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

Re: cladobabble.../Some clarification... ([very]long)

King, Norm (who I should point out I generally aggree with) wrote:

>>. . . "The clade Maniraptora [is a classification which I or others 
>>will continue using], no matter what [animals it might refer to]."
>>About the only thing I know about cladistics is that it is an 
>>intellectual method of classifying (certain?) animals.
        Actually, although this seems to me to be the intent, De Quiroz and
Gauthier go to great lengths to make a distinction between classification
and the recognition of natural groups. PT is the latter.

>I agree that clades do not exist in material form.  Previously some 
>discussants claimed that they are the REAL objects out there, but I 
>disagreed, saying animals themselves and fossils are the only real 
>objects.  I can hit you over the head with a fossil, but not with a 
>clade; clades are mental constructs (hope I used that term correctly).

        I am afraid you have missed the point of yesterday's philosophy
storm. Clades *do* exist in nature. A group of populations through time
united by their descent from a single population is a real entity. Lemme put
it to you this way, I can't hit you over the head with the historical
relationships inherent in a strat column, but they are no less real than the
rocks which elucidate them. I can use various geological methods to discover
that a series of geologic structures result from stress associated with the
same tectonic event, even though I cannot hit you over the head with that
tectonic event. My immediate family is real, because my sister and I are
united by a process of ancestry and descent with my parents, as will her
children (and mine, on the off chance I ever stop studying enough to leave
the Department). Based on this, you could seperate the entire family as "Roy
and Sue Wagner and all of their descendants", a real biological entity. Btw:
that last example shouldn't be taken too far, it's more of a metaphore, as
far as I want to take it.

>Now maybe I am being overly reductionist, but as mental constructs, clades
>would not exist in nature external to human reality.  So, they are either 
>not what I say they are, or not what Tom says they are.
        And the answer is: not what you say they are. :)
        See my last posting on this.

>So, the definition of a clade depends on an earlier determination, which 
>may not be true.
        In one sense yes, in that the decision to recognize a clade as a
phylogenetic taxon is dependant on the namer's determination that the group
is important enough to require a name. However, the clade *exists*
independant of that determination. Bazillions of clades exist or have
existed in the past. We may not have observed them all or evaluated their
content correctly, but that does not affect their existance.

>It seems that we define clades in general, then try to 
>see what the specific ones are.  Specific clades aren't really 
>defined--they're identified.  Right?
        I am not clear on what you mean. We can define a phylogenetic taxon
(which must be a clade) by identifying what I like to call anchor taxa (taxa
which must be included or excluded from the group by definition, and whose
common ancestry and pattern of descent are used as guides in the
specification of the exact clade being named, e.g. Maniraptora's anchor taxa
are _Ornithomimus_ and "modern birds", and the clade consisting of their
common ancestor and all of its descendants is named Maniraptora). Anchor
taxa identify a clade, which *must* exist (unless you boff the definition).
We can then use some form of phylogenetic analysis to determine which
animals fall into that clade.

>I don't see any advantage to erecting this bewildering array of clades,
        No more bewildering than the Linnean system. I seem to recall there
being a profusion of brachiopod groups of various ranks. then there is the
redundancy inherent in obligatory use of ranked taxa. Besides, in PT, you
use which taxa you think you need to. I don't have to say an animal is a
reptilian archosaurian dinosaurian theropodous tetenuran avetheropodan
coelurosaurian maniraptoran eumaniraptoran avian ornithothoracine
enantiornithine avisaurid if I can more easliy say it's an avian avisaurid.
But I can put any animal which isn't as close to modern birds into a
reasonable phylogenetic context when I tell you what more inclusive clades
it belongs to.

>with all the undecipherable vocabulary that goes along with it, and the 
        Clade, node, stem, taxon, inclusive, exclusive... ?
        Is "King Peter Can Off Four Glasses Slowly" really that much more
decipherable, or are you just *used* to it? :)
        [for the casual non-science type: Kingdom Phylum Class Order Family
Genus Species]

>interminable debates over what is in each clade.  Maniraptor, for 
        As opposed to the good old days, when no one argued about taxonomy
at all? Really, this debate has gone on for ever. All PT has done is change
the focus from arguing over what fits into a typological, arbitrary linnean
ranked taxon to arguing what animal is included by virtue of propinquity of
descent into an arbitrarily named *real* group. :)

>example, is an unnecessary term, because we already had the same concept 
>in saying that birds have ancestors among theropod dinosaurs.  We have 
        I sense that you are still thinking in Linnean terms. Are you
suggesting that by recognizing this cross-"class" taxon we are somehow
disrupting the proper processes of taxonomy? Is it your contention that we
would be better off leaving  _Deinonychus_ a reptile and _Archaeopteryx_ in
class Aves? This is not something that happens in PT. PT recognizes no
"rank", and no "equivalence" of ranked taxa. Therefore, your objection is
irrelevant. PT taxa may be named with the intent of illustrating
phylogenetic relationships. Indeed, this is one of the reasons some workers
find PT preferable to traditional systematics.

>always tried to identify the specific characters that diagnose birds and 
>to see what their ancestors are.  What's the difference?  Maniraptor 
        The difference is that we can talk about the evolution of
Maniraptora relative to their sister group Arcotmetarsalia. We can make
statements which have phylogenetic meaning. We can test each others
taxonomies by examining the phylogeny of various groups. Our results relate
to real historical entities.

>simply excludes ornithomimes, and puts a limit on how far back we want to 
>go in theropods to be talking about an ancestor in this context (e.g., to 
>the divergence between ornithomimes and the theropod group that went on 
>to produce birds).  T. H. Huxley hypothesized the theropod bird 
        It puts no limit on anything. Practitioners of PT do not limit their
exploration of NAD-AD relationships to Maniraptora. Indeed, you'll find that
many other theropod groups are even *defined* relative to "modern birds"
because of this.

>connection in 1869, and nothing has been discovered in the meantime to 
>disprove it or to require a different phrasing other than development of 
        Well, that depends on who you talk to or whose work you read. I was
rather under the impression that no one bought into this theory for
something like the the first 70% of this century. :)

>Of course, messy debates ensued over the limits and content of 
>conventionally-recognized taxa.
        Debates which could not be resolved, because:
a) there was no objective basis for the definition of taxa, and
b) there was no testable, reproducable, or moderately consistant means for
assessing the distribution of more exclusive taxa into more inclusive ones,
c) there was disagreement among workers on how these taxa should be/are
distribnuted, even assuming methodologies which were relatively similar

>But now we have messy debates over phylogenetic taxa,
        No, we have messy debates about the *content* of taxa. these debates
center not around points a and b, but merely around point c. Sereno doesn't
use some of Holtz's characters. Charig and Milner use some of Holtz's
characters, but reject others. This is science, the critical evaluation of
other people's methods and data to produce a better result. In the past,
point c was limited by the lack of carry-over in methodology from one worker
to the next. Arguments on the point were largely subjective. Now we have
less subjective standard, and our arguments are not about some abstract
typological arrangement of animals. Our arguments, although concerned with
taxonomy, are actually arguments about something biologically menaingful,
the phylogeny.
        E.g. If a PT taxonomist argues that sauropodomorphs and ornithicians
should be combined into one group, it is *NOT* becuase he thinks
plant-eaters should have their own group (which would be artibtraty and
typological in the extreme), it is because he thinks they share a more
recent common ancestor with each other than with theropods.

>and these are largely the result of not having learned from experience,
        No, they are largely the result of science. Science => argument like
tree => leaf.

>of not taking the time and care to erect a new system 
>that might help avoid such messy debates.
        How would you do this? The only way I can see is decree ONE SYSTEM
for all time. How useful is that? We need a system which can change with new
ideas. Messy debates are (as I said above) a part of science. At least now
our messy debates have an actual biological meaning, and aren't just
abstract disagreements over one or the other persons idea of what
"Dinosauria" should be.

>It can't be justified on the 
>grounds that since there were messy debates in the old system, it's OK to 
>have messy debates in the new system.  We had a chance, and maybe there 
>is still a chance, to make improvements for a new system.
        A fundamentally more consistant and reproducable systematics, with a
firmer grounding in modern concepts of biology, a systematics which
recognizes natural groups and channels debate from the arbitrary and
subjective to the specific and objective is *always* going to be more
favorable. Just because PT hasn't solved *ALL* problems does not mean it is
not better.

>Maybe start by adopting a defined procedure for characterizing clades, 
>like node-based.
        Node based definitions are, IMHO, overall of little value without
stem-based definitions. Stem-based defintions allow one to specify animals
which are serially closer to a node-based taxon without resorting to
erecting serially closer node-based taxa. Example:
        ((A, E), (B, (C,D)))
        If I define saurischians = {+A, +D}, and sauropodomorphs = {+C, +D},
theropods = {+A, +E}, what do I call taxon B? It is a "basal saurischian".
No indication is given that it is closer to sauropodomorphs than to
theropods. The node-based solution is to erect a new taxon, "sauropodiforms"
= {+B, +D}. B is a sauropodiform. But what if a new taxon is discovered?
        ((A, E), (F, (B, (C,D))))
        What is F? A "basal saurishican", again with no indication that it
is closer to sauropodomorpha (or sauropodiforms) than to theropods. The node
based solution is to erect *another* taxon, sauripodocoglia = {+F, +D}, so
that F is now a sauropodocoglian.
        The stem-based solution is to start by erecting a stem sauropodigens
= {+D, -A}. Thus, B and F are sauropodigens. Now, you don't know which is
closer to sauropodomorphs, but with only one taxon you have something to
call *Anything* that is closer to sauropodomorphs than to theropods.

>Node-based would probably have the least disagreement 
>with earlier taxonomic concepts and hypotheses of descent, and that's 
        Stem-based taxa don't disagree with any hypotheses or taxonomic
concepts, to my knowledge, at least not any more than node-based taxa. Yes,
they are easier to get a grasp on. A lot of concepts are slightly more
complex than complimentary ones (e.g. correlation versus regression). That
still doesn't mean you can't use both.

> Don't name every clade--the proliferation of names may 
>ultimately produce a structure as as complex as the tree of life itself 
        If you only use node-based taxa.
        Y'know, I'm getting quite sick of this entire "proliferation of
names" thing. How many &^%&% genera are in their *OWN* family in traditional
taxonomies? And *that's* not proliferation?

>(as G. O. has pointed out).  The whole superstructure will get bogged 
>down by its own nomenclatorial complexity,
        As if traditional taxonomy is not. Bear in mind that the complexity
of traditional taxonomy is an artifact of the method. Although it may
reflect a subjective assessment of gross morphological difference, it is
hardly as concrete or reproducable as the complexity of PT.

>to say nothing of the 
>background chaos caused by the inevitable disagreements over diagnostic 
>characters and clade "definition."
        Diagnostic characters do not have a thing to do with it. You could
do a full phylogenetic taxonomy of Animalia without ever once referring to
diagnostic characters (wouldn't do anyone much good, though). As for
definition, recall that we define *taxa*, we *name* clades. The clades are
out there, our definition tells us which one is being named.

>Phylogenetic taxonomy also leads to absurdities such as the Tetrapoda not 
>including all animals with four legs. Now, just stop and think about that 
        And how would you define it phylogenetically so that it did? What is
a "leg" anyway? Where do you draw the line? Is everyone going to draw that
line the same? Yeah, maybe it's silly, maybe we should just have dropped the
name. Then you might be saying "where's Tetrapoda? Isn't it absurd you
aren't using Tetrapoda?" :)

>That simply won't play in Peoria, and if it doesn't, we 
>could all be in trouble.
        Contrary to the movie Contact, sometime 95% of the world *is* wrong.
Access to scientific work for the layman is all well and good, but should we
allow the waning average intellect of an increasingly anti-intellectual
culture dictate the path of scientific discourse?

>Also, as I said previously, any Peorian can see 
>that birds aren't dinosaurs.
        And maybe that's why we should make the point in our taxonomy.
        The public yused to be impressed when we artibrarily seperated out
one animal over another into different "families" arbitrarily. Maybe they'll
be more impressed if we have a reason. Or maybe they'll say we're
evolution-breathing heathens.

>Since birds are such unique animals, 
>there is nothing wrong with emphasizing that uniqueness.  
        We do: taxon Aves.

>Now, that said, would it surprise anyone to know that I demonstrate the 
>reasoning used to conclude that birds are dinosaurs in my class?  I 
>supportively explain how cladistic works in the beginning of the 
>semester.  I use it in other contexts, but then I really drive it home in 
>discussing birds.  They learn about cladistics, they are convinced, and 
>they enjoy the whole discussion.  Finally, none of this in any way 
>obviates the opinions and complaints about PT I expressed above and 
        You are possibly the most impressive example of how science and
academia should operate, and I salute you (not to mention worship the ground
you so objectively walk on). That's really cool! :)

>Perhaps I suffer from dual personality syndrome.
        Perhaps acquired in grad school?
    Jonathan R. Wagner, Dept. of Geosciences, TTU, Lubbock, TX 79409-1053
        "Chimp here does the killing." - Doug Mackenzie