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

Beating a dead horse[ was Re: A NEW REF, SOME TETRAPODS, AND CLADISTICS & PT, Re: Crown clades a-frikkin'-gain]



        Jimany Christmas! Let's take this junk off the list, shall we?

At 06:40 AM 7/20/98 -0400, Jonathon Woolf wrote (quoting me):
>>         We've been through "rationality" and taxonomy before. Please check
>> the archieves under "Arctometarsalia".
>So we have, and you're still wrong. [...] _Homo sapiens_ is supposed to be a
>thinking animal -- so why not think instead of mindlessly following the rules?
        And with that, I think that discussion is best ended...

>A noble purpose.  Too bad some people insisted on throwing the baby out
with >the bathwater.  If membership in a group can't be diagnosed, the group
is >effectively useless -- and in a number of cases there is no known way to
>distinguish members of a crown-based clade from their immediate ancestors.
        Of course there is! Code them, run the data matrix, and find out if
they are descendants of the most recent common ancestor of the crown clade.
What's wrong with that? It is almost the exact same thing, only one step
better. Instead of just looking for the "diagnostic features", which anyone
worth his salt will tell you aren't always useful, you assess the whole
suite of characters. Besides which, workers who apply PT still *do* provide
lists of diagnostic characters for phylogenetic taxa.

>Early mammals, for example.  There is a long, long string of fossil forms
from >the latest Triassic up through the mid Cretaceous that look mammalian
in every >detail -- but can't be called mammals because they date from
before the last >common ancestor of Monotremata and Theria.
        How does one "look mammalian in every way"? Especially, how does
this happen, yet we conclude they are not descendants of the mammalian crown
clade? There MUST be some differences, or noone would ever suggest that they
aren't.
        BTW: I suggest you not use time metaphores in discussing phylogeny.
It is potentially confusing to everyone.

-----Meanwhile, in another post----------------------------------------

>So the name itself has a meaning in whatever language it's taken from.
That >meaning is part and parcel of the name. [...]
>As I see it, a name can have either a positive information content, zero
>information content, or negative information content.
        And "_Kritosaurus_", is this zero, or is it negative information
content? Perhaps I am "choosing reptiles", so I choose _Kritosaurus_ because
it is "the chosen reptile" ["There can be only one!"]. Or maybe I am
assessing the "nobility" of reptiles, and I include _Kritosaurus_, because
it is a "noble reptile" (it's a cow, for criminies' sake!). Isn't this
negative content? Should we sue the estate of Barnum Brown for picking a
misleading name? Oh, forget it, I'm sick of this pointless argument!

>If the thing has all the diagnostic features for the group, it should be
>included in the group.
        Oh dear. You proceed from the assumption that there are certain
diagnostic features which *must* go with a certain taxon. Yet, as has been
demonstrated by others (Rowe et al.), there is no one set of diagnostic
features which define "mammalness". On the other hand, there may be one
character which defines "tetrapodness". Until you look closer:
        At what point do four lobes become four limbs? Answer me that, give
me the exact instant, and maybe I'll continue this argument.

> Did you hear about the molecular trees that place birds and mammals in a
clade >with dinosaurs as an outgroup?
        Where did they get dinosaur DNA?
        As for this general thread of anti-cladistic propaganda, Chris
Brochu has done an admirable job of refute such claims on this list. I
suggest you consult the archieves.

>Did you hear about the three or four different analyses of elephant shrews
-- >all of equal parsimony --
        How do you compare relative parsimony between seperate analyses? I
know of no method for doing so.

>that assigned them to as many different places in the overall mammalian tree?
        This happens. Different workers get different results. Cladistics
isn't a black box that spits out the right answer. It's a method of
organizing information and producing testable hypotheses based on
established scientific criteria. Maybe the four workers now have come to a
consensus, maybe they haven't. It's called science, deal with it.

>Robert Carroll almost got lynched at a conference a couple of years ago
because >he dared to point out that cladistic methods were not infallible.
        I am sure you exagerate. Noone suggests cladistics isn't infallable,
only that it is better than not using cladistics. Why are we having this
discussion again?

>As far as I'm concerned, there's enough contrary data floating around out
there >to question cladistics as it's currently being applied and used.
Until that's >no longer true, until all that contrary data is taken into
account and >explained,
        Until doomsday. If there were no contrary data, there'd be no
science. We'd have all the answers already. Science is the process by which
we gatehr data and test hypotheses to resolve these contrary data.
Cladistics is tool MEANT to help resolve these contrary data. Yes, it also
produces incompatible results. Does that mean we "throw the baby out with
the bathwater" as you so eloquently put it?

>I'm not going to accept cladistics blindly.
        Good, as well you shouldn't.

>"Most of the time" isn't all the time.  So why do people treat phylogenetic
>taxonomy as being infallible, and either ignore or rationalize away the
>exceptions?
        There is no question of "fallability" in PT. PT is a taxonomic
method, and I am not sure there is a way it can fail, except in the
subjective judgement that it is somehow less desirable than other systems.
Objectively, it's theoretical framework is perhaps the most sound structure
ever devised in the history of taxonomy.

>If Hopson is right that monotremes and placentals share a MRCA in the Mid
>Jurassic, then nothing earlier than that can be a mammal --
        Again, let's avoid the time thing.

>including such animals as the morganucodonts, which have every feature that
>could reasonably be used as an apomorphy for mammals.
        Again, if they had every mammalian apomorphy, they would be members
of the crown taxon Mammalia.

>On the other hand, if the idea that monotremes are
>descended from a different line of therapsids than all other mammals is
        This is sort of a cop-out, since therapsids (in this use) are
paraphyletic. What you are really saying is that the "Mammalian" features of
monotremes and other mammals are homoplasies.

>correct, then crown-clade Mammalia would have to include a number of
therapsids >that _don't_ have many of the mammalian apomorphies.
        Fine. So be it. What would you have instead? You'd have two choices,
a polyphyletic Mammalia, or a Mammalia excluding monotremes. It's either
those or the crown clade. Take your pick. Think fast.

>Uh-huh.  In other words, the cladists want me to ignore the evidence and
just >take their word for it.
        What evidence? Evidence that cladistics "doesn't work". Heck, by
your reasoning, I could say 14C dating "doesn't work", because you can have
contaminated samples, resetting, or samples from beyond the resolution of
the method. Does that mean radiocarbon "doesn't work", and I have to "ignore
the evidence and take their word for it" in order to use it? How about
stratigraphy? There's a bunch of controversy in stratigraphy. Should I just
"take their word for it", and assume stratigraphy works even though half of
them think the Maastrichtian Javalina Formation is really part of the larger
Cretaceous-Tertiary Tornillo Formation? Ooh, but the controversy is so
lively, and lord knows I could never go out to Brewster County and just form
my own oppinion. I guess I'll just walk away from it, and lock myself in a
little room with the things I "know", like what makes a mammal. What does
make a mammal, anyway?

>I know "experts" who worship
>cladistics.  I know other "experts," just as qualified if not more so, who say
>that cladistics and phylogenetic taxonomy as paleontologists use them are crap.
        Yeah, we all know controversy. Science, you gotta love it!

>When I hear conflicting opinions from two different groups of "experts," I
>throw away all their opinions and go back to the facts.  And the facts, to
my >eyes, say that the cladists are wrong.
        About what?

        Wagner
----------------------------------------------------------------------------
    Jonathan R. Wagner, Dept. of Geosciences, TTU, Lubbock, TX 79409-1053
                    "...To fight legends." - Kosh Naranek