[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Re: Dino cladistics - long with requotations (was Re: Dinosaur Web...

Dinogeorge@aol.com wrote:

>The reason I couched my post in these terms is to get (browbeat?) people into
>thinking about some of the results that cladistic methodology is coming up
>with and not simply (blindly?) accepting the latest cladograms that the "plug
and chug" computers have churned out.
        I am sure you have read recent papers on the subject, such as Novas
1997, Gauthier 1986 (recent?), Holtz 1994 or any of the plethora of Sereno
archosauromorph papers.  I am therefore perplexed that you say this. Many of
these authors go into great detail on the ramifications of their analysis,
specifically what it means in terms of character transformations and
homoplaisy. Indeed, I was impressed by the number of times Gauthier was
compelled to admit that a certain theropod character state was absent in
birds and/or present in pterosaurs. Hardly "plug and chug".

> As I've said before, a 95% confidence level means that there's still one
>chance in 20 that a particular cladogram is wrong >despite< parsimony,[...] 
        Would someone please explain "confidence"?
        As an aside, I am given to believe (from Gauthier 1986) that
Consistancy Index is # of characters/ # of steps. In that case, would not a
consistancy index of greater than 95% be unexpected and not necessarily
indicative of a good analysis, as it would include less homoplaisy than you
would expect in evolution?

>and if we're going to weed out any wrong cladograms, the place to start
could >well be with those cladograms that, at least at first sight, seem
most >outrageous.
        I believe that Benton (?) has been known to do this. The problem
with your "five big misconceptions" was that three of them are not at all
outrageous (Thyreophorea, Saurischia, and Therizinosauroidea) and two of
them actually are lacking in complete analysis.
        The idea in weeding out cladograms, if we were to do such a thing,
ought to be to pick the cladogram which best represents our understanding of
the data. One can hardly do that in the absense of another cladogram to
compare it to, as in your other two objections.

>I don't think Nature is slavishly parsimonious to the point that, say, 50
>characters >must< outweigh 49 every time;
        The application of parsimony is simply the acknowledgement that the
fewest unsubstantiated or untestable assumptions (character state
transformations, in this case) will be accepted in the final analysis. It
isn't about relative "weight" (whatever that is), it is about the same
adaptation occurring twice. Such determinations are best made from the
results of a phylogenetic analysis, not a priori to one.

>to assert this is to coin another "Natural Law" like Cope's Rule and
>Dollo's Law.
        More like "Scientific Method", ie. assumed homoplaisies are
minimized, but homoplaisy is not assumed to be minimal (see Sereno in
_Dinosaur Systematics_).

>I also [...] should be given far less weight than highly specific similarities 
>in each of a much smaller number of skeletal regions.
        Unfortunately, this leads to a greater chance that your "high
weight" characters are linked functionally (as I am almost sure that some of
the ones I listed linking Oviraptorsauria and Therizinosauroidea are), and
are thus overweighted. Deinonychosauria is an example of the effect
overweighting of characters (and insufficient data) can have.
        By limiting the data which may be used in an analysis, you
drastically reduce the possible objectivity of that analysis.

>but it is just as vexing to give equal weights to all characters
       It is counterproductive to assign weights a priori, because the data
you recover on character distributions is biased by your a priori assumption.

>I'll reply to the rest of your (most thoughtful and interesting) comments as
        No soup for me, eh? :)
| Jonathan R. Wagner                    "You can clade if you want to,     |
| Department of Geosciences              You can leave your friends behind |
| Texas Tech University                  Because your friends don't clade  |
| Lubbock, TX 79409                               and if they don't clade, |
|       *** wagner@ttu.edu ***           Then they're no friends of mine." |
|           Web Page:  http://faraday.clas.virginia.edu/~jrw6f             |