1) All basic protein domains, or most of them, were generated by pure RV. or 2) There is some strange law of nature that erases systematically all molecular intermediaries between basic protein domains, although they were functional enough to be naturally selected and expanded at the time they served as intermediate steps to the transitions. I’m not impressed by you either! Fair enough! :)gpuccio_{October 26, 2011
October
10
Oct
26
26
2011
09:23 PM
9
09
23
PM
PDT}

Thanks, I'll take a look at SCOP.DrBot_{October 26, 2011
October
10
Oct
26
26
2011
02:32 PM
2
02
32
PM
PDT}

DrBot: Please look at my post 44.1.1. If ypou go on the SCOP (Structural Classification of Proteins) site, you can access a tool called ASTRAL. By that tool you can ask for the identifiers of domains sharing less than n% identity. 10% is the lowest value you can set. You get 6311 identifiers that way. A classification based purely on fold structure, much more conservative, gives 1195 independent folds. Superfamilies and families are intermediate classifications. The evalue search gives results similar to the % of identity search setting an evalue >1 (no statistically significant homology for sure): 6680 independent groups. Obviously, not all domains have less than 10% homology. Protein families include often many domains that are quite similar. That's why I always speak of "basic protein domains", that is those groups that are really independent (1195 - 6680 according to the grouping one chooses). For instance, the paper I am referring to uses a classification based on 3464 domains, that is the munber of protein families in SCOP 1.73 (there are 3902 in SCOP 1.75, the most recent).gpuccio_{October 26, 2011
October
10
Oct
26
26
2011
01:51 PM
1
01
51
PM
PDT}

Do all domains share less than 10% identity?DrBot_{October 26, 2011
October
10
Oct
26
26
2011
12:45 PM
12
12
45
PM
PDT}

Prev 1 … 3 4 5 6 7 … 14 Next

You must be logged in to post a comment.

Leave a Reply