|
Author
|
Topic: A Gödelian Argument Against Darwinism
|
Noel Rude
Member
Member # 516
|
posted 11. April 2003 15:56
How does ID's supposed "argument from ignorance" differ from those Darwinian "promissory notes"? Well, the latter issue from the philosophical disposition that would rule out design a priori, the former is the latter's characterization of design raised as an empirical question. Darwinian "promissory notes" are founded on the premise that design is illegal, the cry of "argument from ignorance" seemingly wants to squelch any challenge to that premise.
How to settle the issue? In our definition of "Science"? Or within empirical science? Live and let live I say. Why squelch ID? I thought Daniel Dennett said the really dangerous idea lurks somewhere else.
As for this Gödelian Argument -- hope y'all don't mind a voice from the choir -- if the Platonic world (speaking here to physicists [and Michael Denton?] who tend to be mathematical Platonists) is rife with discontinuities, why not this world? It's a great analogy, if nothing else. And hey! this is brainstorms.
As Prof. Dembski eloquently reminds us elsewhere in his books, we can imagine but three types of explanation: chance (described with statistics), necessity (described with algorithms), and design (which defies both statistics and algorithms). If all three modes represent sides of reality that are elemental (i.e., one cannot be derived from another), then there will indeed be discontinuities in the natural world -- the discontinuities in the Platonic realm will indeed be mirrored in the physical world. The divide between physical law and quantum uncertainty troubled Einstein ("Does God play dice ...?"), and the breach between these and design bothers the Darwinists ("argument from ignorance!").
Mechanistic reductionism will not stand if elementarity obtains in all three realms of explanation.
Physical law accounts for the stability of the world in which agents live, and perhaps the uncertainty of the quantum realm is where mind inserts its authority over matter. We know the equilibrium of history is punctuated by the will of men, and if man's decisions are ultimately his and not reducible to blind forces, then history -- military, intellectual, technological, biological, cosmic -- is where the Gödelian Argument plays out in full.
[ 11. April 2003, 16:11: Message edited by: Noel Rude ]
IP: Logged
|
|
Alix Nenuphar
Member
Member # 686
|
posted 11. April 2003 17:01
Noel Rude:
You appear to be laboring under a misapprehension shared by a number of ID advocates; a misapprehension unsupported by rigorous evidence. Science does not rule out ID a priori, it rules out ID ex post facto. The parsimonious aspects of science (not least of which are grant money, laboratory time, availability of grad students, etc.) preclude research on low-probablitity explanations - which is all that my colleagues and I believe ID to be.
Keep in mind that the quality of the tools available to investigate a given domain also play into the question of how often they are used. The only tools currently available for investigating the design intuition appear to be variations on the probability argument; when they become more rigorous, some researchers might become more interested in using them. At the moment, they are of little value.
I hope you attend to the wording I have chosen: no one is ruling out ID - we witness ID taking place all around us. What scientists are requiring is better evidence for presuming ID by unknown designers in difficult to research situations.
For example, this,
quote:
Darwinian "promissory notes" are founded on the premise that design is illegal, the cry of "argument from ignorance" seemingly wants to squelch any challenge to that premise.
appears to be an emotional and hasty generalisation. What is it based on?
Certainly I have no problem entertaining a Gödelian Argument in brainstorms: forums like this are excellent for stimulating the brain (may I respectfully call attention to RBH's MDT hypothesis?)
The Gödelian Argument is also interesting in that no scientist is actually suggesting that evolution would ever need to be able to connect any two points in a solution space; only that there exists a set of connected solutions. Indeed, it would appear far more suspicious if infinite connections appeared!
I appreciate your calling to mind Dr. Dembski's remarks in this context; they remind me that the great 'trilemma' of chance, regularity, and design is as false as C. S. Lewis' famous counterpart. Why couldn't a designer choose algorithm or chance to create? Since we have no hypothesis about the designers' motives, chance and algorithm might be as effective for their purposes as concrete intervention.
Mechanistic reductionism will not stand if elementarity obtains in all three realms of explanation.
I apologise to the moderators if this post does not appear entirely constructive; but I feel that brainstorms is all the more productive when certain recurring classes of unsupported assumptions need not be re-addressed.
[ 11. April 2003, 17:08: Message edited by: Alix Nenuphar ]
IP: Logged
|
|
gedanken
Member
Member # 594
|
posted 12. April 2003 03:05
Going back to an earlier post of Alonso’s, first quoting myself:
quote:
With respect to your post on "Logical Underpinnings" you make this fatal error:
quote:
require “sweeping the field clean” all alternate explanations of the supposedly “IC” structure.
and
quote:
By that I mean all alternative examples that have reasonable expectation of having happened, for example by proceeding with “selected” steps or other physically realizable mechanisms that are or will be observed as regularly occurring in nature.
Notice that the bolded section is the ultimate in argument from ignorance. As it has been shown in this thread and many others, it is quite easy, with complex systems, to show unselectable steps in a Darwinian pathway. For example, from the C ring to the rod there are many unselectable steps that have no selectable function for natural selection to work with. The analogy with Godel is consistent with this conclusion. Although some of Biological systems can be explained by simple logic, as the complexity of the system rises, it becomes less a likely that Darwinian evolution can explain it.
As for the point about having to sweep the field clean of all alternatives, first of all, theories are not about certainty. They are about research and finding data consistent with the working hypothesis. Thus, this is not a show-stopper even if someday new data will turn up that will suggest a pathway which ID biologists will have to consider and see if it is a likely pathway.
Secondly, the mere possibility of imaginary future pathways is irrelevant. You keep taking Dembski's "sweep the field clean" comment out of context. Here is Dembski's discussion on that:
quote:
Science must form its conclusions on the basis of available evidence, not on the possibility or promise of future evidence. This means that eliminative inductions need to be local inductions, based on detailed testable models and hypotheses that are currently available.
Invisible, untestable, imaginary models are garbage in science.
Sorry for long quote, but this is back on a previous page and I wish to respond to many sentences distributed throughout the passage.
“…it is quite easy, with complex systems, to show unselectable steps in a Darwinian pathway.” But that is not relevant, as the point is not that there are “unselectable steps” in a given pathway, but that the number of possible pathways is very large. Pathways include a great deal of variation in how they become considered “selectable”, based on a great deal of complexity in both the environment as well as the specific complexity of the organism’s structural issues. Alonso keeps claiming that the availability of a result of analyzing a particular pathway is equivalent to a knowledge of the completeness of the list of such pathways -- it is not. I leave the issue of complexity in specific biological examples (“C ring to the rod”) to the biologists to discuss.
Now as to the issue of “sweeping the field clean” and basing hypothesis on currently available information, I think that reading the quoted segment above is very important. (That’s why I wanted to quote it entirely from previous page.) Alonso’s post is a reflection of just what Dr. Dembski wants us to read, in my opinion. But Dembski’s misses the point that scientists may know explicitly that all the pathways have not been examined without having examined them explicitly. So “currently available” information comes in two forms, completeness and results of what has been analyzed. It is the completeness issue that is at the hart of the IC argument, and it is the completeness issue that Dembski’s passage suggests we ignore. It suggests we ignore the completeness issue because it suggests that we examine the “hypothesis” that are currently available for that judgment, rather than examining the meta-knowledge of the completeness of the list of such hypothesis. This is a misdirection -- taking our focus off of what is important. Alonso’s post demonstrates just exactly how this misdirection affects readers who want a positive result for ID.
The term “imaginary future pathways” is very odd. Dembski’s methodology is that the field needs to be “swept clean” of possible pathways in order for the “explanatory filter” to be applicable. Now the discussion of how a given supposedly “IC” structure arose is a discussion of possible actual historical sequences that happened in the past. It is not a discussion about “future” pathways, nor is it a discussion of “imaginary” pathways. But of course the problem is that in order to figure out if a particular pathway was taken, that pathway needs to be described. That is where imagination comes in, in figuring out what potential actual historical pathways might have been. That can be done in part when direct evidence points the way. Or it can be done by figuring out what sort of evidence to look for.
Now ID proponents often wish to examine the construction of structures by making assumptions about the probability of those structures coming into existence. To do so, they pose what they imagine as possible mechanisms for that assembly. If there are multiple possible pathways in consideration that arrive at the same point, then the probability of arriving at the point is a summation of probabilities of fanning into that point over different paths. Of course the historical events could (in most cases) only have taken one of the potential pathways being described by a probability. But for the “explanatory filter” to be applicable, one must find that sum (considering all “probabilistic resources”) to be less than 10^-150. This is Dr. Dembski’s claim, and we are discussing the logical underpinnings of his “theory”! Now I know that actually finding such probabilities may be almost impossible -- and if that is the case then the explanatory filter may not be useful and thus there is no inference of design according to Dr. Dembski’s methodology. But presumably to infer “design” of the “flagellum” for example, one must find the conditions met for the explanatory filter -- in Dr. Dembski’s methodology.
This is where the issue of “local inductions, based on detailed testable models and hypotheses that are currently available.” becomes so vitally important. This is a fundamental portion of the argument in “Logical Underpinnings”.
There is a very significant difference between these two cases: A) There are no known possible pathways for the supposedly IC structure (e.g. flagellum) to have developed., and B) There are a multiplicity of possible pathways for the supposedly IC structure to have developed. Now both case A and B are cases in which the pathway is not known! I do not know how to make this point more strongly, except to bold it and repeat it. Science does not have an answer how the flagellum came about, not because it has no particular pathway in consideration, but because it has too many pathways in consideration! The available evidence does not distinguish between those pathways -- a point that is very relevant to the issue of “untestable models” that Alonso brings up. The models will not be claimed as differentiated by science as long as there is no particular evidence that distinguishes them. But this does not make the various models “imaginary” in the sense of being impossible. Rather it emphasizes the scientific caution at making claims that cannot be distinguished between until sufficient evidence is gathered -- precisely Alonso’s point about testability.
But there is a tremendous difference between cases A and B in Dembski’s explanatory filter. Case B is a case of multiple “probabilistic resources” that must be accumulated! And that is according to description found in “Logical underpinnings” paper! These probabilistic resources are not imaginary, unless Dembski’s methodology as a whole is to be declared “imaginary”.
But there is an even more important aspect of this and Gödelian arguments. The complexity of the possible pathways (in case “B” above) is very high. The result of Dembski’s explanatory filter is supersensitive to the presence of any new “imagined” probabilistic resource. Add one single probabilistic resource of probability even as low as 1 in 10^100 (an exceedingly low probability) and the result of the explanatory filter flips completely from “design” to “chance” inference claim.
Let’s compare this to a question of differentiating two possible pathways to development of the flagellum. Say we have two speculative pathways X and Y. We can examine if there is physical evidence for their actually being the pathway, and any small evidence helps the particular case X or Y accordingly. But now let’s assume that there was some fairly significant evidence that distinguishes Y over X developed. Now how sensitive is that resulting differentiation to coming up with a new pathway Z which is considered to only have probability of 10^-100 chance of having happened? That result would be virtually untouched by the consideration of that new “probabilistic resource” in pathway Z of such a low value. This is why Dembski’s “explanatory filter” is supersensitive to new information. Even considering a pathway that has not previously been thought of could easily produce a possibility of 10^-100 chance it occurred -- thus completely changing the inference from “design” to “chance”.
This is what distinguishes inferences in the “explanatory filter” from normal scientific inferences. Normally such inferences are not supersensitive to a possible explanation that previously was not thought of that is so unlikely as 10^-100, for example. In fact if you use Dembski’s UPB of 10-150 as the cutoff, then any previously unthought of pathway that adds a probabilistic resource of 10^-150 or more completely reverses the result of the “explanatory filter” method.
A point that is very relevant to my “Sequence of tests for IC” thread is that to examine the probability of all the potential pathways to the given structure, one adds pathways as they are described. There is no way to know when all pathways have been found and described in terms of “probabilistic resources”, one simply looks for more pathways. This is very distinct from having some logical evidence that all pathways have been found and examined in some categorical manner. No such exhaustive categorical evaluation of all the possible pathways to the flagellum has ever been carried out -- nor can it. Dembski explicitly suggests ignoring that problem and moving ahead with the EF inference without that analysis. This is why in my “Sequence” thread, I consider such procedures to be “falsifications” but not “tests” for a given definition. They may provide individual falsification that a given definition has been met, but no verification that a definition has been met.
Also Alonso said
quote:
You could have replied telling Frances to focus in this way. But you didn't.
Once again, my initial questions were strictly to determine the intended focus of this thread. It appears at this point that the entirety of “logical underpinnings” is open for discussion, because all of it is relevant to the issues. My point was not to complain about any post’s “focus,” and asking for the thread’s focus is not telling anyone how they should focus. Once again, I did find that backing my point of the wideness of the possible foci with argument was necessary. And Dembski’s opening post did contain a substantial portion of argument against Darwinian evolution in that its major focus was the difficulty of showing completeness of Darwinian pathways. While the link to “Logical underpinnings” goes further, the major points of the opening post were arguments against completeness of Darwinian pathways. To that I think Alix Nenuphar’s point says the most important point in responding:
quote:
The Gödelian Argument is also interesting in that no scientist is actually suggesting that evolution would ever need to be able to connect any two points in a solution space; only that there exists a set of connected solutions. Indeed, it would appear far more suspicious if infinite connections appeared!
Dembski’s “explanatory filter” depends on the probabilistic resources producing a probability of the event below 10^-150 in summation over all such resources. Finding all the resources is a task with a Gödelian component, since it is very difficult to guarantee that all such pathways have been found. However a scientific distinction between two such pathways based on evidence does not have such a Gödelian problem. Dembski’s focus is misplaced in his opening paragraph. Science makes no claim to know all the answers. It only claims to have developed significant explanations for a great many cases.
Dembski’s Gödelian argument is very interesting, it is simply applied backwards. It is the EF that has a logical argument that all probabilistic resources be considered. Drwinian arguments are based on positive evidence of connecting pathways in a great many cases. Only Dembski’s EF requires that all pathways be considered -- only Dembski’s EF suffers from the Gödelian problem.
[ 12. April 2003, 12:18: Message edited by: gedanken ]
IP: Logged
|
|
Nel
Member
Member # 614
|
posted 12. April 2003 14:37
Without the assumption that everything evolved a priori, we just don't have the evidence yet, ID follows naturally from the evidence.
Consider again the bacterial flagellum and the F-ATP synthase. Scientists have stated:
quote:
"More so than other motors, the flagellum resembles a machine designed by a human"
David J. DeRosier, Cell 93, 17 (1998).
and
quote:
"With parts that resemble pistons and a drive shaft, the enzyme F1-ATPase looks suspiciously like a tiny engine. Indeed, a new study demonstrates that's exactly what it is."
Science News vol 151, p173
and
quote:
"Some enzyme complexes function literally as machines, and come equipped with springs, levers, and even rotary joints"
Nature 386:217 (1997)
It looks designed, the fact that the function of these machines are beyond the individual components, etc etc, therefore had to be built together simultaneously. This evidence suggests that these systems were intelligently designed.
That we have no evolutionary history for systems like acquired immunity, photosynthesis, ATP synthase, bacterial flagella, cilia, or even various cell-types themselves, points to intelligent design of these systems. A Darwinist can come along and say, well we don't have evidence of an evolutionary history yet , we don't know how to reduce it yet , ad nauseum, and thus accuses the IDist of having an argument from ignorance.
By the way, just to be clear, Dembski has stated:
quote:
Specified complexity confirms design regardless whether the designer responsible for it is embodied or unembodied.
Ged writes:
quote:
I also note that no one has commented on my post above wherein I suggest:
One need only look at my last response to Ged to know that this is completely false. I did comment on it.
I noted a possible explanation for Dembski not answering your questions is that you misinterpret his "sweeping the field clean" comment to mean invisible, untestable, future explanations rather than what Dembski is actually pointing to:
quote:
Science must form its conclusions on the basis of available evidence, not on the possibility or promise of future evidence. This means that eliminative inductions need to be local inductions, based on detailed testable models and hypotheses that are currently available.
[ 12. April 2003, 15:25: Message edited by: Nelson_Alonso ]
IP: Logged
|
|
Nel
Member
Member # 614
|
posted 12. April 2003 15:02
Ged,
Ged's second reply to me is quite long, but his points are few in number (and sometimes repeated). The assertion that the number of possible pathways is "large" is not supported by the evidence. In fact, the number of possible pathways is constrained by IC, further constrained by the particular components, further constrained by the amount of unselectable steps, and further constrained by the ability to actually test these models. Unselectable steps are not irrelevant, without unselectable steps the Darwinian mechanism becomes moot.
Ged states:
quote:
It is the completeness issue that is at the hart of the IC argument, and it is the completeness issue that Dembski’s passage suggests we ignore.
Actually Dembski wants just the opposite. He wants us to completely test the model , or suggested evolutionary pathway that we actually have evidence for. To suggest that some invisible, untestable pathway exists is an argument from ignorance. For example, with the flagellum, one such pathway is being throughoughly investigated and is being found wanting (unselectable steps, and the fact that the type III system post-dates the flagellum), thus the working hypothesis is consistent with the bacterial flagellum was designed. Of course future data may overturn this hypothesis but that is true for all theories (except unfalsifiable ones).
Ged also says:
quote:
Dembski’s methodology is that the field needs to be “swept clean” of possible pathways in order for the “explanatory filter” to be applicable.
Again this is false. Dembski doesn't point to "all possible pathways" but currently testable ones. Of course, what I mean by "future pathways" is evolutionary pathways thought up in the future that may not even involve the Type III secretory system at all. Imagination is not testable and therefore garbage, which is why it is also not useful in science. We cannot test the statement "the fittest molecules had the most selective advantage", "or the filament just happened to work fortuitously with the export machine".
As far as finding probabilities where the it is under the probability bound, this is not "impossible", and in fact has already been done with various systems like the bacterial flagellum in Dembski's book and with the type III secretory system at ARN. I know several people are also working on more calculations.
Ged also asserts:
quote:
Science does not have an answer how the flagellum came about, not because it has no particular pathway in consideration, but because it has too many pathways in consideration!
And yet not a single one of these pathways is present in the literature, because of the constraining nature of the bacterial flagellum. For example, I challenge Ged to come up with 10, specific, testable pathways to the bacterial flagellum. If there are so many, then this should be an easy task. The only one that is put forth, is one that is in my opinion, failing, the Type III system to Bacterial flagella system.
Ged states:
quote:
The complexity of the possible pathways (in case “B” above) is very high.
But fails to mention even one. The reason is that the possible pathways Ged is thinking of exists only in his mind, and his mind cannot even fathom what one of these supposed possible pathways looks like. Thus no discussion of them is forthcomming. How many of them are there exactly? 10? 37? 59? 19999? If the answer comes back infinite, then you're not dwelling in science anymore, you're dwelling in Darwinian religion.
So as pathways are described to IC systems, pathways are constrained and no new pathways will be spawned, and even if they are, each pathway may or may not lend credence to another. This all has to be determined on a case by case basis. But such tight restrictions on pathways. the number of possible pathways is constrained by IC, further constrained by the particular components, further constrained by the amount of unselectable steps, and further constrained by the ability to actually test these models, further constrained by the fact that IC rules out specific types of pathways (direct ones) makes it unlikely that the complexity of any vague possibility space is "high".
As an aside Alix stated:
quote:
The Gödelian Argument is also interesting in that no scientist is actually suggesting that evolution would ever need to be able to connect any two points in a solution space; only that there exists a set of connected solutions. Indeed, it would appear far more suspicious if infinite connections appeared!
The Godelian argument actually makes the claim that no steps can be connected when the system is sufficienty complex, not that there can't be infinite steps. Evolution does indeed make the claim that there are connecting points in Biology, in fact I can't think of an instance where there is no connection between any two points. But the Godelian argument is that you cannot do this with respect to complex systems.
[ 12. April 2003, 15:24: Message edited by: Nelson_Alonso ]
IP: Logged
|
|
gedanken
Member
Member # 594
|
posted 12. April 2003 21:37
Alonso said:
quote:
Actually Dembski wants just the opposite. He wants us to completely test the model , or suggested evolutionary pathway that we actually have evidence for.
The words say “opposite”, but they confirm my point.
If there are gaps in present scientific understanding that leave multiple explanations, then Dembski demands that we find the actual evidence for which of those explanations (or pathways) is the “actual” pathway. He ignores the most fundamental point.
The most fundamental point is that his method supposes that we show that the possible physical explanations pathways in sum (the sum of “probabilistic resources”) should give a probability of less than 10^-150 chances of the event occurring.
quote:
For example, I challenge Ged to come up with 10, specific, testable pathways to the bacterial flagellum.
I’m not a biologist, and to do so would be a mistake. Others have proposed many potential pathways in many other threads. I shall not repeat their work, nor am I qualified to do so.
But I will comment on the “testable” aspect of that demand. If by “testable” you mean that we have the evidence to distinguish the particular pathway from others, then this may not be possible with today’s evidence -- and thus one cannot come up with 10 “testable” pathways with today’s evidence. If you mean generically testable, then these have clearly been proposed (whether or not you think that they have failed such tests, for example, I am not trusting in that judgment, though I think that they have not been “tested” because there is not sufficient evidence to make a discrimination at the moment).
The fundamental point, to continue the above discussion, is that the proposed pathways have a reasonable chance of being correct -- at least one of them. They certainly don’t meet the predicate of the EF (whatever the value of the EF) that they have below 10^-150 probability of being correct. How can you assure us that there is a less than 10^-150 probability that there were precursors to the Type III secretory system predating the flagellum, for example? You cannot. How can you make an assurance of better than 1 in 10^150 of a matter of dating the first occurrence of such a structure? The very fact of its somewhat later occurrence in observed history confirms that it has rather high probability.
quote:
The reason is that the possible pathways Ged is thinking of exists only in his mind, and his mind cannot even fathom what one of these supposed possible pathways looks like. Thus no discussion of them is forthcomming. How many of them are there exactly? 10? 37? 59? 19999? If the answer comes back infinite, then you're not dwelling in science anymore, you're dwelling in Darwinian religion.
This gets to the essence of the problem. It does not mater how many pathways and detailed sub-divisions of pathways need to be evaluated, with respect to finding answers that fit an evolutionary scenario. If details are found, then they will be found. And if not, then not. It matters not to Darwinian theory how long that takes -- as nothing has been posed that contradicts Darwinian theory so far here.
But that answer has profound implications for ID propositions. If the explanatory filter will change state of its inference with the turning up of even one such pathway that even has a probability of just over 10^-150 of having occurred, then the ID proposition is supersensitive to that analysis. The very fact that the ID proponents don’t know the answer to that question is telling. It is your problem, not mine.
IP: Logged
|
|
|
|
Nel
Member
Member # 614
|
posted 13. April 2003 13:04
Alonso said:
Ged once again confirms my point about "promisory notes" when he points to:
quote:
If there are gaps in present scientific understanding that leave multiple explanations, then Dembski demands that we find the actual evidence for which of those explanations (or pathways) is the “actual” pathway. He ignores the most fundamental point.
Secondly, Dembski does not demand that we find what "actually" happen. Just that when we propose a Darwinian pathway, that it be testable, and not invisible or imaginary, that no appeal to a future explanation would render the ID theory untenable. Thats ridiculous. Imaginary, hopes for future explanations are irrelevant. Imagine if I used that same argument to keep Darwinism out of the literature, we may have evidence that bacteria develop antibiotic resistance due to Darwinian evolution today, but thats only because ID theories don't presently understand. Sometime in the future, an ID theory will be proposed that will explain antibiotic resistance and thus Darwinian evolutionary theory is an argument from ignorance.
What Dembski means by "probabilistic resources" has absolutely nothing to do with summing up invisible and/or future explanations and getting less than 10^-150. The number 10^-150 is itself the reference class of probabilistic resources that we should test our improbabilities against. So once again, you misunderstand the very core of Dembski's thesis.
I noticed that you confirm my assertion that you have no idea how many possible pathways there are, or even what 10 of these pathways look like! And yet you assert that the possibility space is "complex" and that scientists don't know how the flagellum evolved, not because they have no pathways to work with but because they have too many pathways to work with!
With respect to testable pathways, only one has been proposed. And the reason for this may be that IC is constraining such possibility space. And once again you use an argument from ignorance when you point out that you have no evidence or testable pathway "today".
Ged writes:
quote:
How can you assure us that there is a less than 10^-150 probability that there were precursors to the Type III secretory system predating the flagellum, for example? You cannot.
Of course I can. For example:
quote:
"Phylogenies of most or all of the flagellar proteins follow those of the source organisms with little or no lateral gene transfer suggesting
that homologous flagellar proteins are true orthologues. We suggest that the flagellar apparatus was the evolutionary precursor of Type III protein secretion systems."
J Mol Microbiol Biotechnol 2000 Apr;2(2):125-44
With scientific evidence, not only can I show that there were no precursors to Type III secretion systems predating the flagellum, I can show that there were no Type III secretion systems predating the flagellum at all .
Again, if you say someday, somehow we will find evidence of a precursor type III secretion system then you are not only arguing against the scientific evidence (which suggests we may never find such a precursor) but you are using an argument from ignorance.
Secondly, the use of the UPB is for testing what the probability of a certain pathway is, and not for testing the probability of finding an imaginary precursor. Show us the physical precursor and lets "do the calculation."
Ged again amits that it doesn't matter to Darwinian theory whether one pathway is found, or none. In fact, Ged has admitted that a great number unselectable steps doesn't even matter. It's "irrelevant". But unselectable steps from the M ring to the filament is a grand problem for evolution, for the flagellum for example, --- unselectable steps points to pure chance events, tornado in a junk yard, and renders the Darwinian theory moot. Whether the evidence supports Darwinian theory is irrelevant to Ged, that it occured is a fact (although it's a metaphysical fact taken on faith, since he has no idea what these pathways would look like). Lets take a look at the physical precursors, lets take a look at the unselectable steps, lets take a look at the probability for these pathways. The more you suggest, the more work for ID theorists and Darwinian evolutionists, the more we will compete for the better theory, the better for science as a whole in the long run.
Note Dembski's discussion on Godel's theorem has completely went off a ramp on a dirt road and we may get zapped by the moderator pretty soon.
Note also, I have no idea why Ged linked to an essay on "Design and the argument from evil" in this thread".
[ 13. April 2003, 15:33: Message edited by: Nelson_Alonso ]
IP: Logged
|
|
gedanken
Member
Member # 594
|
posted 13. April 2003 15:16
quote:
Note Dembski's discussion on Godel's theorem has completely went off a ramp on a dirt road and we may get zapped by the moderator pretty soon.
I had asked on several occasions for a focus for this thread or what was the intended limitation of subject matter. I was lead to understand that the subject matter of the entire contents of “logical underpinnings” paper was included (and not only by positive response, but because the entire paper is relevant to understanding what would constitute a “Gödelian” argument).
I provided the link to Dr. Dembski’s new paper because it contains the 3rd through 5th paragraph of his opening post.
Making the Task of Theodicy Impossible? Intelligent Design and the Problem of Evil
[ 13. April 2003, 15:46: Message edited by: gedanken ]
IP: Logged
|
|
Nel
Member
Member # 614
|
posted 13. April 2003 15:30
Ged, how is it different? Some of the wording is even identical to his OP.
[ 13. April 2003, 15:31: Message edited by: Nelson_Alonso ]
IP: Logged
|
|
gedanken
Member
Member # 594
|
posted 13. April 2003 15:41
Yea, I just noticed -- I was busy tracking it paragraph by paragraph and updated the above before I saw your post. Now I've removed the quote since it was redundant. This was much more relevant to the "Intelligent design is not mechanistic" thread, which was zapped, and I posted it both places at the same time. It is interesting how Dembski used the argument with regard to "mechanistic reductionism" which Noel Rude referenced above.
[ 13. April 2003, 15:50: Message edited by: gedanken ]
IP: Logged
|
|
gedanken
Member
Member # 594
|
posted 14. April 2003 00:21
quote:
With respect to testable pathways, only one has been proposed. And the reason for this may be that IC is constraining such possibility space. And once again you use an argument from ignorance when you point out that you have no evidence or testable pathway "today".
I think that far more than “one” pathway has been proposed. Alonso seems to think that talking about a “Type III secretory system” as a precursor is “one” pathway -- while it is actually many pathways as the details within that are considered. But the issue is not so much that particular type of system as the precursors to that system. The obvious reason that a precursor would be likely to contain a “secretory system” is that the flagellum itself is constructed by a secretory system, so obviously any precursor that changes slightly to a closer variant must do some part of the action of that construction. This is not a matter of the “ICness,” it is a matter of obvious physical constraints.
No I “have no evidence or testable pathway” because I am not a biologist and I don’t deal with the biological issues at all. My points are about the logical form of the issue, and pointing to others who discuss the issue, how they relate to the “Gödelian” claims that Dembski has made.
Of course Alonso has used a couple of rather effective appearing debating tricks by asking me to discuss areas outside of my expertise, and the use of the term “promissory note”. It is this term “promissory note” that is especially effective in making an emotional appeal that somehow science is failing -- but I think the readers will see through the veil of ignorance that backs that up.
There seems to be quite lively discussion of issues of the development of the flagellum, for example the ARN thread “Flagellum and evolution”. These discussions, appearing in various places, convince me that Dembski, Behe, and other ID presenters have made little headway in providing convincing evidence that even what they do examine has a probability below 10^-150, much less that they consider the available pathways that have been discussed as reasonable alternatives for the development of the flagellum. So clearly the predicates of the “Explanatory Filter,” which is Dembski’s argument of inference of “design”, has not been even closely met.
(For an example on another supposedly IC example, see the lively discussion on “Intelligently Designing Immunity. Once again the IC argument is not holding much promise of showing a probability below 10^-150 of holding for their claims.)
Dembski and others point to the incompleteness of scientific research, and demand that if it is not made complete that a Gödellian argument inherits in the complexity of such research. This hardly compares to the Gödellian implications of the claim of the “explanatory filter” which takes as a logical predicate that we have carefully calculated this exceedingly low probability for the sum of all pathways to the event -- and then claims that Gödellian complexity ensues in the lack of scientists own work rather than considering that same issue applied to their lack of calculation of such a probability. (Wasn’t there some sort of “promisory note” that the calculation would be done for the flagellum? I remember Dembski himself promising that the book No Free Lunch would “do the calculation” for some significant case of importance.)
Alonso will probably repeat that I am demonstrating “ignorance” in my argument, in some fashion, as he has done in the previous post. And of course I am “ignorant” of the details of the subject beyond my area of experience. What is not beyond that area of experience is the ability to examine the logic of the claim that that convincing evidence has been presented that the probability of the flagellum having developed along the lines of one of those proposed pathways is below 10^-150. Does anyone here know how small the probability of one in 10^150 is? That is substantially larger than the number of elementary quantum operations that have occurred in the live of the universe, according to Dembski’s claims.
Alonso may also repeat that Dembski has handled this issue by claiming that we must deal with “presently” available concepts of the pathways to the flagellum. But of course the claim of the explanatory filter is that it applies when this tremendously low probability has been calculated. It is incumbent on the presenter making the new and novel claim (e.g. of “design” using the suspect methodology of the “explanatory filter”) that the conditions have been met. It is hardly fair of Dembski to demand that scientists to his calculation for him, and then claim a failure similar to what occurred in the mathematical level proof of Gödel’s demonstration when scientists have not completed his work for him.
I don’t have to know any biology at all to know that the arguments presented show there is something near even money on some pathway being reasonable -- this is a tremendous distance from a convincing argument that there are less than one chance as compared to the number of elementary particle transitions that have occurred in the entire life of the universe.
The analogy of the incompleteness of current scientific research on the subject of the flagellum (even on any scientific topic) has been shown to be a very weak analogy to the issue of logical proof involved in Gödel's demonstration. Even Alonso has admitted that:
quote:
I was not pointing to the deductive argument by Gödel, or whether the ID argument is empirical in nature. I quite clearly compared Hilbert's realism VS Gödel's idealism with respect to mathematics with IC and Darwinian evolution, which is why I didn't think it was relevant.
If the subject matter of this thread is supposed to be the comparison of the incompleteness of scientific research (a never ending task) to Gödel's demonstration, then this thread indeed may have run out of steam. Unless there is another topic to discuss, I think the subject has been exhausted.
[ 14. April 2003, 01:02: Message edited by: gedanken ]
IP: Logged
|
|
Noel Rude
Member
Member # 516
|
posted 14. April 2003 10:15
Perhaps one last point: Discontinuities may abound in the Platonic realm and in the cosmos, but the mind bridges them all. Which is to say that the demarkationists, such as, say, Gould and his NOMA ("NonOverlapping Magisteria"), who would extend our "Gödelian" discontinuities into the realm of epistemology, are mistaken. Humility demands we say, "I do not know", when such be the case. But it is arrogant to say "It cannot be known" when of course this is precisely what we cannot know: That this or that can never be known.
And this is the glory of science. It is the unknown that entices and beckons and the explorer knows no bounds.
IP: Logged
|
|
Nel
Member
Member # 614
|
posted 14. April 2003 15:07
Ged,
As far as I know, the type III secretory system to bacterial flagellum is only one pathway. Give me a second one utilizing the type III secretory system as a precursor. You will see how the bacterial flagellum itself constrains the possibilities here when attempting to do so.
The flagellum is not constructed by a "secretory system". The export machine of the flagellum only bears resemblance to the type III system. And the reason for that similarity, as my reference shows, is because the secretory system evolved from the flagellum and not vice versa. There is good design reason why there is a need for the export system, the flagellum needs to be built from the inside-out.
Ged writes:
quote:
Of course Alonso has used a couple of rather effective appearing debating tricks by asking me to discuss areas outside of my expertise, and the use of the term “promissory note”. It is this term “promissory note” that is especially effective in making an emotional appeal that somehow science is failing -- but I think the readers will see through the veil of ignorance that backs that up.
Ged, I'm simply holding you accountable for your assertions. You made the claim that the bacterial flagellum does not contradict Darwinian theory and that the possibility space is complex. Due to the pausity of ID advocates on these kind of message boards, you are probably used to making this claim without much resistance. But I will hold you responsible for your claims. If you feel that you cannot support your statement here because it is "outside your expertise" then I suggest you stop making those statements. Otherwise, I will continue to ask you to support it. This is not a "debate tactic" this is about intellectual responsibility.
You also write:
quote:
(For an example on another supposedly IC example, see the lively discussion on “Intelligently Designing Immunity. Once again the IC argument is not holding much promise of showing a probability below 10^-150 of holding for their claims.)
And yet you freely admit that you are unable to discuss these topics when it comes to supporting these kind of assertions. But you have no problem with making them if it supports your position. For example, can you show in your next reply, why you think that acquired immunity is easily evolvable? If so, why was there a Big Bang of immunity instead of evidence for a step by step origin? Why is it that I can't find a single organism with a few antibodies with high specificity? Why is it that I can't find any antibodies without all three VDJ segments? Why is it that if I remove VDJ recombinase, I don't have antibody diversity? Don't you find it interesting that the mammalian pattern of VDJ segments requires the pre-existence of the VDJ recombinase?
With respect to the Godelian argument, again, the Godelian argument does not apply to the filter or to Dembski's claim that things of lower probability then the UPB do not occur by chance. Or to Dembski's want for scientists to put forth testable pathways. Science is about putting forth testable hypothesis with evidence that a particular evolutionary pathway took place. Scientists claim that there are connections between points in Biology and that all of Biology can be reduced to this. The Godelian argument refuted this for mathematics, ID is doing the same for Biology with IC and specified complexity.
Dembski has shown how to apply his calculations to a biological structure (the bacterial flagellum) in his book No Free Lunch, and in fact has even taken the Type III secretory system into account in a recent reply to Miller.
Dembski's claim is that for Darwinian evolutionists to satisfy that their pathways are logical and testable and likely, they must "do the calculations" when they propose a pathway. Asserting that a pathway is above the UPB is only wishful thinking unless the calculation is done. It is ridiculous to think that every time a pathway is shown that Dembski must immediately do the calculation for you. Darwinian evolutionists are grown ups and can do their own work, while collaberating with ID theorists.
I disagree that you don't need to know any Biology in order to know that a certain pathway is not reasonable. For example, you keep saying that the type III secretory system predates the flagellum. But how could it when there is evidence, not only phylogenetically, but other independant lines of evidence concerning how evolution itself works that shows it evolved from the flagellum? That, in and of itself, reduces the probability of this event to near zero.
As for the comparison between empirical and mathematical arguments, I have not "admitted" that the analogy between Godel VS Hilbert is weak, that is a completely false statement. What I stated was that it is an analogy.
An analogy is defined as:
quote:
Similarity in some respects between things that are otherwise dissimilar.
In other words, analogies have these properties:
aaaaX
aaaaY
If you notice from these two sequences, analogies have very similar, perhaps even exact properties between them, but they also have dissimilarities (i.e. the X and the Y bewteen the two sequences). Thus of course there will be differences, to say that this in and of itself makes the analogy weak is ridiculous. If there were no dissimilarities the comparison would be literal rather than analogous.
[ 14. April 2003, 17:46: Message edited by: Nelson_Alonso ]
IP: Logged
|
|
gedanken
Member
Member # 594
|
posted 14. April 2003 22:10
Alonso, if I accidentially wrote that the “Type III” secretory system predates the flagellum, I’ll correct it. I was trusting your statement, and all that I remember writing were cases in which I meant to be indicating that some observation of the flagellum predates that of the Type III. Except for my argument that it had to have occurred at least “simultaneously” in an above post, with which you are disputing that the flagellum is constructed by a “Type III secretory system”. I am certainly not qualified to make such a judgment, and make no claims to be knowledgeable in this area. I can, however, read Miller and Levine when they say:
quote:
On the basis of these homologies, McNab (McNab 1999) has argued that the flagellum itself should be regarded as a type III secretory system.
….
McNab, R. M., 1999. The Bacterial Flagellum: Reversible Rotary Propellor and Type III Export Apparatus. Journal of Bacteriology 181: 7149—7153.
So there certainly is some use of the terminology in the literature in which the flagellum is considered to be constructed by a “Type III secretory system”. This may be a matter of terminology and very refined differences of definition, and there may be disagreement among scientists as to terminology or this issue. For me what is important is the degree of similarity of this action of bacterial to “secrete” things, and the construction mechanism of the flagellum. It seems very similar, whatever the exact details.
I notice you still stick to defending Dembski’s claim that scientists need to “do the calculation” in order to verify that a given path is greater than 10^-150 probability.
I hope that the readers understand just how low a probability that is. And the readers should note that one chance in billions of a particular path being correct (something that would be scientifically irrelevant) would still be many many many orders of magnitude greater than 10^-150 probability.
I think that the demand by Dembski that scientists “do the calculation” is basically irrelevant. Not because having that information itself would be irrelevant, but because the requirement that something be so extremely unlikely as 10^-150 is so incredibly small that any plausible pathway described whatsoever would have a probability of greater than that. If readers are capable of understanding that, they will understand what I said. And if not, then I won’t bother further trying to educate them. (And it is up to the one proposing the novel result to back up his claim.)
As to the analogy, I think that many posts done more ably than I was able have shown the analogy to be very weak, and is more like
abcdX
defgX
than your example
quote:
I was not pointing to the deductive argument by Godel, or whether the ID argument is empirical in nature. I quite clearly compared Hilbert's realism VS Godel's idealism with respect to mathematics with IC and Darwinian evolution, which is why I didn't think it was relevant.
As I read this, all that is being compared as similar or an analog was Gödel’s “realism”, and other aspects were not similar. Gödel’s result was in a matter of mathematical “proof”, Dembski’s claim is not so. I urge the reader to check out page 1 of this thread, including RBH and Rex Kerr’s remarks.
But Dembski’s method (explanatory filter) rests on a set-theoretic point that “design” is within the set in question most likely, and that all other cases have been eliminated by the given argument most likely. (The logic of the eliminative argument is strictly set-theoretic, but the demonstration of real-world conditions matching the set-theoretic constraints is simply an argument that it is likely.) But the real key is that if a mistake is made in evaluating the eliminative argument, then the result of Dembski’s set-theoretic claim is in error. The EF depends on the logical complexity of having covered all pathways in a reasonably likely case -- which is the difficulty.
This gets back to Alonso’s insistence that starting with the “Type III secretory system” is a single pathway. All along, the argument is that there is a common precursor to all -- and this is reasonable to assume even if you assume that the “Type III secretory system” evolved from the flagellum!. That the “Type III” evolved from the flagellum would be demonstration of the type of evolutionary change being considered. And the very “export system” is some sort of process that is very similar in both flagella and other bacterial systems.
I find it interesting that Alonso is willing to claim that the “Type III secretory system” is likely evolved from the flagella -- accepting that sort of evolution -- while insisting that some sort of missing information makes the evolution of the flagellum itself from some sort of export system that was a precursor have a probability of 10^-150.
Once again, the novel claim is the one that needs to be backed up. This is the essential point here. Since one must eliminate cases within a web of extreme complexity, it is the eliminative argument that suffers most from the Gödelian argument.
Alonso, I suggest a new thread. (One that I probably won’t participate much in due to lack of experience in biology). The thread could be a discussion of how constraints on the flagellum force the number of options to be small -- and a specific focus is the evidence that this is a discrete and a finite number of possibilities, such that they can all be eliminated -- and the arguments for that elimination. You could start with whatever Dembski presented in NFL to support that particular point. (I mean all the biological research that Dembski presented to support his point.)
[ 14. April 2003, 22:43: Message edited by: gedanken ]
IP: Logged
|
|
|
Printer-friendly view of this topic