|
Author
|
Topic: Wells' Molecular Phylogenies
|
Josh
Member
Member # 405
|
posted 24. October 2003 12:05
I think Wells' critique of molecular phylogenies will be much harder to sustain after papers like this (Nature 425, 798 - 804 (23 October 2003):
Genome-scale approaches to resolving incongruence in molecular phylogenies
ANTONIS ROKAS*, BARRY L. WILLIAMS*, NICOLE KING & SEAN B. CARROLL
Howard Hughes Medical Institute, Laboratory of Molecular Biology, R. M. Bock Laboratories, University of Wisconsin-Madison, 1525 Linden Drive, Madison, Wisconsin 53706, USA
* These authors contributed equally to this work
Correspondence and requests for materials should be addressed to S.B.C. (sbcarrol@wisc.edu).
One of the most pervasive challenges in molecular phylogenetics is the incongruence between phylogenies obtained using different data sets, such as individual genes. To systematically investigate the degree of incongruence, and potential methods for resolving it, we screened the genome sequences of eight yeast species and selected 106 widely distributed orthologous genes for phylogenetic analyses, singly and by concatenation. Our results suggest that data sets consisting of single or a small number of concatenated genes have a significant probability of supporting conflicting topologies. By contrast, analyses of the entire data set of concatenated genes yielded a single, fully resolved species tree with maximum support. Comparable results were obtained with a concatenation of a minimum of 20 genes; substantially more genes than commonly used but a small fraction of any genome. These results have important implications for resolving branches of the tree of life.
IP: Logged
|
|
Cornelius G. Hunter
Member
Member # 81
|
posted 10. November 2003 12:18
Josh:
Imagine that a physicist hypothesizes that the gravitational force goes as 1/R^2. In experiments, it usually does, but sometimes goes as 1/R. He then argues that because those 1/R cases are rare, that they don't matter. That doesn't follow.
The question is this: why are congruent phylogenies strong evidence for evolution when you have so many inconsistincies which require just-so stories? Yes, most traits line up in a consistent phylogeny. But there are many anomalies, and they are not explained as the result of the normal evolutionary process. Instead, we need explanations such as large amounts of HGT or evolutionary convergence. These are the best explanations under evolution. But they are not compelling, they merely explain away problems. This moves phylogeny from the "strong evidence" category to the "evidence that has problems which can be accommodated" category.
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 10. November 2003 17:16
Cornelius Hunter wrote quote:
Imagine that a physicist hypothesizes that the gravitational force goes as 1/R^2. In experiments, it usually does, but sometimes goes as 1/R. He then argues that because those 1/R cases are rare, that they don't matter. That doesn't follow.
Hunter's analogy is a red herring. That is not the argument that the cited paper makes. The Roka, et al., paper directly contradicts the criticism Hunter raised. (The complete paper is here). quote:
By contrast, analyses of the entire data set of concatenated genes yielded a single, fully resolved species tree with maximum support. Comparable results were obtained with a concatenation of a minimum of 20 genes; substantially more genes than commonly used but a small fraction of any genome. These results have important implications for resolving branches of the tree of life. (p. 798)
Hunter's analogy with the measurement of gravity, to the extent that it is a worthwhile analogy at all, has to do with the power of the instrumentation used to measure gravity and with ways of eliminating noise in measurements. The Rokas, et al. paper speaks to the conditions necessary to increase the power of the 'instrumentation' used to 'measure' (reconstruct) phylogenetic trees, and how that increased power reduces the variability (noise) of those 'measurements': quote:
Although many potential options exist, we explored the effect of concatenating single genes into one large data set (1,27,39). Remarkably, all three methods of analysis of the concatenated sequences yielded a single tree with 100% bootstrap values at every branch (Fig. 4). Furthermore, all alternative topologies generated among the single-gene analyses were rejected (Templeton test, P (lessthan) 0.001 for each of three analyses). Thus, even though the individual genes examined supported alternative trees, the concatenated data exclusively supported a single tree. This level of support for a single tree with five internal branches is, to our knowledge, unprecedented; we conclude that it accurately represents the historical relationships of these eight yeast taxa and will be referred to hereafter as their species tree. The maximum support for a single topology regardless of method of analysis is strongly suggestive of the power of large data sets in overcoming the incongruence present in single-gene analyses. (p. 800)
Far from merely explaining away problems, as Hunter asserts, the cited paper shows how the problem dissolves when the 'measuring apparatus' is improved. Hunter's final sentence was quote:
This moves phylogeny from the "strong evidence" category to the "evidence that has problems which can be accommodated" category.
On the contrary, the findings of the cited paper solve the problems, thereby strengthening the phylogenetic evidence. The anomalies are shown to be artifacts of incomplete data sets, not intrinsic properties of the phenomena being measured. It is necessary that the anomalies be 'real' in order for Hunter's critique to hold, but Rokas, et al. show that the anomalies are an artifact of inadequate data sets.
RBH
[ 10. November 2003, 17:28: Message edited by: RBH ]
IP: Logged
|
|
Cornelius G. Hunter
Member
Member # 81
|
posted 10. November 2003 17:45
RBH:
I'm happy to discuss issue with anyone interested in learning something and exploring the issue, but I won't be responding to personal attacks and invectives that mischaracterize my point as well as the issue.
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 10. November 2003 18:24
Cornelius Hunter wrote quote:
RBH:
I'm happy to discuss issue with anyone interested in learning something and exploring the issue, but I won't be responding to personal attacks and invectives that mischaracterize my point as well as the issue.
I've reread my posting several times. I see no "personal attacks" nor any "invective." I see statements taking strong issue with Hunter's physics analogy and with Hunter's apparent interpretation of the cited paper as merely another instance of "accommodating" known problems with reconstructed phylogenies. (Was Hunter not addressing that paper?) Nowhere do I see an attack on Hunter himself nor do I see any invective.
Perhaps some observer would be good enough to point out what in my posting constitutes a "personal attack" and/or "invective," as distinguished from disagreement, sometimes phrased firmly, with statements, claims, assertions, and interpretations.
RBH
[ 10. November 2003, 18:29: Message edited by: RBH ]
IP: Logged
|
|
Mesk
Member
Member # 630
|
posted 11. November 2003 01:43
RBH,
I also fail to see any sign of personal attacks in your post. All of your criticisms are clearly directed at Hunter's arguments, and not at him personally. Either Hunter is extraordinarily thin-skinned, or he was simply looking for an excuse to escape the thread - perhaps not an entirely bad idea, given the weakness of his argument.
Mesk.
IP: Logged
|
|
Cornelius G. Hunter
Member
Member # 81
|
posted 11. November 2003 02:49
Mesk:
Can you explain why you think my point is weak?
--Cornelius
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 11. November 2003 13:23
Not to anticipate anything Mesk might say in response to Cornelius Hunter, let me expand a bit on my critique of Hunter's gravity analogy. He wrote quote:
Imagine that a physicist hypothesizes that the gravitational force goes as 1/R^2. In experiments, it usually does, but sometimes goes as 1/R. He then argues that because those 1/R cases are rare, that they don't matter. That doesn't follow.
Suppose in the physicist's experiments he finds that when he measures the force of gravity operating on a lead shot in the earth's gravitational field by measuring its acceleration in free fall in an open space in his laboratory, he finds that the gravitational force goes as 1/R^2. However, when measuring it for a feather using the same experimental procedure, he finds that it goes by 1/R. Does it somehow weaken his case for 1/R^2 to observe that air resistance differentially affects the lead shot and feather, which affects his measurement of acceleration and hence his estimate of the effect of gravity as a function of R, and to explain the anomalous measurement with the feather by invoking air resistance? Is that to merely "explain away" problems rather than genuinely explain them?
Note that the air resistance explanation implies testable hypotheses; it's not merely some sort of ad hoc "accommodation" to discrepant results. And it identifies conditions under which the original measurement procedure does not produce reliable results.
The same logic (and an analogous research program) applies to the biological example. Invoking horizontal gene transfer, for example, is not to "explain away" a discrepant finding. It is to identify a genuine explanation for the discrepancy. And it is a testable explanation. For example, it implies a search for methods to detect how horizontal gene transfer might distort the reconstruction of phylogenetic trees and how those distortions can be taken into account, as in this paper: quote:
Abstract
A new approach for comparative analysis of multiple trees reconstructed for representative protein families is proposed. This approach is based on the hypothesis of gene duplication, gene loss and horizontal gene transfer and makes use of stochastic methods and optimization. We present a species tree of 40 prokaryotic organisms obtained by our algorithm on the basis of 132 clusters of orthologous groups of proteins (COGs) from the GenBank of the National Center for Biotechnology Information (USA). We also present a computer technology intended to determine horizontally transferred genes. Some application results of the technology, based on comparative analysis of protein and species trees, are given.
And from the Conclusion: quote:
Conclusion
A well known method of comparing the gene and species trees based on simple counting of the number of duplications and losses was modified. Our method considers the lengths of the edges that define duplication or loss events. The local minimization algorithm of the nearest neighbor branch swapping supplied with a module generating random initial species trees yields species trees similar to the best known trees. When special a priori probability distribution over the trees, computed in advance, is used to generate random initial trees, the resulting consensus tree remains robust under different long runs of the algorithm. These results show that our model of gene duplication and gene loss and the corresponding algorithm present a sufficiently adequate tool for tree comparison and can be used for the detection of horizontal gene transfer. (Emphasis added)
I make no claims about the authors' methodology beyond the fact that it represents genuine research on the problem that Hunter's analogy implies is claimed by biologists to not matter. Clearly biologists think it matters, and equally clearly there is research aimed at resolving it.
Similarly, the Rokas, et al. paper cited in the OP demonstrates an approach to overcoming the problem of discrepancies in reconstructed phylogenies by using sufficiently large numbers of genes to do the reconstruction. It shows that doing so can resolve discrepancies among phylogenies reconstructed from one or two genes. It is (very roughly!) analogous to the physicist measuring the behavior of a large number of different objects in a 1-g field and finding that in the aggregate, the measurements show that gravity varies approximately as 1/R^2 and that one can statistically describe the error of estimate.
Again, Hunter's analogy is not a good one. A slightly better analogy (though still not completely parallel) would be the case in which a microscopist increased the power of his instrument and found that he could now resolve details of a tiny structure that were blurred in his previous lower-powered instrument. With that increased resolution he can test hypotheses about those details.
RBH
[ 11. November 2003, 14:15: Message edited by: RBH ]
IP: Logged
|
|
Cornelius G. Hunter
Member
Member # 81
|
posted 11. November 2003 23:12
RBH:
Evolution predicts convergent phylogenies. Often one finds a partial convergence where the majority of the characters line up but there are some that don't. Sometimes it is worse, with little or no convergence observed. To explain the non convergent data, evolution has a variety of biological and analytical mechanisms to draw on. The former include mechanisms such as convergence on a macro scale, massive HGT, massive gene loss, high evolutionary rate, and non uniform evolutionary rate (of various sorts). Some of these are not observed empirically. The latter include mechanisms such as sample size too small, sample size too big, incorrect analytical method etc. All of these mechanisms are used in some instance, and no single mechanism is sufficient to explain all the inconsistencies.
The fact that there are many non convergent characters and many explanatory mechanisms suggests the possibility that a reliable, trustworthy phylogeny may not always be obtainable. In fact, it has been found that in some cases increasing the sample size did not help, but brought on problems. This particular problem of bigger samples not seeming to converge seemed resolvable (at least to me) but interesting. It would have been awkward (though not unexplainable) for evolution if ever-increasing samples did not provide for a convergence to a single phylogeny.
Now the findings in the Rokas paper are that, for the particular yeast data set they assembled, a large sample does provide for convergence. This does not mean there are no non convergent characters. There are. It means that those characters were not very significant in the analysis. That is, assuming evolution is true, the resulting estimated phylogeny was of high probability. The non convergent characters are still there, and the various explanatory mechanisms are still used. But because they were swamped by the agreeable data, they did not cause the estimated phylogeny to be of lesser probability. I do not see how this has anything to do with instrument power. The greater "power" added by the Rokas analysis does nothing to further explain the anomalies.
Phylogenetic convergence is claimed to be powerful evidence for evolution. The problem with this claim is not that there is an extreme lack of convergence. Though there are many non convergent characters, the majority of the characters do tend to agree. The problem is that those non convergent characters need to be explained, and the explanations require mechanisms that are not empirically observed. Indeed, one could just as easily argue that the lack of convergence makes this evidential claim serve to falsify evolution. There are non convergent characters that require explanatory mechanisms that have never been observed. One could argue this turns the evidence against evolution. But at the very least, as I see it, we must admit the evidence is not strongly in favor of evolution.
I tried to illustrate this with my gravity analogy. The analogy was not meant to deceive, it was meant to illustrate the problem that anomalous, even if rare, observations need to be explained. Increasing the sample size with agreeable observations, to the point that the anomalous observations are driven into the noise, does not resolve the issue.
This is not a criticism of the Rokas paper as you seem to have assumed. The Rokas paper is operating within the evolution program. It is presupposing evolution and addressing a technical problem that occurs within that program (namely, how best to recover the true historical phylogeny assuming evolution has occurred). It addresses the problem and provides a potential solution to the problem.
--Cornelius
IP: Logged
|
|
Pim van Meurs
Member
Member # 541
|
posted 12. November 2003 00:56
CH: This is not a criticism of the Rokas paper as you seem to have assumed. The Rokas paper is operating within the evolution program. It is presupposing evolution and addressing a technical problem that occurs within that program (namely, how best to recover the true historical phylogeny assuming evolution has occurred). It addresses the problem and provides a potential solution to the problem.
Of course the fact that so many phylogenies are congruent strongly supports the assumption that evolution has occurred. Of course phylogeny is not the only evidence that strongly supports evolution.
If as CH claims: I tried to illustrate this with my gravity analogy. The analogy was not meant to deceive, it was meant to illustrate the problem that anomalous, even if rare, observations need to be explained.
Then I would argue that science has provided for some good reasons and hypotheses, backed up with actual evidence, to support why in some cases congruency seems to be lacking. So if I understand CH correctly he accepts that there are a majority of agreements with some disagreements. Science is working on resolving them, and in many cases has resolved many of them.
[ 12. November 2003, 00:59: Message edited by: Pim van Meurs ]
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 12. November 2003 02:39
Hunter wrote quote:
Phylogenetic convergence is claimed to be powerful evidence for evolution. The problem with this claim is not that there is an extreme lack of convergence. Though there are many non convergent characters, the majority of the characters do tend to agree. The problem is that those non convergent characters need to be explained, and the explanations require mechanisms that are not empirically observed. Indeed, one could just as easily argue that the lack of convergence makes this evidential claim serve to falsify evolution. There are non convergent characters that require explanatory mechanisms that have never been observed. One could argue this turns the evidence against evolution. But at the very least, as I see it, we must admit the evidence is not strongly in favor of evolution.
I'm having a little trouble following the line of argument here. Would Hunter be so kind as to provide a reference or two illustrating what he means by "these non convergent characters" and by "mechanisms that are not empiricially observed" and by "explanatory mechanisms that have never been observed"? Specifically what are they?
Second, there appears to me to be some slippage in Hunter's use of the term "convergence." This paragraph contains several of the senses that seem to be used: quote:
Now the findings in the Rokas paper are that, for the particular yeast data set they assembled, a large sample does provide for convergence. This does not mean there are no non convergent characters. There are. It means that those characters were not very significant in the analysis. That is, assuming evolution is true, the resulting estimated phylogeny was of high probability. The non convergent characters are still there, and the various explanatory mechanisms are still used. But because they were swamped by the agreeable data, they did not cause the estimated phylogeny to be of lesser probability. I do not see how this has anything to do with instrument power. The greater "power" added by the Rokas analysis does nothing to further explain the anomalies.
"Convergence" does not appear in the Rokas, et al., paper, though I suppose one could say that with increasing numbers of genes used in the analysis, the resultant trees converged on a single form. That seems to be the sense of "convergence" in Hunter's phrase "a large sample does provide for convergence." The Rokas, et al., paper was directed at the question of differing tree structures derived from genetic data. In evolutionary biology the trees are interpreted as phylogenies - a map of the kinship relationships among lineages deriving from a single ancestral lineage. The question was whether the several trees derived from a few genes would resolve to one tree if enough genes were used in the analyses. And they found that is the case. That is not driving anomalous observations into the noise; it's more akin to increasing the number of data points to better describe a curve of unknown shape. There's a substantial difference between driving something into the noise associated with estimation and reducing the noise associated with estimation. As I interpret it, the Rokas, et al., paper shows that the noise can be reduced.
Hunter's second usage(s) of "convergence" is in the phrase "phylogenetic convergence" in the first quotation above It's not clear whether Hunter means 'convergence of characters displayed by lineages related by common descent on the evolutionary hypothesis' or 'convergence of molecular and/or morphological data on a common tree of descent.' In the sentence "Phylogenetic convergence is claimed to be powerful evidence for evolution" Hunter appears to mean the observation of several independent lines of evidence that converge on a common phylogeny showing the forms of relationships expected on the hypothesis of common descent, which is one of the cluster of theories that comprise "evolution."
The third usage is in the phrase "non convergent characters." Here I can't tell whether Hunter means "characters in putatively related lineages that differ,' or 'characters that generate incongruous phylogenetic trees.'
So I'd appreciate some clarification and a reference or two for the three claims noted in the first paragraph of this posting.
RBH
[ 12. November 2003, 02:48: Message edited by: RBH ]
IP: Logged
|
|
Josh
Member
Member # 405
|
posted 12. November 2003 09:53
This should not be surprising to anyone. If two fish have fins, or several yeast species grow very similarly and inhabit similar environments, we should expect to see similarity in their genes. Simply put, similar phenotypes will have a basis in genotype that will also be similar. Wells gives us the impression that molecular analysis will never be resolved and leads us to confusing phylogenies. This may be the case for some examples, however he fails to mention that the Nobel Prize in 2001 was awarded to researchers in the cell cycle field who proved that human genes can replace genes in yeast. This means that molecules for cell division have great similarity throughout life and that they perform quite similar tasks that are conserved. More than likely, genes involved in eyesight will not have any function in yeast, so there is a general trend that will emerge just based upon phenotypes. So, there may be some variation/ incongruencies, but there may also be alot of general trends that argue for common descent (and probably more examples that support the notion of common descent than not will show up.) Even Behe doesn't have great qualms with common descent, but this may just be because his specific focus is IC. The real question is whether this proves that all life descended from one another or if some IDer generated life based upon similar blueprints.
Question for the evolutionists. How do you falsify common descent based upon these molecular phylogenies? Every single observation (i.e. congruencies and incongruencies) will fit neatly within the plastic theory of evolution such that no observation could possibly lead to the conclusion that common descent is false. An unfalsifiable theory may not be the best.
IP: Logged
|
|
Cornelius G. Hunter
Member
Member # 81
|
posted 12. November 2003 14:08
RBH:
The research results presented in the Rokas paper does not explain how non convergent characters arose. It assumes they arose via the standard explanations used by evolution. As I pointed out in an earlier post, some of those explanations are not empirically observed. The paper argues that (given this assumption) they believe that increasing sample size is an effective way to solve the convergence problem. This has been a point in question amongst evolutionists. As an aside, the paper by no means resolves this problem since it is looking at only a few species, but it is a good start.
You concluded that "There's a substantial difference between driving something into the noise associated with estimation and reducing the noise associated with estimation. As I interpret it, the Rokas, et al., paper shows that the noise can be reduced."
You are failing to address my point. If you feel that "reducing the noise" is a better way to describe the paper's results, then that is fine. But this does not address my point, and the gravity analogy. One can reduce the noise with additional measurements supporting the 1/R^2 relationship, but this does not remove the 1/R observations. They still must be explained.
As for the word "convergence," I am using it in the same way the paper uses the word "congruence." When different characters produce the same phylogeny, then there is a convergence, or congruence. So, for a simple example, if one character is very different in two different species, and another character is also very different in the two species, then they are convergent. As for references, here are a few:
Michael Balter, "Morphologists Learn to Live With Molecular Upstarts," Science 276 (1997): 1032-1034.
Jason Raymond, et. al., "Whole-Genome Analysis of Photosynthetic Prokaryotes," Science 298 (2002): 1616-19.
Tim M. Berra, Evolution and the Myth of Creationism (Stanford, Calif.: Stanford University Press, 1990) 62-6.
--Cornelius
IP: Logged
|
|
Grape Ape
Member
Member # 399
|
posted 12. November 2003 15:26
Josh writes:
quote:
Question for the evolutionists. How do you falsify common descent based upon these molecular phylogenies?
The lack of a nested hierarchical pattern, or a pattern which was anti-correlated to a nested hierarchy.
quote:
Every single observation (i.e. congruencies and incongruencies) will fit neatly within the plastic theory of evolution such that no observation could possibly lead to the conclusion that common descent is false.
I'm not sure what you mean. If you're saying that the lack of congruence fits "neatly" into an evolutionary framework, then I disagree. It's true that a lack of congruence can be accounted for and is even expected under some circumstances. For example there is an expected noise level, and there are situations under which one expects the possiblity of conflicting signals. (And there are good reasons for these -- it is not a mystery why trees can sometimes end up incongruent.) But if there were really a complete lack of congruence, this would be very problematic. If a nested hierarchy were not a genuine feature of species, then it would be impossible to build a phylogenetic tree with any degree of statistical significance. It is, for example, impossible to impose such a classification on atomic elements, planets, and most man-made objects. But such a classification scheme can be imposed on living organisms, and it can be done so with a very high degree of statistical significance, even though there are occasionally disagreements in the data.
As it stands, even incongruent trees are clustered on a very small subset of the overall tree space, which is monstrously huge for any data set containing more than a handful of taxa. So congruence is pretty strong, even when there are disagreements. If there were no such congruence, then common descent would be in trouble, and creationists would waste no time pointing it out.
quote:
An unfalsifiable theory may not be the best.
Agreed. ![[Wink]](wink.gif)
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 12. November 2003 16:53
In response to my request to provide references substantiating his use of the phrases "these non convergent characters" and "mechanisms that are not empiricially observed" and "explanatory mechanisms that have never been observed," Hunter provided the following: quote:
Michael Balter, "Morphologists Learn to Live With Molecular Upstarts," Science 276 (1997): 1032-1034.
Jason Raymond, et. al., "Whole-Genome Analysis of Photosynthetic Prokaryotes," Science 298 (2002): 1616-19.
Tim M. Berra, Evolution and the Myth of Creationism (Stanford, Calif.: Stanford University Press, 1990) 62-6.
The first is a news story (not a peer reviewed research or review paper) in Science (available here) that describes the sometimes heated debate that occurs in sorting out the various methods of determining phylogenetic relationships. The main focus of the story is on reconciling morphological and molecular methods of reconstructing phylogenetic trees. The article says (nor clearly implies) nothing about the three phrases Hunter used. In particular, there's nothing about empirically unbserved mechanisms. Perhaps the most interesting paragraph from ID's point of view is this: quote:
And puzzles there are. Few groups of plants or animals have had their evolutionary, or phylogenetic, trees worked out with complete confidence. Debates still rage over when and how flowering plants split off from their nonflowering ancestors, the relations among orders of amphibians, the origins of rodents, and a host of other issues. And while a number of talks at the meeting showed the considerable power of molecular data to tease out elusive phylogenetic relations, there were also warning signs that molecular evidence can lead to misleading and embarrassing errors.
That's a sign of the kind of healthy disagreement that generates research to settle the questions and elucidate the errors and develop the methodologies. The first step in correcting errors is to identify them, and that's what's going on.
The abstract (full text here) of the second is:
quote:
The process of photosynthesis has had profound global-scale effects on Earth; however,its origin and evolution remain enigmatic.Here we report a whole-genome comparison of representatives from all five groups of photosynthetic prokaryotes and show that horizontal gene transfer has been pivotal in their evolution.Excluding a small number of orthologs that show congruent phylogenies, the genomes of these organisms represent mosaics of genes with very different evolutionary histories.We have also analyzed a subset of "photosynthesis-specific" genes that were elucidated through a differential genome comparison. Our results explain incoherencies in previous data-limited phylogenetic analyses of phototrophic bacteria and indicate that the core components of photosynthesis have been subject to lateral transfer.
Far from substantiating Hunter's use of the three phrases, the paper appears to clearly contradict his point.
I don't have immediate access to Hunter's third reference, a book, but its date (1990) makes me cautious of its current relevance to questions of phylogenetic determination. The last 12 or 15 years have been the most productive with respect to elucidating the relationships and processes at issue, so I'm wary of the relevance of something written that long ago.
So I ask again, are there references to substantiate Hunter's use of "these non convergent characters," and particularly his use of "mechanisms that are not empiricially observed" and "explanatory mechanisms that have never been observed"? What are these "mechanisms" that Hunter claims are invoked that are not empirically observed, and where in the professional literature of evolutionary biology are they invoked?
By the way, I don't disagree at all with Hunter's observation that anomalies need to be accounted for. I do disagree with the clear implication of his gravity analogy to the effect that evolutionary scientists are not interested in doing so. A whole slew of research directed specifically at resolving those anomalies says otherwise. (References available on request.) In fact, Hunter's own references demonstrate that's not the case. Directly contradicting Hunter's claim that "He [the hypothetical physicist] then argues that because those 1/R cases are rare, that they don't matter," Hunter's own references show that the anomalies are the focus of vigorous debate and research! It's figuring out the anomalies that earns the Nobel prizes!
RBH
[ 12. November 2003, 17:07: Message edited by: RBH ]
IP: Logged
|
|
|
Printer-friendly view of this topic