|
Author
|
Topic: Mathematical Impossibility of RM&NS Evolution
|
warren_bergerson
Member
Member # 262
|
posted 14. December 2002 17:03
Daniel, on another thread, suggested that it would be useful to first establish the ‘basis’ for scientific truth before attempting to determine scientific truth. One of the issues associated with determining scientific truth is the type of scientific theory. The validity of the "mathematical impossibility of RM&NS demonstration" depends on the type of theory being considered.
Evolutionary theories involve models or theories of changes in state. A change of state model or theory has the general form F where F(St)=St+l . Three distinct types of scientific theories can be identified based on type of change of state function. A type 1 or non-mathematical, or soft science theory is a theory which identifies factors or variables associated with the change process, but where the identified factors do not translate into an explicit testabke mathematical expressions for F. A type 2 theory, or a ‘physics’ type mathematical theory involves an explicit algebraic expression of the function F.
A type 3 theory is defined in terms of a solution space SS which contains a set of possible functions or algorithms. A type 3 mathematical theory is some defined subset SSt of the solution space SS. The terminology used here may be unfamiliar, but theory expressed in terms of solution spaces describes the general approach used in engineering.
To my knowledge, no one is suggesting that a TOE can be expressed as a type 2 theory. Most commonly, TOE is treated as a type 1 theory. Natural selection and random mutation are general concepts which in some non-mathematical and none precise fashion impact evolutionary change.
It is, however, possible to define both Darwinain and neo-Darwinian evolutionary theories as type theories. Specifically, it is possible to define a solution space SS involving a set of operations including preservation processes(heritability), variance processes, and selection processes. This solution space also requires a number of additional types of processes. A type 3 RM&NS theory is constructed by limiting variance to ‘random variance’ and selection processes to natural or whole organisms selection. As I demonstrated earlier, it appears mathematically impossible for a type 3 neo-Darwinian theory to produce evolutionary change. A Darwinian type 3 theory is defined by limiting selection processes to Natural or whole organism selection.
The demonstration of the mathematical impossibility of RM&NS type evolution is based on neo-Darwinian defined as a type 3 theory. Any counter demonstration also needs to be based on RM&NS precisely defined as a type 3 theory.
The irrelevant references and unsupported claims presented by Frances clearly do not qualify as a counter demonstration.
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 15. December 2002 09:48
I suggested above that Darwinian and neo-Darwinian Theories of Evolution(TOE) can be formally expressed as type 3 predictive, hard science, mathematical theories. Some individuals who appear to object to any explicit formulations of TOE, will undoubtedly find these testable hard science formulations objectionable, but others may find the concept interesting.
As outlined above, a type 3 scientific theory of evolutionary or adaptive changes involves 1)a very precisely defined solution space SS consisting of a set of possible mathematical algorithms of the form F(St)=St+1 which could simulate or predict observed evolutionary/adaptive changes in state, and 2)a precisely defined subset SSt of SS which identifies the set of algorithms proposed as possible by the theory being considered.
It will be noted that a type 3 theory can be refined by both positive evidence (evidence shrinking the size of SSt by demonstrating that what ‘is’ is smaller than what ‘could be’, and by negative evidence(falsification) showing that what actually ‘is’ is outside what is predicted by the current SSt. Using type 3 theories, scientific research is an ongoing process of refining the size and shape of the current theory space SSt.
Darwinian theory of evolutionary or adaptive change, it appears, has its origins in the Greek concept of teleological or purposeful change. Purposeful change, the ancient Greeks suggested, is produced by the interaction of three types of processes- preservation processes, variance processes and selection processes. [Actually modeling or simulating teleological change requires far more than three types of processes. My simplified or idealized model of teleological change involves 200 plus types of processes or operations.]
For the sake of discussion here, we can use the simplifying assumption that a solution space SS for evolutionary/adaptive change is defined by the three processes -preservation, variance, and selection. Starting with this solution space, the Darwinian TOE can be defined as the subspace defined by 1)limiting ‘preservation’ to heritability (preservation between generations) and 2)limiting selection to Darwinian Natural Selection. The neo-Darwin TOE is defined by adding the additional requirement that variance is limited to random variance. (And by suggesting that evolutionary change to defined and measured or quantified in terms of genetic change. )
Although not always recognized as such, Darwinian and neo-Darwinian theory represent very severe restrictions on the mechanisms of evolutionary change. In particular Darwinian and neo-Darwinian theory suggest that with respect to:
1. PRESERVATION- The heritability restrictions suggests that the adaptive or teleological states responsible for the survival of an organism are preserved in genetic code and passed from generation to generation. This excludes from evolutionary theory adaptive changes which are reproduced or recreated within each generation. Although it is often glossed over in the literature, preserving or inheriting ‘the ability to generate and find adaptive states’ is materially and fundamentally different from ‘preserving and inheriting adaptive states’. The heritability component of Darwinian theory excludes consideration of ‘within life-time evolutionary/adaptive change’.
2. SELECTION- There are many powerful forms of selection in addition to Darwinian Natural Selection. Specifically, there are types of selection operating directly on sub-processes and mechanisms rather than on the organisms as a whole. Limiting selection to Natural Selection severely limits the potential speed of evolutionary/adaptive change.
3. VARIANCE- The alternative to random variance is directed variance. As discussed earlier, the combination of a large search space, random variations, inter-generation heritability and whole organism or Natural Selection would make evolutionary change mathematically impossible.
SUMMARY
It has been argued that evolutionary change processes are too complex to express as type 2 or physics type scientific theories (it seems likely that physics is to complex to reduce to a type 2 theory). It does, however, appear feasible to express scientific theories of evolution as what I call type 3 theories. If, as outlined above, Darwinian and neo-Darwinian theories are expressed as type 3 mathematical theories, then it appears likely that both types of theories will be falsified. The Darwinian and neo-Darwinian restrictions on preservation processes, selection processes and variation processes all seem to be overly restrictive.
IP: Logged
|
|
Evan
Member
Member # 164
|
posted 15. December 2002 10:33
Given that, as the moderator has pointed out, much of what Warren offers has been offered in the past, I will keep my comments short:
1) Warren writes that the type of theory he proposes “involves 1)a very precisely defined solution space SS consisting of a set of possible mathematical algorithms of the form F(St)=St+1 which could simulate or predict observed evolutionary/adaptive changes in state”
My comment is that this “precisely defined” solution space has not in fact been defined at all in biological terms. So step 1 is still missing.
2) Warren writes, “My simplified or idealized model of teleological change involves 200 plus types of processes or operations.”
I would be interested in seeing a sample of these - a dozen or so would suffice, because I don’t know exactly what Warren is talking about.
3)Warren writes, “The heritability component of Darwinian theory excludes consideration of ‘within life-time evolutionary/adaptive change’”
The reason for this is that no evidence has been found that in fact what an organism learns during its lifetime is passed on to the next generation. This is where actual scientific investigation comes in, as opposed to theorizing. This aspect of evolutionary change is rejected not because of any dogmatic adherence to a particular view, but because it has been investigated and found to lack supporting evidence.
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 16. December 2002 15:30
Evan
Quote: My comment is that this "precisely defined" solution space has not in fact been defined at all in biological terms. So step 1 is still missing.
You need to differentiate between the solution space relevant to the ‘mathematical impossibility of RM&NS’ which is well defined, and the solution space relevant to all biological information processing which, I suggest can be precisely defined, but for which a precise definition has not yet been offered.
[Note: A solution space is a set of functions or algorithms satisfying some criteria. It is a fairly trivial action to precisely define such a set. The challenge is not is defining a solution space, but it find a space that leads to useful models and theories. ]
To repeat if you missed it, solution space for the RM&NS impossibility analysis is the set of search functions which search a space with N options to find one adaptive solution using processes involving some type of selection process and some type of variance process. That is a precisely defined set of mathematical functions.
As you are aware, this type of search function or routine has been studied extensively. Since you claim to be knowledgeable in this area, you must be aware that for functions relying on natural selection, the search speed slows down as the size of N increases. The mathematical impossibility of RM&NS is based on extrapolating this well known fact to systems which have levels of complexity approaching the complexity of real world biological systems. Since you have raised no objection to the mathematical conclusion presented, can I imply you are in agreement?
As I clearly stated, I offered a demonstration of the mathematical impossibility of RM&NS type evolution in very complex systems. (The only existing demonstrations that RM&NS evolution works involve relatively simple systems.) If someone can offer a counter demonstration showing RM&NS can in fact work with for very complex systems, then there would be a basis for claiming that RM&NS works for very complex biological systems.
To make the claim that RM&NS works in real world biological systems, but is a mathematical impossibility in precisely defined mathematical systems, would be an ‘interesting’ scientific stance.
As I discussed yesterday, RM&NS impossibility demonstration has a number of interesting implications and applications. It does not, however, seem appropriate to address more complex issues until there is some consensus on the basic issues raised.
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 16. December 2002 18:37
Addressing Evan, warren wrote quote:
To repeat if you missed it, solution space for the RM&NS impossibility analysis is the set of search functions which search a space with N options to find one adaptive solution using processes involving some type of selection process and some type of variance process. That is a precisely defined set of mathematical functions.
As you are aware, this type of search function or routine has been studied extensively. Since you claim to be knowledgeable in this area, you must be aware that for functions relying on natural selection, the search speed slows down as the size of N increases. The mathematical impossibility of RM&NS is based on extrapolating this well known fact to systems which have levels of complexity approaching the complexity of real world biological systems. Since you have raised no objection to the mathematical conclusion presented, can I imply (sic) you are in agreement? (emphases added)
I see two problems with this analysis, indicated by the italicized sections in the quotation above. Both problems are symptomatic of a deeper misconception of evolution by random mutation and natural selection that is embodied in warren's analysis.
In the first italicized phrase "search a space with N options to find one adaptive solution," there is a presumption that the process is in search of a single solution. It is not. It is not in search of a solution at all, say nothing of "one adaptive solution." Evolution by RM&NS is a local adaptation process where the population is dispersed more-or-less widely over some area on the surface of a (high-dimensioned) fitness landscape. Through differential reproduction as a function of relative fitness the population 'moves' on the landscape, migrating to (not "toward"!) nearby higher levels on the topography of the landscape. In that process evolution often (but by no means always) "finds" better solutions, but it does so as a by-product of differential reproduction as a function of relative fitness, not as a goal.
The second phrase, "the search speed slows down as the size of N increases," is meaningless without detailed ceteris paribus clauses that I have not yet seen spelled out. Those clauses have to refer to assumptions about rates of introduction and dispersal of variability in the population, the density of acceptable solutions in the search space, the probabilistic topography of the search space, the size of the population, and so on. It also depends on the assumption of search for "one adaptive solution," which is a false assumption.
The underlying misconception is touched on above. Evolution by RM&NS, while it is often analyzed as a "search" process, is not in fact a search process. In many applications in engineering and elsewhere EAs are employed as search algorithms, but in biology evolution is not a search "for" something: It is a natural process that has as a side-effect the (occasional) finding of adaptive solutions (plural) to challenges posed by varying selective environments. To model evolution by RM&NS as a search process with "one adaptive solution" as its goal is to misrepresent it, and therefore I see no reason to believe that the "mathematical conclusion" tells us anything at all about the possibility or lack thereof of evolution by random mutation and natural selection in biology.
RBH
[ 16. December 2002, 19:01: Message edited by: RBH ]
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 17. December 2002 13:01
RBH,
Quote: In the first italicized phrase "search a space with N options to find one adaptive solution," there is a presumption that the process is in search of a single solution.
The search for one solution requirement is a assumption used to simplify the mathematics in evaluating the adequacy of a RM&NS search processes. You get the same result with a lot more work if you assume that solutions involve sets of sets solutions within design tolerances. The demonstration presented here was based on the simplifying assumption that the set of possible assembly instruction which satisfy design tolerances can be treated as a single option. It is reasonably obvious that the conclusion reached is not impacted by the use of this assumption. You are welcome to demonstrate that a different result would be obtained using continuous assumptions.
Quote: The underlying misconception is touched on above. Evolution by RM&NS, while it is often analyzed as a "search" process, is not in fact a search process.
As was clearly defined, the demonstration presented is based on a common type of evolutionary change which could be labeled adaptive redesign. This is the commonly occurring phenomena where the physical shape of a feature such as a wing or beak or internal organ is redesigned and where the range of possible redesigns can be identified with a reasonable degree of accuracy. As was also discussed in detail, changes in these phenotype features are modeled as changes in assembly or design implementation processes. The type of evolutionary change being discussed can legitimately be described and modeled as a search process. Again, you are welcome to demonstrate that there is method of modeling evolutionary changes which 1)does not use search process models, 2)which incorporates the criteria applicable to the demonstration performed, and 3)which produce different results.
Since RM&NS models purport to describe all types of evolutionary change, RM&NS can be said to describe those types of evolutionary change which can be modeled and analyzed as search processes.
Quote: To model evolution by RM&NS as a search process with "one adaptive solution" as its goal is to misrepresent it, and therefore I see no reason to believe that the "mathematical conclusion" tells us anything at all about the possibility or lack thereof of evolution by random mutation and natural selection in biology.
This is the type of statement, I suggested yesterday, that no serious scientist would be willing to make. You are saying, if effect, that you can arbitrarily and subjectively reject and ignore mathematical analysis because it does not satisfy some arbitrary and non-substantive formatting requirements. You then suggest you are justified in making the claim that the mathematical/logical process RM&NS works in biology even though 1)there are demonstrations it does not work in mathematics, and 2)there is no demonstration that it will work under the conditions defined.
You are in effect saying that you can use your own personal subjective requirements as a basis for rejecting mathematical and scientific findings. Such behavior may be acceptable in evolutionary biology (although I doubt it), but it is not acceptable in design science.
Quote: The second phrase, "the search speed slows down as the size of N increases," is meaningless without detailed ceteris paribus clauses that I have not yet seen spelled out.
This statement is very misleading. It suggests that you are rejecting the ‘speed slows down with increasing N’ for a legitimate mathematical reason, when, in fact, you are rejecting a well known and highly relevant observed phenomena based on an arbitrary and irrelevant basis.
To begin, as a mathematician who has worked around search routines including RM&NS routines, you are know that ‘the time required to complete a search increases as the size of the search area increases’. Even non-mathematicians know that in general it would take longer to find a ‘needle in a haystack’ than to find ‘an elephant in a pick-up truck’. As was pointed out, there are other factors involved, but the general rule that ‘search speed slows down as the size of N increases’ is obviously valid whether from a mathematical or common sense basis. To reject or ignore the obvious because of some trivial and irrelevant technicality is, IMO, game playing.
If you recognize the obvious relationship between complexity and speed, then you must also recognize that some search routines will become mathematically impossible given 1)certain sizes of search areas and 2)certain limitations on search times. In other words, you must recognize that the test defined here is a mathematically/scientifically valid test of both RM&NS and Natural Selection models and theories.
SOME GENERAL COMMENTS
The ‘mathematical impossibility of RM&NS evolution’ conclusion is based on design science analysis which differs in a number of key respects from traditional analysis used in evolutionary biology. To understand why or how design science can produce results which appear to directly contradict evolutionary science results it is useful to note the following key differences:
1. OBJECT OF EVOLUTION- TELEOLOGICAL INFORMATION VERSUS CODES
Most current theories assume that evolutionary processes operate on sets or strings of codes. A strand of DNA is an example of a set which can be readily translated into a mathematical code. Design science by contrast, assumes that evolutionary processes operate on teleological causal relationships. An assembly instruction would be an example of such a causal relationship. Most basic concepts of information can be defined using either ‘code concept’ or the ‘teleological relationship concept’.
As I see it, the key difference between two approaches is that the code concept of biological requires ‘assuming’ the existence of process or mechanism for translating from code to real world traits( the assumption of a genotype-phenotype map process, for example). The teleological causation concept defines biological information in terms of actions and does not require assuming a translation process(the definition includes the translation). The teleological causation concept, as can be demonstrated, provides an observable measurable genotype to phenotype mapping mechanism. [This might in itself be an interesting topic for discussion.]
2. TYPE 3 VERSUS TYPE 1 THEORIES
As briefly introduced earlier, design science analysis is based on formulated scientific theories as type 3 mathematical theories. Evolutionary biology generally expresses theories as type 1 or non-mathematical theories. As has been discussed, both Darwinian and neo-Darwinian theories can be expressed as type 3 theories. The use of type 3 theories means that it is feasible and practical to apply much more rigorous scientific and mathematical standards in performing analysis.
3. CAUSAL DENSITY
Design science is based on a mathematical concept of causation which is quite different from the concept which appears to be used in evolutionary biology. The conceptualization of causation implied in most analysis suggests that the universe, or changes in the universe, are controlled by a relatively limited number of causal relationships or laws. The purpose of science, from this perspective, is to discover the laws or causal relationships or laws governing the universe.
In design science, the operation universe is viewed as being controlled or described by an infinitely dense set of causal relationships. The role of science, from this perspective, is to find some small subset of this dense set which 1)can be subjected to scientific testing, 2)which produce reliable predictions, and 3)which are useful in solving practical problems.
The causally dense universe perspective is compatible with the analysis of complex causation, because complex causal relationships are defined in terms of the interactions of very large sets of simpler causal relationships where the component relationships take a number of different forms.
CONCLUSION
In theory math is math and science is science. A conclusion reached in design science or design science mathematics should not be incompatible with conclusions reached in evolutionary biology or the fields of mathematics associated with evolutionary biology.
Design science clearly represents a very different perspective from which to view and analysis biological design processes. Design science produces results which appear to be both 1)logically and mathematically sound and 2)in direct conflict with conclusions reached in evolutionary biology. Interesting.
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 17. December 2002 14:13
warren wrote quote:
This is the type of statement, I suggested yesterday, that no serious scientist would be willing to make. You are saying, if effect, that you can arbitrarily and subjectively reject and ignore mathematical analysis because it does not satisfy some arbitrary and non-substantive formatting requirements. You then suggest you are justified in making the claim that the mathematical/logical process RM&NS works in biology even though 1)there are demonstrations it does not work in mathematics, and 2)there is no demonstration that it will work under the conditions defined.
My point was very simple: If a mathematical model does not veridically represent the entities, processes, and relationships that are defined as relevant by a substantive theory of a domain of inquiry, then conclusions based on the the model's behavior do not answer questions about the substantive theory. My observation is that by representing biological evolution as a process in search of "one adaptive solution," warren's model misrepresents the substantive theory of interest and thus the behavior of the model is uninformative with respect to that theory. There is nothing arbitrary or subjective about making the assessment; one looks at the way the terms and operators of the math are mapped into the entities, processes, and relationships of the theory. If the math appropriately maps the territory, then it can be useful in generating and evaluating hypotheses about the substantive domain of interest. If the mapping is invalid then the math is uninformative (and potentially misleading) about the substantive domain. The issue is not the validity of the math, it is the veridicality of the mapping of the math onto theory and substance. Put in warren's terms, "the conditions defined" do not represent evolutionary theory, and thus his model is irrelevant to that theory.
Warren also misreads my remarks about the necessary ceteris paribus clauses. For example, one such clause must posit that the number of needles in a haystack is a constant regardless of the size of the haystack. But in biology that may well be false: the proportion of adaptive targets may be constant over search space size and in that case search time will not scale linearly with search space size. To claim that one has mathematically disproven evolutionary theory and to assert that evolution by random mutation and natural selection is in principle an inadequate explanation of the observed biological phenomena is also to claim to have adequately considered such questions. I see no evidence in his postings that warren has adequately considered them and he has provided no reasons to disregard them.
RBH
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 18. December 2002 13:57
RBH raises three issues which needs to be addressed. These issues are 1)mathematical validation, 2)scientific validation, and 3)biological reality.
MATHEMATICAL VALIDATION
The issue in question can be expressed as
1. Evolutionary biology makes the claim that evolutionary change can be produced by RM&NS type processes.
2. I offered a demonstration that under specified conditions RM&NS type process would be to slow to produce evolutionary change.
3. No one has offered a counter demonstration showing that an RM&NS process could produce evolutionary change under the defined conditions.
ISSUE: Is it sound or acceptable mathematics to claim that 1 is a mathematically sound and valid assertion even a)no one has demonstrated a mathematical flaw in 2 and statement 3 is valid. In terms of simple logic, can it be asserted that (2+2=5) is mathematically sound if there is a demonstration that (2+2=4 (and not 5)) and there is no demonstration that (2+2=5).
As far as I can determine, RBH is not questioning the mathematical argument that is the basis for this thread. He is not questioning that there is a demonstration(not a proof) that RM&NS type processes appear incapable of producing evolutionary change under the conditions defined. He also, as far as I can determine, is not questioning that it would by mathematical standards, be improper and misleading to make the mathematical claim that ‘evolutionary change can be produced by an RM&NS process’ if there is a demonstration to the contrary and no demonstration supporting evolution by RM&NS under the defined conditions.
SCIENTIFIC VALIDATION
If the ‘lack of a mathematical demonstration supporting the possibility of RM&NS evolution is accepted as mathematically valid we can now look at the question of scientific validation. The scientific issue comes down to:
1. Evolutionary biology makes the ‘scientific’ claim that evolutionary change can be produced by RM&NS type processes.
2. Item 2 from the mathematical validation above.
3. I offer a set of precise definitions showing that for the model in item 2 can be used to model certain identifiable types of evolutionary change.
4. No argument or evidence is offered showing that the particular definitions used are invalid or unsound.
5. There is no alternative set of definitions offered which make it possible to demonstrate that RM&NS type evolution can occur under the conditions defined.
ISSUE: Is it scientifically sound to continue to claim 1 is a valid scientific claim or hypothesis given a)a demonstration of a set of conditions when 1 does not appear to be valid and b)an absence of alternative definitions and demonstrations that 1 is valid?
The argument/evidence provided would seem to 1)provide a classic falsification demonstration for RM&NS and 2)support for the alternative hypothesis that RM&NS can not, in most instances, produce evolutionary change in biological systems. It will be noted that falsification does not require ‘proving’ that RM&NS is impossible. Falsification simply requires a demonstration that RM&NS does not work under certain conditions (and arguments addressing any counter-arguments.
Not to be naïve, we know that in practice many individuals, before accepting a falsification argument will wait until the subject has been discussed. We also know that individuals will reject falsification demonstrations for personal, subjective reasons. We also know that it is not unheard of for peer review groups to reject falsification arguments for subjective reasons. That is not the issue here.
The issue here is whether a ‘scientist’ would be willing to knowingly and intentionally reject a falsification argument for what are known to be arbitrary, capricious, irrelevant, and subjective reasons. Your ‘search for 1 solution is invalid’ argument, as I pointed out, comes under the heading of irrelevant and subjective. IMO, a scientist would not reject a falsification argument on subjective grounds, if such a rejection was perceived as involving a significant risk of getting caught.
3. BIOLOGICAL REALITY
There is a general suggestion that the demonstrations offered here are not valid because they do not reflect biological reality. There can be no doubt that ‘assembly instructions’ are part of biological reality. This assembly instructions perspective or visualization is compatible with known facts and is thus a biological reality.
Some individuals appear to be using the concept of ‘biological reality’ as a euphemism for ‘viewed in terms of generally accepted concepts and terminology’. It will be recognized that science requires only that a subject be viewed from a perspective consistent with the facts. There is no scientific requirement that a subject be viewed from a single perspective imposed by some group of authority figures.
It is highly questionable whether the generally accepted perspective for viewing evolutionary change is consistent with the facts and thus reflects biological reality. Is, for example, there any reality to the assumed phenotype to genotype and genotype to phenotype maps. Is there any reality to the claim that ‘genes determine the phenotype of an organism’. As it stands today, there is no evidence that these assumed phenomena are any more real than Santa Claus or the Tooth Fairy.
SUMMARY
Just as a reminder, this thread introduced to discuss the hypothesis that ‘given the level of complexity present in biological systems, it is not possible for evolutionary change to be produced by an RM&NS type process’. This hypothesis is an extrapolation of the common sense observation that ‘if you search for a needle in a haystack by looking at each particle one at a time, then the bigger the haystack, the longer it will be expected to take to find the needle’. Using the ‘assembly instruction’ perspective, I showed the level of complexity in biological systems is such as to appear to preclude the possibility of RM&NS type evolution. No one has yet to offer an explanation of why the argument presented is not valid. No one has yet developed a demonstration showing how RM&NS type processes could produce evolutionary change under the conditions defined.
As I have said on a number of occasions, RM&NS is in itself not a very interesting topic. It does, however, provide an opportunity to introduce a number of subjects which might otherwise be considered irrelevant. The ‘assembly instruction’ or ‘biological information processing’ model provides some interesting and potentially useful techniques for analyzing biological design. The engineering concept of type three theories appears to provide a practical method of formulating and actually testing mathematical theories. Finally, the simple and easily understood mathematics used in this demonstration should make easier for more people to evaluate the flaws in some of the widely used mathematical arguments supporting neo-Darwinian evolution.
IP: Logged
|
|
Frances
Member
Member # 169
|
posted 18. December 2002 14:53
Warren continues his irrelevant claims
quote:
1. Evolutionary biology makes the claim that evolutionary change can be produced by RM&NS type processes.
2. I offered a demonstration that under specified conditions RM&NS type process would be to slow to produce evolutionary change.
3. No one has offered a counter demonstration showing that an RM&NS process could produce evolutionary change under the defined conditions.
Warren has failed to show that his RM&NS process has any relevance to evolution. I and others have shown how evolutionary processes applied in a scientific manner have shown to be quite able to produce evolutionary change
CONCLUSION
Warren's hypothesis is faulty aka a STRAWMAN
[ 18. December 2002, 14:54: Message edited by: Frances ]
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 18. December 2002 15:56
I doubt that there's anything more to be usefully said here. I'm accustomed to a mode of doing science in which theories have contact with data, systematically gathered in field or laboratory, and in which mathematical models are evaluated in the light of the veridicality of their mapping of the theories they are to model and the match between the model's behavior and the data. I see neither in warren's "design science." Even as a brainstorm it seems peculiarly bereft of any contact with reality.
RBH
IP: Logged
|
|
Moderator
Administrator
Member # 1
|
posted 18. December 2002 17:01
I'm afraid that this thread is not making any progress. I'm going to close it.
Two warnings:
1. Warren: remember, this is a brainstorming forum and therefore we are not concerned about the impossibility of things. Nor are we interested in assertions of proof/disproof. Such assertions only serve to provoke others and to constrain the discussion.
2. Frances: I've warned you before about this. Your continual battle with Warren shows me that you have in interst in governing our discussion board and setting the record straight. Though this is sometimes legitimate, you've gone overboard. Only critique somebody when you are interested in helping with their argument - not with short, pithy remarks that belittle. Otherwise, ignore and contribute elsewhere.
IP: Logged
|
|
|
Topic Closed
Printer-friendly view of this topic