|
Author
|
Topic: Evolving Inventions
|
gedanken
Member
Member # 594
|
posted 11. March 2003 22:16
quote:
Nelson:
The the ID position here is scientifically accurate. You cannot remove any component from the flagellum without losing motility.
I’m sure that is correct. I was speaking about whether it was relevant.
quote:
Now as far as the claim that the further a tree branches the more likely it is that a solution will be found, I disagree.
But of course I made no such claim. What I was explaining is that different solutions to different problems will be found on different branches. In fact I commented that upon “rewinding the tape” so to speak that if evolution was run all over again that a different set of characteristics would arise. But of course we do observe organisms that do not have a flagellum, and yet are motile, so we know that different methods of giving an advantage of motility can be found. No one has demonstrated that there is a need for this particular “solution,” either, so its lack in a previous generation does not seem to imply that the organism would not survive in some niche.
Alonso, I have a question. What function whatsoever of any organism does not have an “IC Core”?
[ 11. March 2003, 22:30: Message edited by: gedanken ]
IP: Logged
|
|
Nel
Member
Member # 614
|
posted 11. March 2003 23:01
quote:
Nelson:
The the ID position here is scientifically accurate. You cannot remove any component from the flagellum without losing motility.
Ged:
I’m sure that is correct. I was speaking about whether it was relevant.
Nelson:
You can see the relevancy when you are forced by Behe's thesis to invoke co-option.
quote:
Now as far as the claim that the further a tree branches the more likely it is that a solution will be found, I disagree.
Ged:
But of course I made no such claim. What I was explaining is that different solutions to different problems will be found on different branches.In fact I commented that upon “rewinding the tape” so to speak that if evolution was run all over again that a different set of characteristics would arise. But of course we do observe organisms that do not have a flagellum, and yet are motile, so we know that different methods of giving an advantage of motility can be found. No one has demonstrated that there is a need for this particular “solution,” either, so its lack in a previous generation does not seem to imply that the organism would not survive in some niche.
Nelson:
Yes, there are different ways of getting a motile structure, but the one we currently see cries out for an explanation, and alternative methods seem to be also IC. For example, we don't see bacteria moving using a simpler, homologous motility structure. We only see the bi-directional motor. There are of course other solutions that can be arrived at but most likely, none of these will be arrived at through the trial and error process of RM&NS.
Ged:
Alonso, I have a question. What function whatsoever of any organism does not have an “IC Core”?
Nelson:
ADP synthesis (I think).
[ 11. March 2003, 23:23: Message edited by: Nelson_Alonso ]
IP: Logged
|
|
Frances
Member
Member # 169
|
posted 12. March 2003 00:04
Hi Nelson,
You have identified one of the major shortcomings of ID based on the IC argument namely that IC itself is not sufficient to eliminate Darwinian pathways and since ID does not propose its own pathways, ID seems to be not much different from 'we don't know'. So far testable or falsifiable ID hypotheses seem to be lacking, ID seems to be infered from negative evidence and arguments that use IC or CSI seem to be based on "begging the question" approaches. So far ID has failed to make for itself a convincing case.
I may also want to point out that you may while Dembski may be making some very sweeping claims, such as CSI can only increase through ID, when looking closer at the arguments it appears that much is missing. I do not blame you for being taken in by some of the arguments but I would propose that we carefully take apart the arguments and determine if there is any foundation to the claims. An example raised by you is wrt the probability calculations of the flagellum. You claims include This quote states my point quite clearly, and I'm basing this claim on the perturbation and tolerance factors of Dembski's calculation. or The logical possibility of an indirect route does not speak to the improbability of such routes. Again, it is quite possible for random chance alone to get a flagellum. But even evolutionary biologists reject this. But indirect routes smuggles in too many random chance events , as shown by irreducible complexity and Dembski's calculation, then, although it too is possible, it should be likewise rejected..
So what did Dembski actually calculate when discussing the probabilities of the flagellum? Did Dembski take into consideration any effects of selection? Nope. Did Dembski take into consideration any evolutionary pathway? Nope. Instead he focused on a purely random argument.
Several people have pointed this out:
Richard Wein
quote:
Nevertheless, since Dembski does not state clearly that he has based his calculation on a hypothesis of purely random combination, I will describe the calculation briefly in order to demonstrate that this is the case.
Perhaps Nelson can explain to us why he believes these calculations are relevant and what they pretend to show? Preferably in his own words.
The argument from absence of evidence/design is well exposed by Nelson's statement:
Showing that indirect pathways are unlikely is enough to warrant a design inference.
In order for this to be true however all pathways need to be shown to be unlikely and other than random chance, Dembski has done little to show the probabilities of such pathways
So when Nelson states You have to change the function in order for your story to work. he seems to appreciate the strawman of IC. By limiting IC to a particular function, IC arguments ignore potential evolutionary histories.
more interesting assertions which seem to be irrelevant but I would like to ask Nelson to provide some supporting evidence:
quote:
Exactly. So with this ancestral function that has absolutely nothing to do with motility,(protein secretion), natural selection would conserve it. Unfortunately, what you want is a flagellum that functions for motility, not protein secretion. To get that, you have to invoke multiple chance events.
What multiple chance events? Pure chance or chance/selection? Is Nelson using the term chance here as defined by Dembski in NFL?
A circular argument
quote:
Furthermore, that Darwinists have laid their cards on the table about how the flagellum evolved, we can , by looking at the systems in question, and whether they are irreducibly complex, ascertain their probability.
ICness is not helpful in assessing the probability or improbability of a pathway. I understand that these concepts can become confusing since they are being used in a not always very rigorous manner, and sometimes even circular.
When I stated that
Francis:
Science thus must form its conclusion s not on the possibility or promise of future evidence of intelligent design.
Nelson responded:
ID theorists don't base it's conclusions on promises of future evidence for intelligent design. I regard the bacterial flagellum as evidence for intelligent design right now. Which is why it has no evolutionary history. What I hear more often then not, and even you stated this in the front-loading thread, is that "just because we don't have an explanation for it's evolution now doesn't mean we won't in the future".
Perhaps I should have said instead of future promises, Nelson's feelings about the flagellum? But whatever you regard evidence or not does not make it so. In fact NO evidence of ID for the flagellum has been presented other than some circular argument "that's why it does not have an evolutionary history", without even investigating possible historical pathways. Now if Nelson could provide us with some POSITIVE evidence of ID for the flagellum? But given the inherent shortcomings in the ID approach, that may seem to be unlikely.
So to recap:
Specified complexity is easy to assign to almost any event whether it be a hand of poker cards, or the after the fact assignment of specifications to the flagellum. Thus not only false positives are trivially easy to generate but we do not have ways to compare ID proposals for the origins of lets say the flagellum and evolutionary explanations.
ICness as well is unable to be used as a reliable identified of ID in that it cannot eliminate direct and/or indirect routes.
So if CSI can not be used and ICness cannot be used without risks of false positives, an eliminative argument based on negative rather than positive evidence can never be succesful.
We can add to this Murray's observation that since ID cannot differentiate between front loading and intervention, combined with trivial ease of providing a specification and we see how ID will fail to provide a scientific alternative to methodological naturalism.
On the other hand I do see that ID can have a positive contribution to science in that it encourages scientists to look more closely at some interesting structures in nature. Of course even if we may never know their evolutionary history, that should not be seen as any evidence _for_ ID.
While ID arguments may initially appear to be convincing, when looking at the details it becomes clear that they suffer in many instances from lack of mathematical rigor or that they make claims which cannot logically be supported.
I would be very interested in exploring some of these concepts which are raised by Nelson such as the flagellar protein calculations and their relevance (or absence thereof) to evolutionary pathways.
Until ID can propose a positive research program rather than arguments to absence of evidence, it will allow itself to be vulnerable to scientific criticism.
Since ID so far seems to be based on negative evidence I presume that Nelson can walk us through the calculations which led him to state that
quote:
Yes, there are different ways of getting a motile structure, but the one we currently see cries out for an explanation, and alternative methods seem to be also IC. For example, we don't see bacteria moving using a simpler, homologous motility structure. We only see the bi-directional motor. There are of course other solutions that can be arrived at but most likely, none of these will be arrived at through the trial and error process of RM&NS.
But even if the process is not RM&NS, it would still be a natural pathway and would thus not be much help to the ID inference. We need some rigorous arguments here since ID has set itself the task of rigorously rejecting all relevant hypotheses.
The problem of course is that "all relevant" may include yet unknown pathways and since ID cannot allow for false positives it cannot reject the logical possibility of such a pathway which means that negative approaches will not be helpful here unless one can reject large sets of hypotheses in one grand swoop. TRIZ and IC were attempts to provide for such markers but in both cases they have been reduced to almost meaningless circular arguments that have lost most of their relevance to the issues at hand.
[ 12. March 2003, 00:47: Message edited by: Frances ]
IP: Logged
|
|
gedanken
Member
Member # 594
|
posted 12. March 2003 03:21
quote:
Ged:
Alonso, I have a question. What function whatsoever of any organism does not have an “IC Core”?
Nelson:
ADP synthesis (I think).
But explain this in detail. How can it possibly not fail to function after removing sufficient parts, first reducing to an essential subset, then removing one more?
IP: Logged
|
|
Rex Kerr
Member
Member # 632
|
posted 12. March 2003 04:07
Gedanken, if the essential subset isn't complex, it isn't irreducibly complex. Behe seems to grant that single proteins of typical length are not complex, and thus even if they are irreducible, they are not irreducibly complex.
Nelson said:
quote:
No ID critic has ever accepted this challenge.
This is because the theory predicts that doing this is frightfully unlikely on time scales that are experimentally tractable.
No ID supporter has shown a primitive replicating cell created from pure elements by intelligent designers in a lab, without any reference to modern biology.
I think both failings are completely understandable.
quote:
Now as far as the claim that the further a tree branches the more likely it is that a solution will be found, I disagree. I disagree because IC eliminates so many of these possibilities by it's very own nature. As we see it eliminates all direct pathways. Direct pathways lies in a zone where the potential for probable pathways mostly exist.
Instead, what do we have? We only have indirect pathways. All indirect pathways to date have been shown to both not exist and/or be completely unlikely.
Where has this "showing" taken place? I haven't seen it. I haven't even seen the indirect pathways seriously considered, much less considered, calculated, and rejected.
Someone (maybe Dawkins?) likened evolution to a probabilistic ratchet. This is a particularly apt analogy in this case, and serves to illustrate why indirect pathways cannot simply be waved aside. One can read Climbing Mount Improbable or whatnot for more details. For now, let's consider the mathematics in more detail.
Evolution does not act on individuals; it acts on populations. Let's say the population size is N at any given time, and we have M generations throughout history. This means that we have about N*M distinct individuals, or N*M chances for something to happen all at once.
Suppose we're wondering about a transition A->Z that has probability q of occuring in one step. We would expect to see this one about q*N*M of the time (a very small number if 1/N*M >> q). Let's call this number E(A->Z), the expected number of times we'll see A->Z.
Now suppose we identify a functional intermediate M, and find that p(A->M) = 0.01*sqrt(q) and p(M->Z) = 0.01*sqrt(q) so that the probability of going from A->M->Z is 0.0001*q. This seems to have not helped at all!
But let's look more closely. Halfway through history, we've had N*M/2 organisms. We'd expect to have gotten M with a probability of 0.01*sqrt(p)*(N*M/2) by this point. If we get M at this point, then by the end of history we expect to reach Z with a probability of 0.01*sqrt(p)*(N*M/2). We expect to see A->M->Z 0.01^2*q*(N*M/2)^2 of the time. Doing a bit of algebra gives E(A->M->Z) = 0.0001^2*N*M/4*p*M*N = E(A->Z)*N*M/40000. Conclusion: if N*M > 40000, it was much more likely to go through this intermediate than to go in one step.
If you do the math a little more carefully, realizing that you can switch over from A->M to M->Z anywhere between the beginning and end of history, and p(A->M) = r and p(M->Z) = r', then E(A->M->Z) ~= r*r'*N^2*M^2/2. If M is about half as big of a complex as Z, you'd expect r to be about sqrt(q), but since it has to be functional there's presumably an additional functionality penalty k (and likewise a penalty k' for r'). Thus E(A->M->Z) = k*k'*N*M*q*N*M/2 = k*k'*N*M/2 * E(A->Z).
If there are W such indirect pathways, you can reject them as insignificant (relative to one-step creation) if 1 > W*k*k'*N*M/2. In order to do this, you need to estimate (1) the number of indirect pathways W, (2) the functionality penalties k and k', and (3) the total number of individuals throughout history N*M. Dembski has not, to my knowledge done such a calculation. (Strictly speaking, W and k and k' are not independent, so you should really compute sum(k_i*k'_i,i=1..W).)
And this is only supposing one indirect intermediate. You then have to consider cases with 2, 3, 4, ... intermediates. These are important to consider because the (N*M) factor will go up exponentially in the number of intermediates, but the (k*k') penalty will not necessarily keep pace since presumably as intermediates become more closely spaced, the penalty will be reduced. (I've not proven it, but I believe the general formula for J intermediates will be k_1*...*k_J*(N*M)^J/(J+1)!.)
The problem with evaluating this is that we are particularly lousy at imagining intermediates. I would be extremely interested to see any progress by Dembski or anyone else at evaluating this quantity for a system that is supposed to be IC.
Also, there's another major problem with the original version of irreducible complexity. The flagellum is supposed to require 30 subunits or so. But why? Why can't flagella that we see now have the kind of partition of function as seen in the eng1/eng1b example? If we think about the bare minimum requirements for a flagellum, we need (1) a pore, (2) a motor, (3) a filament, and (4) a link from the pore to the cytoskeleton so the pore won't just spin around in the membrane. That's potentially as few as four subunits. The remaining 26 could just be embellishment--it's really hard to know. The original version of IC doesn't even seem to consider this possibility. Dembksi's revision does, to an extent, but he doesn't seem to fully consider the implications when he does the flagellum calculation.
IP: Logged
|
|
Argon
Member
Member # 276
|
posted 12. March 2003 10:53
Rex comments:
quote:
Gedanken, if the essential subset isn't complex, it isn't irreducibly complex. Behe seems to grant that single proteins of typical length are not complex, and thus even if they are irreducible, they are not irreducibly complex.
I agree that this is what Behe thinks. But I'm not sure that what he thinks is necessarily correct in this particular case. If we accept the idea of an "IC core" or of criteria based on the level of "complexity" (no matter how difficult this is to determine), then we must also accept that there are different levels of "ICness". But having different grades of ICness and consequently, different grades of evolvability is not what the concept of IC was all about -- IC is about providing binary, "yes/no" determinations for classification of systems and for questions of evolvability. The notion that a system is not IC because it is not "complex enough" is begging the question of what constitutes sufficient complexity. And that appears to be a highly subjective criterion (As we know, "system" and "function" as similarly subjective criteria at times). Further, if we can assign a level of complexity, we then must determine what portion of that complexity can be attributed to SC steps vs. steps that provided via "additive complexity".
[ 12. March 2003, 10:54: Message edited by: Argon ]
IP: Logged
|
|
gedanken
Member
Member # 594
|
posted 12. March 2003 11:04
quote:
quote:
Gedanken, if the essential subset isn't complex, it isn't irreducibly complex. Behe seems to grant that single proteins of typical length are not complex, and thus even if they are irreducible, they are not irreducibly complex.
I agree that this is what Behe thinks. But I'm not sure that what he thinks is necessarily correct in this particular case. If we accept the idea of an "IC core" or of criteria based on the level of "complexity" (no matter how difficult this is to determine), then we must also accept that there are different levels of "ICness".
Thanks Argon, your point is what I was thinking.
For example put that single protein in a petri dish. Does it “function?”
(Also everybody’s comments have deprived me of putting Alonso on the spot. I find the argument he presented to be self-contradictory. I wanted to see which side he took on parts of his argument. Alonso, please consider my question as still open, as I want to examine your version with regard to your longer response above. You may assume that I will respond as I have so far to what has been presented so far.)
Also thanks to Rex for details in the clarification. (Please note that while I accept everything Rex said, and find it very useful to my understanding, I also agree with Argon’s point in its entirety.) Rex also said:
quote:
If we think about the bare minimum requirements for a flagellum, we need (1) a pore, (2) a motor, (3) a filament, and (4) a link from the pore to the cytoskeleton so the pore won't just spin around in the membrane. That's potentially as few as four subunits. The remaining 26 could just be embellishment--it's really hard to know. The original version of IC doesn't even seem to consider this possibility. Dembksi's revision does, to an extent, but he doesn't seem to fully consider the implications when he does the flagellum calculation.
This relates back to my point about probability and branching of trees.
Let’s consider what (for example) in this forms a “specification” and not a “fabrication”. Even in this simple of a description, we have described a combination of a “motor” and a “filament” and a “pore”. Now these are not terms that would normally come up in human design together, rather the combination only would tend to occur because of our prior knowledge of the bacterial flagellum. (Thus such a description would be painting a bullseye on the result after the fact -- a “fabrication” in Dembski’s terminology.) Even pairwise these terms are not part of normal human communication as to form a specification in the normal sense, so in triple parts even less so.
What does appear to be part of human “specification” is the outboard motor, with some sort of propeller. (The essence is energy conversion and rotation in the mechanical form of the converted energy.) But this “propeller” must be taken in its most general form to be part of normal human “specifications” outside of observing the specific biology after the fact. So the probability analysis should ask what is the probability of arriving at a combination of a rotating motor and a form in general that can be mapped to a propeller -- and all such possible biological examples thereof.
(But then of course we have many examples of a non-rotating motor and propeller -- all the fish tails and fins. Maybe the human designed the propeller from observing of such non-rotating motor-propeller systems, and simply added the “rotation” as a recognition that the motor could be put in a rotating form. Thus the TRIZ contradiction was back at the invention of the wheel, wherein the axle was first considered. Oops, the ancient Egyptians used rolling logs, looks a lot like an axle. Humm…)
In any case to get back to the topic of a probability calculation -- when we combine these disparate elements, “pore,” “motor,” “rotation as the form of the energy transfer,” “filament,” we have different dimensions that could occur independently in some sort of structures. So there might be many different “solutions” to varying problems involving these elements. So there is no necessity that any organism have these connected elements to solve any problem. Every problem solved by any single or combination of these elements have been solved in nature in many different ways.
The only contradictions are in producing the particular solution observed based on a viewpoint of something prior, but wherein we apparently at the moment don’t know the pathway. This gets to the branching of the tree. What Rex showed in probability calculations is (as I presume from what was shown) a Bayesean posterior relationship of pathway to a given particular organism (looking backward). The part I wanted to focus on is the density of possible forms that could occur working forward.
Now given any particular position in the branching tree, selection may favor some pathways in greater probability over others. Thus selection forms a strong constraint on possible pathways. But remember that this is in the context of the evolving environment. (There were discussions about NFL theorems not considering the parallel effects of other organisms evolving, for example.) The pathways still depend to a considerable degree upon the random descent with modification operations -- and thus the branching tree of what actually evolves could be considered to be quite variable if the “tape” were rewound and “played over” (evolutionary trials rerun).
Once again, there has as far as I can tell no accounting for the general case of probability of a rotating motor in biology, These are the essential “specifications” that might be of interest -- the combinations with unrelated issues like “filament” and “pore” are not particularly interesting as combinations and could have easily occurred in different patterns. The individual patterns that actually occur are indeed of very low probability -- in fact each set of unrelated or partially unrelated aspects like “motor,” “filament,” “pore” will have very low probability of occurring on the branching tree probability calculation if one starts far enough back on the tree. And if one starts sufficiently near to the organism or combination observed, the probability would undoubtadly be much higher. This is the problem of claims of “low probability,” it is not specified in terms of starting basis. As soon as the starting basis is considered to be far enough back in the tree, the probability is necessarily low of the particular complex combination occurring.
[ 12. March 2003, 12:13: Message edited by: gedanken ]
IP: Logged
|
|
ASCSCommanding
Member
Member # 682
|
posted 12. March 2003 13:06
Nelson,
I’d like to comment on the following from an earlier post of yours because I find the ID take on “direct” and “indirect” paths to be very confusing.
quote:
I disagree because IC eliminates so many of these possibilities by its very own nature. As we see it eliminates all direct pathways. Direct pathways lies in a zone where the potential for probable pathways mostly exist.
My understanding of the ID position on these pathways comes largely from DBB and some more recent essays as well on the comments on this board. If there are other sources that clear up these issues, please direct me.
From DBB it seems that for some proposed transition from state A to F there is a direct path through B, C, D, and E. There may be other possible state A1, A2, A3, …, B1, B2, B3, … etc. that the system could go through, so A to A1 to B1 to C and on is possible, but these are presumably less likely than the direct path A, B, C, D, E, F. It further more seems that the direct path is A, B, C, D, E, F and any other path passing through any of the other states is an indirect path. My first observation is that even if the probability of going from A to B directly is 80% of the probability of getting from A to B at all, and likewise for all the other transitions, most likely the system will transition via an indirect route, even though the direct route is more probable than any indirect route. The probability of staying on the direct route must be vastly more likely than the total probability over all indirect routes at each step, or else the expectation is that some indirect route will be taken. Note also that this applies to all sorts of systems; including the development of reducibly complex systems that all agree can be produced by Darwinian mechanisms. Indeed, in any system where the transitions are in part either random or can be modeled as random.
I would expect that in trying to determine the likelihood of ending in final state F if starting in initial state A is the sum of the probabilities over all possible paths from A to F, not just a comparison of the probability of some direct path vs. the probability of some particular indirect path. To state it another way, if you divide possible paths into either direct or indirect, then even if the direct path is much more likely than any indirect path, if the probability over all indirect paths is high, one would expect some unlikely, indirect path to be taken.
I have not seen an ID argument with regard to the total probability over all indirect paths. What I’ve seen are arguments to the effect that the direct path is proportionally more likely that whatever indirect path is proposed. This argument is irrelevant. Furthermore, given that the direct path has zero probability, it doesn’t make sense to claim that it has a higher probability than the indirect path. The ID argument seems to require that the probability of the direct path be both fairly high and zero.
I think the basic problem is that the ID “direct” pathway is a misnomer and has little to do with how biological systems actually change.
IP: Logged
|
|
gedanken
Member
Member # 594
|
posted 12. March 2003 13:21
quote:
I have not seen an ID argument with regard to the total probability over all indirect paths.
This is exactly correct.
But my point above is also that the starting point determins the probability -- it is not independent. In above, point "A" is the starting point over which all pathway probabilities must be calculated (if that were even remotely possible). Choose an intermediate point, and one gets a different answer.
Choose a point prior to "A" and one gets an even lower probability. Go back far enough, and one can get a probability that is below the UPB (10^-150). This is true for all final points. What is the basis for starting at point "A"? This is arbitrary.
What is important (and well explained by ASCSCommanding) is that different pathways must be accounted for, and to that I add that the densitity of "destinations" that will be interesting organisms also needs to be considered in the forward direction from any point such as "A".
IP: Logged
|
|
ASCSCommanding
Member
Member # 682
|
posted 12. March 2003 16:34
gedanken,
I agree very much with what you say.
As you point out, the probabilites of getting to state F from the starting state depend upon where the starting state is in time. Additionally, one must know the starting state in some detail to have any chance of calculating the probabilities.
For both these reasons, calls to calculate the probability of something like the flagellum are impossible to meet, not because the theory of evolution has problems, but because required information is not available. The relevant probabilities cannot be calculated.
IP: Logged
|
|
Nel
Member
Member # 614
|
posted 12. March 2003 17:23
Ged:
But explain this in detail. How can it possibly not fail to function after removing sufficient parts, first reducing to an essential subset, then removing one more?
Nelson:
I don't have time to get into the details of the system I mentioned but you can see how this is possible with some functions that we already discussed, for example, the membranes that Dawkins talked about, the function can be reduced to one membrane. Here you see that the addition of additional membranes just makes the function more efficient. The 9+2 pattern of MTs in cilia are not IC, it's just more efficient than the 3+0 pattern.
IP: Logged
|
|
Nel
Member
Member # 614
|
posted 12. March 2003 18:19
Rex:
Gedanken, if the essential subset isn't complex, it isn't irreducibly complex. Behe seems to grant that single proteins of typical length are not complex, and thus even if they are irreducible, they are not irreducibly complex.
Nelson:
I'm not sure if Behe thinks that single proteins are not complex. However, I agree with you that single proteins may not be irreducibly complex. Especially in the case where, say a 3-protein complex, where the addition of the second and third protein that simply makes the function of the single protein more efficient, where removing that second and third protein simply makes the function less efficient, is not IC.
Rex:
Nelson said:
quote:
No ID critic has ever accepted this challenge.
This is because the theory predicts that doing this is frightfully unlikely on time scales that are experimentally tractable.
Nelson:
I don't agree that the experiment is not tractable due to timescales. An experiment involving thousands of generations of selective pressure for motility, involving something like this:
Papadopoulos D et. al. Genomic evolution during a 10,000-generation experiment with bacteria.
Proc Natl Acad Sci U S A. 1999 Mar 30;96(7):3807-12.
and we can increase the mutation rate. This should (theoretically) produce a flagellum-like IC motile structure that is at least beneficial, if we start with a structure that, as Darwinists say,the flagellum was co-opted from -- an ion channel. I don't see anything particularly difficult about this experiment. Except that the stakes are pretty high here for Darwinian accounts of evolution. Even thinking about how this would work boggles my mind.
Rex:
No ID supporter has shown a primitive replicating cell created from pure elements by intelligent designers in a lab, without any reference to modern biology.
Nelson:
I think that studies in AL, nanotechnology, and the fact that we invented the motor before we found one attached to bacteria is effectively an experiment that supports the ID inference here. However, we've never seen RM&NS build a motor from a export device.
Nelson:
Now as far as the claim that the further a tree branches the more likely it is that a solution will be found, I disagree. I disagree because IC eliminates so many of these possibilities by it's very own nature. As we see it eliminates all direct pathways. Direct pathways lies in a zone where the potential for probable pathways mostly exist.
Instead, what do we have? We only have indirect pathways. All indirect pathways to date have been shown to both not exist and/or be completely unlikely.
Rex:
Where has this "showing" taken place? I haven't seen it. I haven't even seen the indirect pathways seriously considered, much less considered, calculated, and rejected.
Nelson:
As I have repeatedly said, the indirect pathways were seriously considered, in my opinion, here:
http://www.idthink.net/biot/flag1/index.html
And taking the T3SS into account, the calculation is still way below the probability bound.
Rex:
Someone (maybe Dawkins?) likened evolution to a probabilistic ratchet. This is a particularly apt analogy in this case, and serves to illustrate why indirect pathways cannot simply be waved aside. One can read Climbing Mount Improbable or whatnot for more details. For now, let's consider the mathematics in more detail.
Nelson:
Dawkins showed in Climbing Mount Improbable exactly why IC systems are inaccessable to Darwinian evolution --- you don't get a more efficient flagellum by adding parts to a single component. Any number below 20 is not a less efficient motility structure, it is a non-functional motility structure. But of course that gets us back to the indirect pathways argument.
Also, consider some of the points raised by Mike in his essay, the irreducible complexity of the folding of P Pilus, not to mention the sophisticated mechanisms, donor strand exchange and donor strand complementation. The pilus itself is made up of 5 parts, PapK PapA,PapE,PapK, and PapG.
quote:
As if having your supporting evidence shown to be irrelevant was not bad enough, there are more problems. For example, let's imagine that with enough luck, somehow a P-pilus-like materializes. After all, such pili are the most common. And therein lies the problem, because while the P-pilus makes a great attachment organelle, it's probably a dead-end if one wants to evolve a flagellum. For one reason, the P-pilus has not been observed to secrete proteins. This could be because the channel is so small . Or it might have something to do with the energetics of the system, as P pili formation is independent of cellular energy. It's not surprising that the P-pilus looks very different from the bacterial flagellum (or even things like type IV pili).
Finally, there is yet another fact that suggests flagella did not arise in the manner that the EFM proposes. Whether we're talking about simple type I pili or more complex type IV pili, what they all share in common is being built from the bottom-up. The flagellar filament, in stark contrast, is built from the top down. And the manner in which this done is yet another amazing story in microbiology. How amazing? Robert Macnab is an expert on the flagellum and has been working on them his whole life. As such, you might expect him to be used to the complexity and sophistication of the flagellum.
http://www.idthink.net/biot/flag2/index.html
and
quote:
The latest technical discoveries in flagella fascinate biologists such as Robert Macnab, a professor of molecular biophysics and biochemistry at Yale University who also studies flagella. He marvels at how organisms as simple as bacteria have evolved such complex methods to develop propelling features, especially since motility in bacteria is not directly necessary for survival, like DNA replication or protein synthesis. "We think it would not be possible for the system to work with any significantly lower complexity."
Rex:
The problem with evaluating this is that we are particularly lousy at imagining intermediates. I would be extremely interested to see any progress by Dembski or anyone else at evaluating this quantity for a system that is supposed to be IC.
Nelson:
Not to belittle your good work here, but I don't think the existence of possible intermediate steps or how good we are at imagining intermediate steps is what makes these indirect routes so unlikely, what makes them unlikely is the number of pure chance, unselectable steps along the way to the flagellum.
You can think of the filament, the ion channel, and the motor, as whole multi-part systems. This whole-sale recruitment of complex systems doesn't seem like the likely scenario that you paint to be. Also consider another point made by Mike Gene's essay:
quote:
The rotor must have access to the proton/ion through the ion channel. What type of fortuitous change is going to pry open this ion channel and then glom it onto the proto-rotor?
and
quote:
Of all the ways to mutate an ion channel, the number of ways that would result in its interacting with the base of some filament is surely in the distinct minority. And of all the ways to mutate an ion channel that gloms onto a filament, the number of ways to mutate it such that rotation does not occur is probably much higher than the number of ways to elicit some rotation.
Thus:
quote:
This [mutation] allows some ion channel to glom onto the base of a filament and open its channel and expose the ion flow to the proto-rotor in such a way that a set of electrostatic interactions just happen to form and elicit significant rotation. Suffice it to say that such an improbable mutation has never been observed in nature or the lab.
Next he points out that the motor complex itself is IC, requiring fliG, motA, and motB. Thus to get around this you would need to propose more than an abstract existence of possible intermediates or imaginary scenarios that gloss over these kind of details.
Rex:
Also, there's another major problem with the original version of irreducible complexity. The flagellum is supposed to require 30 subunits or so. But why? Why can't flagella that we see now have the kind of partition of function as seen in the eng1/eng1b example? If we think about the bare minimum requirements for a flagellum, we need (1) a pore, (2) a motor, (3) a filament, and (4) a link from the pore to the cytoskeleton so the pore won't just spin around in the membrane. That's potentially as few as four subunits. The remaining 26 could just be embellishment--it's really hard to know. The original version of IC doesn't even seem to consider this possibility. Dembksi's revision does, to an extent, but he doesn't seem to fully consider the implications when he does the flagellum calculation.
Nelson:
I don't see how grouping all the subfunctions of the filament or the motor to one part really reduces the complexity of the flagellum. This mistake is most expemplifed by the reducible mousetrap example (the two-part mousetrap). You would literally have to reshape and re-engineer the seperate parts of the motor and the seperate parts of the filament and the seperate parts of the export machine in order to group them all up into seperate parts. But after you have done so, so what?
quote:
In the real world of biology the staples, bends, and so forth would be features of molecules, of proteins in particular. If two proteins don't bind each other in the correct orientation (aren't stapled right), if they aren't placed in the right positions, if their new activity isn't regulated correctly, if many details aren't exactly correct, then the putative Darwinian pathway is blocked. Now, it's hard, almost impossible, for persons without the appropriate science background to tell where such difficulties would occur in Darwinian scenarios for blood clotting or ciliary function or other biological systems.
http://www.discovery.org/viewDB/index.php3?program=CRSC%20Responses&command=view&id=446
Such a scenario begs the question of where the irreducible complexity of each subfunction comes from in the first place.
[ 12. March 2003, 19:57: Message edited by: Nelson_Alonso ]
IP: Logged
|
|
GP
Member
Member # 570
|
posted 12. March 2003 18:21
Greetings Nelson Alonso,
I was wondering if you could provide references for your J. Thomas citations, as is typically customary to give credit to the work being reproduced. And I also like some background information on her research pursuits.
Thanks in advance,
GP
PS: I accidentally posted in the wrong thread. But luckily it seems that you are active in this one too. With the moderator's indulgence, I will post a copy on the relevant thread, and he/she may erase a copy.
[ 12. March 2003, 18:22: Message edited by: GP ]
IP: Logged
|
|
Nel
Member
Member # 614
|
posted 12. March 2003 19:09
I can't find anything in Francis's post that I havn't already addressed. I'll leave replies to Francis's post as an excercise for the reader. So I'll respond to ASCS and call it a night.
By the way fellas, we're getting into IC here pretty deeply, does this call for a new thread? I don't see how this has much to do with Bracht's paper anymore. I'd start it but I have no idea what to call it or if the moderater would even allow it.
Nelson:
I disagree because IC eliminates so many of these possibilities by its very own nature. As we see it eliminates all direct pathways. Direct pathways lies in a zone where the potential for probable pathways mostly exist.
ASCS:
I would expect that in trying to determine the likelihood of ending in final state F if starting in initial state A is the sum of the probabilities over all possible paths from A to F, not just a comparison of the probability of some direct path vs. the probability of some particular indirect path. To state it another way, if you divide possible paths into either direct or indirect, then even if the direct path is much more likely than any indirect path, if the probability over all indirect paths is high, one would expect some unlikely, indirect path to be taken.
Nelson:
This is why we have to take into account how natural selection works instead of getting into these abstract possibilities. For example, if A and B are conserved by natural selection, natural selection would be selecting for that particular function. The relationship between that system and natural selection is not random. To quote Dawkins again, what we have with IC is an abrupt precipice where any number below 20 is not inefficient, it does not work. Each step in a direct pathway confers a selective advantage all the way to the top of the mountain for the same function. Even if you were correct, that indirect routes would most likely be taken by a natural process to any particular function to does mean that the indirect route itself is more likely then a direct route, where you are just selecting parts that make the original function more efficient.
ASCS:
I have not seen an ID argument with regard to the total probability over all indirect paths. What I’ve seen are arguments to the effect that the direct path is proportionally more likely that whatever indirect path is proposed. This argument is irrelevant. Furthermore, given that the direct path has zero probability, it doesn’t make sense to claim that it has a higher probability than the indirect path. The ID argument seems to require that the probability of the direct path be both fairly high and zero.
Nelson:
I don't see this argument as irrelevant. For example, an indirect path looks something like this. A does some job which natural selection has locked on and conserved. B comes along and not only starts working interactively with A, but changes the function of the two part system to something other than what A was originally being pruned for by natural selection. The new function is now "irreducible", removing B changes the function of AB we cannot preserve it. Now skipping over the details of how one part can start interacting with and become well-matched to another, all the way spontaneously changing the original function that natural selection locked onto, is unimportant, whats important here is for a 20 part system, you have to invoke these pure chance events multiple times. C comes along and starts working with ABC maybe making it more efficient but then another part, D comes along and starts interacting with ABC all the while changing the function of ABC. Then E comes along maybe making the ABCD function more efficient, so natural selection adds on ABCDEFG, but then antother part H comes along and starts acting interactively and becomes well-matched to ABCDEFG while spontaneously changing the function of the ABCDEFG system and on and on until you get the IC F function that requires 20 parts.
ASCS:
I think the basic problem is that the ID “direct” pathway is a misnomer and has little to do with how biological systems actually change.
Nelson:
The term "direct path" are in the peer-reviewed literature and with respect to IC to boot. The most common pathways are Serial Direct Evolution, Parallel direct Evolution. These are the two pathways that Darwin inferred from looking at finch beaks, and antibiotic resistance to drugs. A gene duplication that causes a group of genes to produce more estersase is making the original function more efficient and therefore more able to resist large portions of pesticide and therefore selectively advantageous.
[ 12. March 2003, 19:39: Message edited by: Nelson_Alonso ]
IP: Logged
|
|
gedanken
Member
Member # 594
|
posted 12. March 2003 20:03
quote:
Ged:
But explain this in detail. How can it possibly not fail to function after removing sufficient parts, first reducing to an essential subset, then removing one more?
Nelson:
I don't have time to get into the details of the system I mentioned but you can see how this is possible with some functions that we already discussed, for example, the membranes that Dawkins talked about, the function can be reduced to one membrane. Here you see that the addition of additional membranes just makes the function more efficient. The 9+2 pattern of MTs in cilia are not IC, it's just more efficient than the 3+0 pattern.
But this simply confirms what I already thought. The basic claim of IC is that if one can figure out the potential pathway, then it must not have been IC after all.
Thus IC is based on gaps in scientific knowledge. As soon as the gap is filled, it is inherently no longer IC. Thus unfalsifiable.
IP: Logged
|
|
|
Printer-friendly view of this topic
