|
Author
|
Topic: Front Loading
|
RBH
Member
Member # 380
|
posted 15. January 2003 12:18
I knew I should have resisted temptation!
warren wrote quote:
The 'we have too many other important things to do' argument is not very credible. Darwinian theory characterizes evolution as a one cycle per lifetime teleological process. The theory predicts that 1)the more complex the adaptive problem (the larger the search area and the fewer the adaptive solutions) and 2)the less time available to find an adaptive solution(the shorter the length of time until the environmental change necessitating the adaptive change kills the organisms) the less likely the organism or species is to survive. Since evolution is limited to one cycle per lifetime, an organism should have no opportunity to evolve an adaptive solution if the time required for adaptation is less than one lifetime.
Let me deconstruct that a bit.
1. It is not a "too many other important things to do" argument. It's a description of the domain of inquiry of evolutionary biology. Newtonian celestial mechanics does not purport to be a theory of biological evolution nor a theory of turbulence in fluids. Is that therefore a legitimate critical claim for the inadequacy of Newtonian mechanics? And, to be strictly accurate, I pointed out at least one 'within-lifetime' mechanism (gene expression affected by environmental variables) that operates within the lifetime of an individual organism, and I also pointed out the disciplines that explicitly concern themselves with "within lifetime" adaptation.
2. warren claims that quote:
Darwinian theory characterizes evolution as a one cycle per lifetime teleological process.
That is false in two respects. First, the finest-grained temporal unit of analysis in evolutionary theory is not "lifetime," it's "generation." Second, in evolutionary theory it's not characterized as a teleological process.
3. warren asserts that quote:
The theory predicts that 1)the more complex the adaptive problem (the larger the search area and the fewer the adaptive solutions) and 2)the less time available to find an adaptive solution(the shorter the length of time until the environmental change necessitating the adaptive change kills the organisms) the less likely the organism or species is to survive.
The assertion that as search space grows there are "fewer adaptive solutions" is not a prediction of evolutionary theory, and in fact it may be false. The proportion of "adaptive solutions" may be constant in search space size, or it might even increase. Evolutionary theory itself says nothing at all about that issue: That's a fact (not a prediction) in the mathematics of search algorithms, not evolutionary biology.
It is true, however, that if the search space grows but the proportion of "adaptive solutions" decreases, then if the topography of the search space is uncorrelated, search time will grow. But as noted above, that's not a statement in the theory of biological evolution, nor does evolutionary biology assume it or require it, and it is not discomfited by it. Warren is here making the same conceptual error that 'random assembly' probability calculations universally make, and that is refuted by Gedanken's remarks elsewhere (I don't have the URL of the thread right here, but it's on ISCID somewhere): No "search" for adaptive solutions in biological evolution starts from a random point in the search space! Evolutionary searches start from already-adaptive points in that space. Assuming only that fitness landscapes are correlated (meaning that within 'neighborhoods' on the fitness landscape the points look more or less similar to the fitness function), the probability of finding an accessible adaptive solution required by a changing environment is much higher than is implied by a 'random-landscape' estimate.
4. warren says quote:
Since evolution is limited to one cycle per lifetime, an organism should have no opportunity to evolve an adaptive solution if the time required for adaptation is less than one lifetime.
Leaving aside the "lifetime" error, warren is here correct in one thing: If the selective environment changes too fast for a population to track it, it's likely the population will go extinct. And that happens all the time! The history of life on earth is positively littered with the traces of extinct populations that failed to sufficiently swiftly track a changing selective environment. I have no idea where warren gets the notion that evolutionary theory rules out the possibility of extinction.
Added in edit: Rereading, I notice one more egregious misrepresentation of evolutionary theory in that quotation: "an organism should have no opportunity to evolve." An organism does not evolve. Populations evolve. The unit of analysis is population genomes, not an individual organism. That's a common misrepresentation and leads to all sorts of nonsensical claims about evolutionary theory's alleged predictions.
5. Along those lines warren repeats that quote:
The actual data available, however, clearly contradicts this Darwinian prediction. Organisms are fully as capable of generating complex solutions to adaptive problems requiring rapid adaptation as they are of generating complex solutions to long term adaptive problems. Darwinian theory offers no explanation (or a hand wave, we'll figure it out some day explanation).
I repeat, the world is full of death, both of individual organisms and of populations, due to their failure to adapt quickly or effectively enough. This morning my cat brought a dead sparrow to the door. That sparrow failed the test of quick adaptation, and it is dead. Its genes no longer form part of the sparrow population's genome, and that genome is therefore (however slightly) altered as a result. Evolution is occurring all around us all the time. So is the failure of organisms to adapt quickly enough to within-lifetime changes in their environments. That sparrow paid the price of its failure to adapt to the presence of the cat and it died.
If warren purports to be criticizing contemporary evolutionary theory, it would be more than helpful if it were accurately represented in his critique. This is my last posting in this thread. The return on the effort required is too small to spend my time on it.
RBH
[ 15. January 2003, 13:51: Message edited by: RBH ]
IP: Logged
|
|
Frances
Member
Member # 169
|
posted 15. January 2003 13:09
Nelson,
First of all, I would like to apologize to you if my reference to the AIG website caused you any distress. My comments led to me being banned from posting for one day because among others,according to the email from one of the moderators, may have been interpreted as an attempt to 'smear[ed] [your] face in the mud of creationism'. If that's the case then I would like to point out that I had no intention to smear your face in such mud and will openly apologize if such impression was raised by my posting. Rather I wanted to point out that your quote of primary research was erroneous. It was my mistake that I concluded that you had quoted from AIG rather than from Denton.
I will from now on strengthen my attempt to be more patient in my replies in order to encourage an open discussion of ideas on this forum. I thoroughly enjoy our discussions here and I hope that others feel similarly.
Per moderator's suggestion I will also not provide any more links to the archive of my postings.
I hope that these steps can be helpful in generating an atmosphere in which we can discuss ID ideas in a constructive manner.
Now on to my response to your posting
And while at smaller resolution indeed this may seem to be perfect, reality shows that at high magnification they are hardly that perfect.
Another example of these 'perfect squares' in prawn shrimps
I would like to point out that my main objection is to Nelson's claims that no evolutionary pathways or selective advantages have been shown for the evolution of eyes, particularly in lobsters and Pecten. By relying on second hand resources which have been shown to be of doubtful accuracy in many other areas, Nelson has been furthering an argument for which he does not seem to have any first hand information. And yet he is willing to make claims that would go beyond what would be supportable without any knowledge of first hand sources. This has led Nelson to make such assertions as 'perfect squares' when in fact the photos show that they are hardly such. Other mishaps have been documented elsewhere and in the rest of this posting.
The moral of the story is that if one wants to argue that t is the failure of natural selection and RM alone to account for these eyes that opens the door to front-loading. one should be familiar with the actual evidence and not some second hand resource.
Nelson still seems to be unable to grasp the simple fact that the observation that Pecten needs to see in and outside the water may explain the selective advantage of an eye that can see in both environments. The difference between the Pecten and its precursors need not be that large when one realizes the likely pathways.
quote:
As noted in Chapter 1, the case of Pecten is fascinating. It has evolved an eye with two separate retinas placed
next to each other but separated from the tapetum and other elements at the posterior of the optical orb. The
tapetum has become a reflective mirror in a catadioptric optical system consisting of the objective group and the
tapetum. When the eye is immersed in sea water, the cornea is ineffective but the crystalline lens and the tapetum
combine to form a catadioptric optical system bringing light to focus on one of the two retinas. When the eye is
not immersed in sea water, the cornea of the objective group is effective and the cornea and crystalline lens operate
as a dioptric optical system with the other retina. This provides an animal living in an estuary with focused vision
under both aquatic and terrestrial conditions.
From "PROCESSES IN BIOLOGICAL VISION by James T. Fulton"
First of all the reflector. It should be noticed that
quote:
The tapetum sheet can evolve to form a variety of functions depending on the animal. It is generally a passive
layer. Normally, it can aid in the absorption of stray light that has passed through the retina. In some cases, it
consists of small groups of cells that act as a retro-reflector to direct light back through the retina. As seen in the
case of the mollusc, Pecten, the cells can also be used to form an optically coherent sheet of cells that form a
reflecting optical element in a catadioptric lens system.
Ibid
Now the retina
quote:
The individual photoreceptors are similar in structure to those of Arthropoda, i.e., the chromophoric material is
found in rods exuded from the sides of the photoreceptor cells. .... Whereas the rhabdom of Arthropoda
exhibits a circular symmetry with respect to the centerline of the assembly, this is much less evident or
nonexistent in Mollusca. The limited data available indicates an orthogonal grouping of photoreceptor cells
to achieve a higher sensitivity to the polarization of the incident light.
Ibid
quote:
It is the failure of natural selection and RM alone to account for these eyes that opens the door to front-loading.
In fact you have failed to show that RM&NS are a failure to account for these eyes so there goes your justification for front loading but you seem to still be unable to describe to us what front loading really is, other than not RM&NS..
Describe your front loading scenario in some detail please. And explain what mechanisms you have in mind that played out for times > t_0 (the time of front loading). I would encourage you to check out some of the works on Intelligent Design that would allow you to familiarize yourself with the concepts and their strengths and weaknesses.
quote:
3. You say: " Spondylus Until Nelson can show us from the original research papers what the eye of the Spondylus looks like as compared to Pecten we have no real way to discuss this."
You did this yourself:
Francis:
In the Spondylus and Pectinidae, the eyes are quite well developed consisting of a cornea, lens and retina.
[quote]
That suggests to me that Nelson is not familiar with the primary sources that describe the eyes of Pecten and Spondylus. In fact as far as I have been able to tell Spondylus eyes are not like Pecten eyes at all.
For instance from Ibid:
[quote]
use of two separate optical forms within the available physical envelope in Pecten, including introduction of an
entirely new optical form to animal physiology--a catadioptric lens system.
Pecten seems to be unique in this aspect. Perhaps Nelson can present us in some more detail Spondylus? Which species of Spondylus btw is Nelson refering to?
quote:
Anyway, my discussion of front-loading and eye development center around pax-6. Pax-6 has remained more or less the same in most branches of life. In every animal examined that has eyes this gene proves to be involed in eye development. Moreover, it is necessary for function. In both Mice and flies pax6 mutations severely affect development of eyes. Drosophila has two closely related pax6 genes, eyeless (ey) and twin of eyeless (toy). Expresssion of either of these genes in antennal, leg and wing imaginal discs in drosophila causes well formed ectopic eyes. In xenopus embryos ectopic injection of pax6 into blastomeres causes ectopic eye-like structures. So not only is Pax6 required for eye formation, it is also the author of eye development. Pax-6 genes control both upstream and terminal functions in the gene network for eye development. Since pax-6 genes also exist in primitive organisms like sponges, a prediction of FL (Mike or Warren can correct me if I'm wrong) would be that it plays a non-essential role in these organisms.
So far so good so we have evidence that evolution shaped the expression of the Pax-6 gene as well as many other hox genes. But how does this show evidence of front loading? That essential genes have changed little over time but have been quite able to lead to different evolutionary shapes shows how a simple variation on the timing of embryological development may have a significant impact on the resulting form. But as with Mike Gene, Nelson is now painting an arrow around the bullseye by arguing that evidence of common descent is suddenly evidence of 'front loading'. Well we all agree that at t=t_0 pax-6 gene existed as an initial condition and how RM&NS played a role in shaping life there after. If Nelson wants to argue front loading then he surely has to accept the role of natural forces shaping evolution for t>t_0 or he is arguing for intervention rather than front loading. But while evolution can explain in a non ad hoc manner the evidence, Nelson seems to want to argue, without much evidence, that this front loading or initial condition required intelligent design. Yet in the actual discussion Nelson seems to be wavering between front loading and intervention and so far has been unable to provide us with the necessary details.
[ 15. January 2003, 13:17: Message edited by: Frances ]
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 16. January 2003 10:09
Quote: If warren purports to be criticizing contemporary evolutionary theory, it would be more than helpful if it were accurately represented in his critique.
Just to clarify, I am not purporting to criticize current Darwinian theory but to offer an alternative theory. I am not misrepresenting current theory or at least RBH has presented no arguments or evidence demonstrating any misrepresentation relative to the subject being discussed. My comments may not be entirely consistent with certain stylistic conventions often used in evolutionary biology, but such inconsistencies have nothing to do with the substance of the subject being discussed.
I found it interesting, although obviously not necessarily meaningful, that RBH declined to address my claim that you can use front loading to generate very large increases in evolutionary/design processing speed and capacity. This is clearly both a subject on which he would be expected to have some knowledge and a subject which suggests an alternative to basic Darwinian theory.
While I don’t claim to be an expert on current Darwinian evolutionary theory, I am somewhat skeptical that RBH is accurately reflecting the current thinking on short term adaptive change. It is my understanding that the ‘evolution of evolutionary processes’ is generally accepted as a reality. If evolutionary processes could evolve, then logically processes to produce short term adaptive changes could have evolved. While I doubt if there is a consensus on the issue, I would guess there are a lot of scientists who believe short term adaptation is part of evolution. They simply do not know how to reflect short term adaptation into a predictive mathematical theory.
Quote: Second, in evolutionary theory it's not characterized as a teleological process.
I am well aware that biologists do not like to refer to Darwinian evolution as a teleological process. However, the technical/scientific definition of teleological process or causal relationship is 1)a process or relationship which achieves a goal or purpose, and 2)which is produced by a complex iterative process involving selection, variation, and preservation. The general concept of ‘scientific or materialistic teleological causation’ predates Darwin by a couple of thousand years. Whether biologists like it or not, Darwin defined evolution as a teleological process.
Quote: Warren is here making the same conceptual error that 'random assembly' probability calculations universally make, and that is refuted by Gedanken's remarks elsewhere (I don't have the URL of the thread right here, but it's on ISCID somewhere): No "search" for adaptive solutions in biological evolution starts from a random point in the search space! Evolutionary searches start from already-adaptive points in that space.
The above includes two ‘arguments’ which purportedly explain how very weak Darwinian teleological processes can generate very complex design. It may be useful to address these flawed arguments.
The random assembly argument claims, accurately, that random assembly does not accurately measure the complexity of the assembly or manufacture process. The incorrect or misleading suggestion or implication made from this observation is the actual manufacture process is less complex than the random assembly estimate. In actual practice, if you account for front loading, random assembly usually dramatically underestimates the complexity of the assembly process.
Second, the ‘starts from an existing adaptive position’ argument is also misleading. In the type of evolutionary/adaptive changes being discussed here, it is assumed that some environmental factor such as weather or a competitor or a disease has changed and the organism must change or adapt in order to survive.
Starting from an adaptive position only reduces the complexity of an adaptive change if the systems is using a structured, directed or non-random search. The complexity of a search can be greatly reduced if the non-adaptive state is due to one of a set of variables, the system knows which variable is mal-adaptive and can generate variance on that specific variable and not others.
The complexity of finding an new adaptation, using a directed search, is only reduced if the number of mal-adaptive variables is small. In biological systems, variable are interrelated and even seemingly small adaptive changes will involves changes in large numbers of variables. (consider as an obvious example the large number of variables associated with a ‘simple’ change in size).
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 16. January 2003 15:38
This is going to be the kind of posting that the Moderator doesn't like, but I'll give it one more try, though I scarcely know where to begin with the mis-statements, errors, and misrepresentations in warren's last posting.
warren wrote quote:
Just to clarify, I am not purporting to criticize current Darwinian theory but to offer an alternative theory. I am not misrepresenting current theory or at least RBH has presented no arguments or evidence demonstrating any misrepresentation relative to the subject being discussed.
I pointed out several respects in which warren misrepresents current evolutionary theory. Two examples are his attribution of evolution to individual organisms rather than populations and identifying the time unit of analysis as "lifetime" rather than "generation." Neither is a "stylistic convention;" both are real and important differences.
warren wrote quote:
While I don't claim to be an expert on current Darwinian evolutionary theory, I am somewhat skeptical that RBH is accurately reflecting the current thinking on short term adaptive change. It is my understanding that the 'evolution of evolutionary processes' is generally accepted as a reality. If evolutionary processes could evolve, then logically processes to produce short term adaptive changes could have evolved.
They have. It's called "learning." I identified the several disciplines that deal with those sorts of "adaptation." Further, it's not "the evolution of evolutionary processes" that's generally accepted, it's the evolution of evolvability. They're different. And how "the evolution of evolutionary processes" implies the evolution of short-term within lifetime adaptive capabilities is beyond me.
warren went on to write that quote:
While I doubt if there is a consensus on the issue, I would guess there are a lot of scientists who believe short term adaptation is part of evolution. They simply do not know how to reflect short term adaptation into a predictive mathematical theory.
As I noted above, those sorts of adaptations are addressed in several disciplines. Within evolutionary biology, the interactions between environmental events and the expression of genetic instructions is an area of interest and research.
There are mathematical difficulties in modeling a system's dynamics when they occur simultaneously at several disparate time grains, particularly when the dynamics at the several levels are loosely and non-linearly coupled as they are in biological and evolutionary systems. (See the Baldwin effect for an example of such coupling.) In fact, John Holland, who is the 'father' of genetic algorithms, has asserted that those mathematics do not yet exist. If warren has invented them I wish he'd publish them. There are lots of electrical engineers and physicists (along with computational biologists) who would be eager to see that.
I'd need to see something more than warren's unsupported opinion that "a lot of scientists" who believe that short term (within-lifetime) adaptations are part of the theory of evolution. I consulted several biologists, and they said "Huh?" Clearly the kind of short term adaptations exemplified by learning from experience have a heritable component, but are not themselves part of the explanatory apparatus of evolutionary theory. (See my remarks on domains of inquiry above.)
warren wrote quote:
I am well aware that biologists do not like to refer to Darwinian evolution as a teleological process. However, the technical/scientific definition of teleological process or causal relationship is 1)a process or relationship which achieves a goal or purpose, and 2)which is produced by a complex iterative process involving selection, variation, and preservation. The general concept of 'scientific or materialistic teleological causation' predates Darwin by a couple of thousand years. Whether biologists like it or not, Darwin defined evolution as a teleological process.
The phrase "which achieves a goal or purpose" is true of evolution, but does not connote teleology. Has a goal or purpose that was extrinsically imposed or carries out someone or something's goal or purpose both are properly characterized as teleological, but merely achieving a goal or purpose does not. For example, a waterfall "achieves the goal or purpose" of transferring water from a higher point to a lower point, but to argue that the waterfall is therefore a teleological process is to so distort "teleological" as to make it meaningless. "Function" and "purpose" are not synonyms.
warren wrote quote:
Second, the 'starts from an existing adaptive position' argument is also misleading. In the type of evolutionary/adaptive changes being discussed here, it is assumed that some environmental factor such as weather or a competitor or a disease has changed and the organism must change or adapt in order to survive.
Yup, and my remarks about starting from an initially adaptive point and tracking a changing environment depends on the rate of introduction and dispersion of variability in the population. If it's fast enough, then through natural selection the population can track the changing environment, which is called "evolution by random mutation, recombination, and natural selection." If not, the population goes extinct. And a whole lot have done so. Though he has asserted it often, warren has never shown that observed biological adaptations in real populations have actually occurred too swiftly to be accounted for by 'orthodox' evolutionary processes or by within-lifetime processes like learning, habituation, sensitization, biological development, and the like.
Finally (for me, anyway) warren wrote quote:
The complexity of finding an new adaptation, using a directed search, is only reduced if the number of mal-adaptive variables is small. In biological systems, variable are interrelated and even seemingly small adaptive changes will involves changes in large numbers of variables. (consider as an obvious example the large number of variables associated with a 'simple' change in size).
Might I commend to warren's attention the readily available material on control genes and their effects on morphology during development? Changes in a relatively few control genes can cascade into the number and variety of variables involved a 'simple' change in size. That's not real new stuff and it's readily available in the dreaded (and by warren, derided) peer-reviewed literature.
warren says he doesn't claim to be an expert on contemporary evolutionary theory, yet he regularly makes extravagant claims about what evolution can or cannot do and what biologists know and don't know. Simply cataloguing those false claims requires more time and effort than it's worth. I really quit this time!
RBH
Edited to insert a missing quotation mark.
[ 16. January 2003, 18:07: Message edited by: RBH ]
IP: Logged
|
|
Irving
Member
Member # 535
|
posted 16. January 2003 16:39
Here's an even closer picture of a lobster eye. I haven't really been following the discussion here that closely, but I did like the pretty pictures and thought this one might help. I'll leave it to Frances and/or Nelson to comment on it's importance.
Also, I hope people aren't confusing design with manufacturing. The use of either hand-hewn beams, or saw-mill lumber shouldn't detract from the architectural significance of a structure.
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 17. January 2003 11:05
Using the concept of front loading, I introduced a general theory of evolution which I suggest offers an alternative to Darwinian theories. We can label the proposed theory the:
STRONG TELEOLOGICAL THEORY: Under ideal conditions, biological design processes will produce or cause the type of front loading most likely to be compatible with the goal or purpose of survival.
This can be compared to
DARWIN’S WEAK TELEOLOGICAL THEORY: Under ideal conditions, evolutionary change is produced by the interaction of variance, heritability and Natural Selection.
or the even more restrictive
NEO-DARWINIAN WEAK TELEOLOGICAL THEORY: Under ideal conditions, evolutionary change is produced by the interaction of random mutation, heritability, and Natural Selection.
The theories expressed above can be characterized as general, informal or descriptive theories. The discussion to date has focused on which of the above theories best fits the informal evidence or observations on the operation of evolutionary or biological design processes.
As demonstrated by the discussion, informal analysis can and often does deteriorate into a discussion of the proper way to view or interpret the available evidence. The ‘you have to interpret the evidence my way’ arguments generally involve authoritative rather than objective scientific issues.
Ultimately, formal analysis is the only way for science to address issues of which theory is superior. In order perform formal scientific analysis, it is first necessary to transform a general or informal theory into a formal ‘testable, predictive, mathematical/logical’ theory.
As was at one time widely recognized, translating Darwinian and neo-Darwinian general theories into formal mathematical predictive theories that can withstand formal testing is problematic. [It is easy to translate Darwin and neo-Darwin theories into a simple predictive, mathematical theories, but such interpretations are easily falsified. ] To paraphrase Dembski- Darwinian theory represents a 150 year old unfulfilled promise of a formal mathematical theory of evolution.
While it is not appear to be feasible to develop a valid mathematical model or theory of the weak teleological theory, it is, I suggest, possible to develop valid, mathematical, predictive theories based on the strong teleological theory. As I propose, this can be accomplished if the permanent and universal interpretation of scientific determinism is replaced with the ‘dynamic and teleological’ interpretation. (Permanent and universal causal relationships are a special instance of dynamic and teleological causal relationships).
Proposing a change in the interpretation of the scientific determinism principle raises 1)a philosophy of science issue regarding the legitimacy of the proposed interpretation, and 2)a design science issue of ‘how’ is the interpretation applied to formal testing of evolutionary/biological design theories. I will be glad to discuss either issue if anyone is interested.
IP: Logged
|
|
Frances
Member
Member # 169
|
posted 17. January 2003 11:24
Hi Warren,
Lets try again to get this discussion moving into a direction in which we get details instead of vague termininology.
What's the difference between your 'weak teleological' theories and Darwinism or neo-darwinism? And more importantly how do they allow themselves to be distinguished in the real world. Your repetitive and unsupported claims about Darwinism seem to still remain unsupported. Perhaps it may be helpful if you look into present day evolutionary theory and let us know why you object.
As far as the strong teleological principle, how does it differ other than that you call the solution to be 'front loading'.
So in other words, your 'theories' seem to be nothing different from present day theories, lack in specificity and have a lot of excessive terminology.
So how does your theories compare to the real world and how do they compare to the existing scientific theories?
[ 17. January 2003, 11:25: Message edited by: Frances ]
IP: Logged
|
|
gedanken
Member
Member # 594
|
posted 17. January 2003 14:37
Mark Elkington, Rex Kerr, and myself have asked what I consider to be important and essential questions that relate to the topic of this thread, in the ISCID thread Information creation and transcendence. It is my personal view that the entire foundational concepts of ID are in crisis, because no one is willing to tackle these definitional problems.
For example, the question of “front loading” must depend significantly on the definition of “information” itself, and how such information could be carried through a physical system if it were indeed “front loaded”! If the processes and concepts used in ID of how design occurs and transfers from place to place in the universe is defined in inconsistent and contradictory ways, how can concepts of “front loading” escape these difficulties? Those who are interested in these problems might look in.
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 20. January 2003 14:57
Some way above warren wrote: quote:
The actual data available, however, clearly contradicts this Darwinian prediction. Organisms are fully as capable of generating complex solutions to adaptive problems requiring rapid adaptation as they are of generating complex solutions to long term adaptive problems. Darwinian theory offers no explanation (or a hand wave, we'll figure it out some day explanation).
I would not have responded further had I not run into this reference on antievolution.org. I strongly recommend that warren read it - all of it - before making more claims about what evolutionary theory does and does not explain.
RBH
IP: Logged
|
|
Nel
Member
Member # 614
|
posted 20. January 2003 23:38
Before I reply to Francis' post, I came across an article in nature that is on the web:
http://www.nature.com/cgi-taf/DynaPage.taf?file=/nature/journal/v404/n6779/full/404709a0_fs.html
that might fit well with the idea of front-loading. For example, could the existence of viruses have been anticipated?
The article says:
quote:
When making proteins, cells opt for high speed at the cost of large amounts of waste. But the waste-recycling process has a happy side effect — it gives the immune system a head start in its battle against viral invaders.
quote:
In mammalian cells, however, the waste may have an important function in the fight against viral intruders, as shown by Schubert and colleagues1 and by Reits and co-workers2 on pages 770 and 774 of this issue.
IP: Logged
|
|
Frances
Member
Member # 169
|
posted 21. January 2003 00:33
Nelson
How would this be an example of front loading? How would you establish such front loading? What is exactly front loaded? When do you estimate this front loading happened? Is the front loading unique to certain organisms or can it be traced back to a common ancestor(s)? Was the waste of proteins meant to eventually be used by mammals only? What makes you jump to FL?
Why do you even suggest front loading in this example?
My question really comes down to: How can we distinguish between ad hoc after the fact hypotheses of front loading when we are in fact merely observing that what may have seemed wasteful at first might have evolved through adaptation.
So far we have the following observations: A natural system of protein productions, a wasteful production system, and 'mammalian cells use these waste products to provide advance warning about the proteins currently in production '
[ 21. January 2003, 00:34: Message edited by: Frances ]
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 21. January 2003 10:48
As with any discussion, there are always a number of sub-topics. The primary topic here is the relevance of front loading to theory construction. A interesting sub-topic is the issue of how biologists interpret or misinterpret front loading. Nelson’s quote is a useful starting point- Quote: "In mammalian cells, however, the waste may have an important function in the fight against viral intruders, as shown by Schubert and colleagues1 and by Reits and co-workers2 on pages 770 and 774 of this issue."
In biology, a consistently observed relationship between two events is interpreted as a causal relationship or causal chain. If human genes are consistently observed associated with humans then human genes cause human beings. In the example quoted- the incidental presence of waste products speeds the fight against viral intruders, therefore waste products are front loading factors in the fight against viral intruders.
Biologists tend to forget that they are using massive simplifying assumptions in interpreting these observed relationships as causal relationships. They also tend to forget these simplifications can lead to massive misinterpretations of the observed results.
Nelson is probably correct in identifying waste products as one the physical features which speed up or front load the teleological search for adaptive responses to viral intruders. He is, however, misinterpreting the results by leaving the issue with this observation. We know from computer simulations that ‘finding an adaptive solution to a viral intruder’ must involve extremely complex information processing. Since we also know that teleological problem solving can be modeled in terms of ‘variation-selection’ search processes, we can estimate the complexity of such a process in terms of numbers cycles of variation-selection needed to find a solution. We can similarly estimate (roughly estimate given current knowledge) the amount of savings in information processing or the amount of front loading produced by the presence of waste products.
Using information processing concepts, as opposed to the ‘simplified macro assumptions’ not only makes it possible to quantify front loading, but it makes it possible to demonstrate that front loading is ‘dynamic and teleological’. Front loading evolves, and as a result of the evolution of front loading, the information processing capacity or intelligence of biological systems is increased.
Treating observed macro transformations, for example the transformations from genes to active organisms, as simple deterministic transformations leads, apparently, to much of the problems with existing evolutionary theory. Although the observed transformations may occur with a high degree of consistency, we know from closer examination that what appears on the surface to be a simple transformation is in fact the result of and is dependent on very complex information processing. These very complex information processes are in turn dynamic and teleological- they evolve.
There are obviously applications where it is appropriate and productive to employ simplifying assumptions. It can, however, also be very misleading not to recognize when the use of simplifying assumptions may be seriously distorting results. The treatment of biological information processing by evolutionary biologists is clearly a situation where mixing information processing concepts and macro simplifications has resulted, it appears, in major misinterpretations of the observed facts.
RBH- This is a subject, the mathematical implications of front loading on information processing, on which in theory you should have something useful and constructive to say. Clearly condescending personal attacks, references to literature without explaining relevance, and insistence on conformity to your stylistic preferences are not positive contributions to the discussion.
You seem to have a general problem with how I interpret Darwinian theory. I do at times offer my own interpretations. This is in large measure because supporters of Darwin consistently fail to provide explicit definitions of what theory they are actually supporting. I share with any number of individuals the belief that ‘current Darwinian theory’ is essentially a ‘vague belief system’. Current Darwinian theory is clearly not an explicit predictive mathematical theory. If you are unable or unwilling to explicitly define and share your version of Darwinian theory, then you are welcome to share my interpretations.
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 21. January 2003 15:31
warren,
What I object to is your casual and contemptuous dismissal of the work of hundreds, thousands of very bright, knowledgeable, and innovative people. For example, you say that quote:
This is in large measure because supporters of Darwin consistently fail to provide explicit definitions of what theory they are actually supporting. I share with any number of individuals the belief that 'current Darwinian theory' is essentially a 'vague belief system'. Current Darwinian theory is clearly not an explicit predictive mathematical theory. If you are unable or unwilling to explicitly define and share your version of Darwinian theory, then you are welcome to share my interpretations.
Thank you, but I find yours so at variance with the empirical data and theoretical structure of biology as to provide no useful guidance in either my reading or my research.
With reference to your claim that "Current Darwinian theory is clearly not an explicit predictive mathematical theory," and to your claims of vagueness and handwaving, I will provide just one example (of many that could be cited) to demonstrate the falsity of that claim. This is the description of Math 583/EEB585, "Mathematical Evolutionary Theory," at the University of Tennessee: quote:
There is an enormous variety of different mathematical models and approaches in evolutionary biology. We will consider populations of biological organisms rather than individual organisms or organism's parts. A major emphasis will be on dynamical aspects of evolution (rather than static, statistical, etc.). We will consider simple models (where the word "simple" refers to the model's description and formulation and not necessarily to its analysis). We will focus on models that can answer biological questions, that can provide biological insights, that can train our intuition about complex evolutionary phenomena, can provide a framework for studying such phenomena, and to identify key components of complex biological systems. We will be most interested in models that allow for analytical investigation (rather than pure numerical). We will consider both classical and some very recent approaches. (All emphases original)
You'll notice that the course is cross-listed in the Department of Math and in the Department of Ecology & Evolutionary Biology.
I have no problem with you developing your own approach to biological phenomena from whatever perspective you wish. But I do object when you misrepresent, distort, or ignore what evolutionary theory is, does, and accounts for. Do your own thing, by all means, but don't misrepresent what other people do. If your ideas are worth anything they'll win adherents because of that worth, not because you insult people working in a discipline in which you yourself say you are not an expert. Knowledgable professionals who are experts read the distortions and come to believe you don't know what you're talking about. One would think that's not a perception you want to encourage, particularly as you are attempting to develop a novel approach for which I presume you want a hearing. Insulting others is hardly ever a useful debating technique.
You're correct: I know a dab about evolutionary algorithms, and about the likely effects of "front-loading" of various kinds on the behavior of those algorithms and other kinds of search algorithms. But I'm not interested in participating in the development of an approach that I think both mis-states the problem and is propounded without reference to the substantial amount of work that's been done over the last 150 years. It is simply not the case that biology and several other disciplines have ignored the kinds of "adaptation" you refer to on the time grains you are apparently interested in.
When I was teaching freshmen in college decades ago, I used to tell my classes that there would be little or no class discussion for the first half-year. Only in the second semester, when they had learned enough to speak with some knowledge of the central problems and data and methods of the field would we discuss issues more often. Before that they had nothing coherent to say and it wasn't worth spending class time on. That's not condescending; it is recognizing what one is faced with. So I decline to participate any more in your threads.
RBH
[ 21. January 2003, 15:37: Message edited by: RBH ]
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 22. January 2003 10:46
I read somewhere that search routines based on evolutionary algorithms are sometimes ‘fooled’ by local optima. The search routine, under appropriate conditions, will converge on the optimum within a small segment of search area and fail to find or consider a much better solution or optima located in a more distant part of the search area. It is well known that the same type of problem occurs with human problem solving. RBH is right in suggesting that I am proposing that the solution to the problem of biological design/evolution is well outside the search area that has pursued by scientists for the last 150 years.
I am in fact suggesting that evolutionary biologists and geneticists have been ‘looking in all the wrong places’ for the scientific answer to how and why evolutionary change occurs. Darwin, I suggest, correctly recognized that evolution was the product of biological information processing or ‘the interaction of selection, variation and storage’. For reasons that probably made sense given the technology of his day, Darwin attributed evolutionary change to external or between generation information processing. Given the focus on between generation information processing, scientists failed to recognize and seriously consider the role of within lifetime information processing.
In the world of science that I grew up with, people recognized that the current Darwinian explanation of evolution was far from complete or satisfactory. It was also recognized that there was a very good possibility, even probability that everyone was looking at the problem from the wrong perspective. Apparently in the view of many modern day ‘scientists’ it is unacceptable to address the problem of evolution by approaching the problem from a perspective that might suggest today’s authority figures might have been wrong. If, from this perspective, you don’t accept the local optima, you are not qualified to participate in the search.
The topic of this thread, it is worth recalling, is the claim or suggestion that the flaw which prevents scientists from solving the ‘problem of evolution’ is much older than Darwin. The ‘inappropriate restriction of solution space’ goes back to the time of Newton and Galileo. The problem, I suggest is the assumption that the laws of nature, or the laws of nature subject to scientific analysis, must have the permanent and universal form. I am claiming that 1)the laws governing biological systems have, or can be expressed as having a ‘dynamic and teleological’ form, 2)scientific theories can be formulated for natural laws with the dynamic and teleological form and 3)the scientific paradigm can be applied effectively for such a theories.
RBH is wrong in suggesting I am trying to develop an unconventional approach to biological design. What I am trying to do is find a way to communicate an unconventional approach to people who are insistent that solutions can only exist in the narrow search area in which they are looking.
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 22. January 2003 12:47
warren wrote quote:
RBH is wrong in suggesting I am trying to develop an unconventional approach to biological design. What I am trying to do is find a way to communicate an unconventional approach to people who are insistent that solutions can only exist in the narrow search area in which they are looking.
Then I suggest that as a first step, you accurately characterize the "search area" in which they are working. Mis-characterizing their work and then contemptuously dismissing the mis-characterization is not an effective communication tactic.
I'll tell you a story from my personal experience with exactly the same problem. With a colleague I devised a radically different approach to modeling phenomena in a fairly circumscribed but interesting sub-domain of inquiry involving complex interacting adaptive systems. The approach was different from any I could find in the literature of that field, a field in which I had done no previous academic or professional work. With my colleague, I spent roughly 8 years working out the detailed structure of the modeling technology, applying it to specific questions in the domain of inquiry, testing, revising, testing, revising, etc., etc., etc. Three years of that time I spent working in a research capacity for a company interested in the approach, so I had a three-year apprenticeship inside the very field I was criticizing. I learned from that experience what the professionals in the field actually did, how they did it, and was thus able to accurately describe and knowledgeably criticize their approach. That was essential to my professional credibility in advocating a new approach.
Eight years after starting concentrated work on the new approach I wrote a long technical paper describing current modeling techniques and the problems with them, and sketched our new approach and showed some actual results - data - that demonstrated that this new approach did indeed add something significant to previous approaches that they could not by themselves produce.
I spent nearly a year hunting around for a journal in which to publish, finally found one, worked for another 6 months or so with an editor getting the paper into a form that was both publishable and preserved the core description and data of the new approach. It got published finally, nearly 10 years after we started work on the approach.
Subsequently I was able to publish several more papers in rapid succession responding to critics, elaborating the approach, and drawing implications. This led to funding to continue the work and to develop applications of it in the real commercial world. Now, just six months ago, after something like 12 years of work, the approach began to be used on a large scale in a significant applied context.
The lesson is that if (as you say) you're trying to find a way to communicate an unconventional approach to people who are on what you think is an unproductive 'local maximum,' then (among other things) you have to convince them that you know what you're talking about when you criticize their approach. You have to know the field you're aiming to transform! This doesn't mean "know" informally, casually, or superficially from reading popular books for lay people, but know at an expert or near-expert level. That both increases the credibility of the critique of existing approaches and informs the development of your own approach.
RBH
IP: Logged
|
|
|
Printer-friendly view of this topic