Charles Darwin

I agree with your sentiment, though not with your numbers, but we can let that go. Let us agree that, for some hypothetical data set, you could infer a set of phylogentic trees using clustering methods or by choosing objective criteria. And it would be quite likely you would not obtain a single tree that unambigously fit the data the best, but rather a set of trees. And there might be some significant topological differences between the set of trees judged to fit the data well. Nonetheless, that set of trees would exhibit a high degree of similarity, when compared to the set of all possible trees.

Now, why exactly is this odd? You say this is "to a highly, highly statistically significant degree." Compared to what? [Hint: answer = randomly arranged traits]. So what? [Hint: the test is meaningless].

You have ignored the many mismatches and their attendant explanations, and have found great meaning in the fact that the species exhibit correlated traits. Amazing.

Charles Darwin

Very interesting. Would you care to explain this secret knowledge that you have. I'd be fascinated to know why a nested hierarchy would not come from creationism. Oh and also, remember not to "imply viewpoints and then categorically dismiss them."

Well you're good at berating me. By the way, what is it that I don't understand about the practice and methods of phylogenetic reconstruction? I thought I had a good handle on it. You say that the mismatches are nothing more than:

"numerical statistical analysis using algorithms designed to isolate the most parsimonious cladograms displaying the greatest character-support at each node, they demonstrate that it is often not easy to establish phylogeny."

Yes, of course this is the case. And there are a dozen or so explanatory mechanisms for every case. So why is this such strong evidence for evolution?

And of course, I never said they were. Let's not "imply viewpoints" OK?


I don't know what you are talking about. Making things up, or referring to those "creationists" is irrelevent.

Charles Darwin

What is it about this sentence that you don't understand:

"Animal relationships derived from these new molecular data sometimes are very different from those implied by older, classical evaluations of morphology. Reconciling these differences is a central challenge for evolutionary biologists at present." Science, 279:505

How did I utterly fail? Please be specific this time.

Doubting Didymus

Actually, the comparison would be to a number of things. For example, a situation where species share genetic similarity because of morphological similarity would predict a certain tree. This is a falsifyable possibility. This is also a falsified possibility. apart from your egregious assertions earlier that all functionless DNA actually does have undiscovered functions, you have not dealt with the problem of independant confirmation from trees built from junk DNA comparison for the 'similar morphology => similar genes' hypothesis.

Doubting Didymus

First of all, WHAT new molecular data is the article referring to? also, what is the problem with new molecular data updating old morphology - based trees? Did you expect phylogeneticists to have everything perfect first try? Also: WHICH classical morphology trees have been updated by this new molecular data? In what way have they changed? Is this a mismatch or an update?

You see, you can't just grab desperately at every tasty sentence where a scientist has said "we had such-and-such a problem", and claim to have obliterated an entire scientific field. Your quote is a meaningless soundbite, until it is placed in context. It says nothing at all about anything specific. In other words, another utter failure to provide any evidence of problematic mismatches that are evidence against the theory of common descent. You have evidence of a scientist saying "hmm... ahhh" and nothing more.

Urvogel Reverie

This torturous parody of cladistic methodology for phylogenetic reconstruction is simply too egregious to let go, and despite the exasperation of having to provide you with a tutorial of how the process actually works, it would seem that at least something along those lines is necessary. You seem to imply that algorithmic analysis of character-sets and the resultant phylogenetic maps are slapped together piecemeal, and thus that parsimonious maps underwritten by multiple characters at each node, are meaningless as their statistical likelihood is being compared to whimsical aggregates of random traits. This is the most idiotic drivel. Cladistic analyses of any taxon start with the elaboration of a character set to be analyzed, and from that point generate bifurcating patterns by which those characters may have appeared, governed by what we know of the process and pattern of evolution. Subsequent maps are thus potentially equally valid, and as they all are formulated from numerical study of a uniform set of characteristics, there is no "fudging" of the data, as it were--i.e., the analysis is not biased so that one set of parsimonious maps emerges which are lent an illusory credibility by inaccurate comparison to other maps. A parsimonious map, underwritten by a preponderance of characters at each node of the cladogram, is statistically likely to be accurate for the entirely different reason that out of alternatives of equal potential, this map is the most substantiated by the data at hand. Your attempt to introduce reasonable doubt into the methods of cladistics to show that phylogenies only appear to corroborate evolution (via an algorithmic sham), is quite simply bunk. One would suggest that you actually read the literature pertinent to the matter of phylogenetic reconstruction, before attempting to refute it.

Urvogel Reverie

Doubting Didymus

By the way, I would appreciate it if you could place publication dates on some of your Science journal references. I am getting a vague feeling that some of them might be quite old and outdated.

Urvogel Reverie

A nested hierarhcy is diametrically at odds with any anti-evolutionary model for organic diversity yet advanced, in that their one unifying factor, is the assertion that taxonomic units are typological. If one is to deny evolution, it is a concomitant logical necessity that one maintains morphologic stasis over time. After all, there is nothing about phenotypic variation which would preclude a speciation event, at least to our knowledge, and since speciation is not allowable under anti-evolutionary models, that which would cause it, must be defined out of existence or denied outright. Hence the tired impression of Platonic essentialism into taxonomy by all manner of anti-evolutionists. Therefore, when you claim that a nested hiearchy could be evidence for creationism just as well as anything else, it reveals astonishing ignorance of your OWN worldview.

As for the matter of your lack of understanding about phylogenetic reconstruction--your entire series of posts on the matter has smacked of such. You obfuscate either intentionally or because you lack requisite knowledge, the principles and methods of cladistics, most egregiously in your claims about comparative analysis of cladograms which are formulated during numerical statistical analysis of a character set, which I addressed in an earlier post.

You never claimed they were? Did you not specifically say: "Meanwhile, the absurdity continues: the most amazing of devices are to have merely arisen?" If this is not a statement which indicates the opinion that cladograms merely assert changes in a lineage appeared as if by magic, then I have rarely seen one.

Making up the charge that creationists constantly assert that evolutionary biologists merely say that derived characters appeared without providing evidence of quantifying data? Not at all. Surely I cannot know the creationist argument better than a creationist. All one need do is read the work of Gish, Sarfati, Walker, listen to Hovind, Baugh, or any other number of creationist alumni, to find that very strawman repeated ad nauseum.

Urvogel Reverie

Coragyps

That quote is in the introductory paragraph of a short "where we are today" review. The rest of the article is about what's being done to address the "central challenge."
Register, for free, at www.sciencemag.org , and use their search function to read the whole thing. (volume 279, page 505 is the easiest search, but it is possible to also give full citations for literature one uses in arguments (/one pet peeve)):
"The Coming of Age of Molecular Systematics"
Laura E. Maley and Charles R. Marshall
Science 1998 January 23; 279: 505-506.

Charles Darwin

I'm sorry that you missed my point again. You seem to be repeatedly inferring that I am trying to cast doubt on the underlying algorithms and methods of phylogenetic analysis.

Once again, I must ask you to actually read my posts if you are going to make a relevant response. I said nothing about fudging the data or algorithmic shams. I very much trust the phylogenetic methods in use.

The point I was making to the gentleman was in reference to his claim that the results are odd and highly statistically significant. No one is doubting that the best-fit tree is indeed the best fit tree, and that the different characters are reasonably consistent. So what? Why is this so odd and what inference is so signficant?