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

RE: Classification: A Definition



(hopefully this does not get truncated, sent from an alias as plain
text)

David Marjanovic wrote:

>Compared to Linnaean classifications, all that phylogenetic 
>nomenclature will take away is the illusion that some taxa are somehow 
>at some kind of same level and therefore somehow comparable (for 
>example, that it's possible to count families). (...)

OK, confessed, I belong(ed) to those many people that thought that a
phylogenetic tree could indeed be translated into a classification, the
extremities of the tree then being the lowest taxonomic levels (usually
species).

But all the same, it is exactly the above statement that I question. I
will try to explain what I mean.

Do not misunderstand me: I do not contest the scientific 'correctness'
of phylogenetics as opposed to the largely 'artificial' nature of
(traditional) classification, and phylogenetics usefulness for
indicating descendence and relatedness, but I question its practicality
for handling large numbers of taxa, i.e. that it would make
classification entirely *redundant*, as you state.

If, as you state, phylogenetics does away with any absolute ranking and
comparison of taxonomic level, and everything indeed becomes only
relative, then in the everyday practice this may become very impractical
and even almost impossible to tackle large numbers of taxa; for
instance, how to deal with: species lists, biodiversity analysis and
mappings, standard works of reference (encyclopedias), without a
practical classification and listing? Sure, the number of names, as you
state, only grows bigger (potentially even without limit, for a
potentially limitless number of clades?), but they loose their
significance with regard to any taxonomic ranking/classification.
If I understand you well, then, when applied consistently, there should
no longer be any reference to families (not even to mention higher
ranks), maybe not even genera, and in its most extreme form even the
species itself becomes questionable (though I can hardly imagine that
anybody would drop that concept).

For example, besides the two mentioned amphibia websites, large standard
works (just a few that I happen to know and use) such as Wilson & Reeder
Mammal Species of the World and Walker's Mammals of the World, would
become very impractical, loosing overview without classification in
orders, families, etc.
And if I may say, even the website 'The Speculative Dinosaur Project'
(fascinating site I might add) should not make reference to families,
but just limit itself to trees.

Is there not a kind of 'reconciliation' possible, for practical reasons,
e.g. 'cut-off' ponits/lines in trees, indicating what we call a
(particular) family, genus, etc. Yes, this would be classification, but
an improved one. Or is this heresy again?

Bottom-line is: how to present (overview of) large numbers of taxa in
trees, without (tabular) classification?

By the way: do not confuse AmphibiaWeb and Frost, they are not exactly
the same source, Frost is the source of the on-line version of
'Amphibian Species of the World' of the AMNH. AmphibiaWeb is of Berkeley
and partly follows, but does in many parts also deviate from Frost.

Comment and clarification is appreciated.


-----Oorspronkelijk bericht-----
Van: David Marjanovic [mailto:david.marjanovic@gmx.at]
Verzonden: ma 14-5-2007 12:27
Aan: dinosaur@usc.edu
Onderwerp: RE: Classification: A Definition

Probably most people only got the "message truncated" error, so I repeat
the whole post:

-------- Original-Nachricht --------
Datum: Mon, 14 May 2007 10:47:00 +0200
Von: "Botterweg, Rrp" <ronald.botterweg@capgemini.com>
Betreff: RE: Classification: A Definition

> David Marjanovic wrote:
>
> "It's not really an alternative. Phylogenetics and PN together make 
> classification completely superfluous; it just drops out of the 
> equation; neither it nor any alternative is needed. "How should I 
> translate a tree into a classification" turns out to be a wrong 
> question, like Lowell & Dingus's famous "why did Napoleon cross the 
> Mississippi"."
>
> Unless I misunderstand something here (and would love to be informed),

> I tend to say that this goes a bit too far, at least as far as the 
> lower taxonomic levels are concerned (family and particularly below), 
> for a very practical reason: overview.
>
> Since phylogenetics/cladistics show relative distances between 
> taxa/clades, it might become rather impractical when dealing with 
> large numbers of taxa (genera, species), one tends to loose overview 
> (such as the large numbers of Amphibian species I am contributing to 
> at AmphibiaWeb and 'Amphibian Species of the World', on-line version, 
> at the AMNH), but I admit I am approaching this from a database angle.
>
> So, although phylogenetics/cladistics definitely constitute a more 
> 'natural' system and 'traditional' classification an much more 
> 'artificial' one, I think the latter still has its merit from the 
> practical point of maintaining an overview of large numbers of taxa, 
> hard to achieve with phylogenetic/cladistic trees.
>
> Unless you maintain that PN can do the same, but in that case I would 
> say that it does offer an alternative classification and 'translation 
> from tree into classification'.
>
> Please clarify and thanks in advance.

>I don't quite understand what you mean. Maybe you make the 
>misunderstanding Benton made in 2000, that the end of classification 
>would also be the end of nomenclature. It isn't, and that's where the 
>term "phylogenetic nomenclature" comes from. There will not be fewer 
>names for clades, there will be more.

>You mentioned AmphibiaWeb. Even without phylogenetic nomenclature, 
>Frost et al. coined lots and lots of rankless clade names in their huge

>work, because they want to talk about their tree and suffer from the 
>fact that there are never enough ranks to give every interesting taxon 
>a rank. Unfortunately,  almost none of these names have a phylogenetic 
>definition; this means that, when fossils are inserted into the 
>topology they found, there is in most cases no way to find out whether 
>they belong to the same taxon as their closest extant relatives or not;

>furthermore, most of the names by Frost et al. are completely 
>inapplicable to any topology except theirs -- for the same reason.*

>Compared to Linnaean classifications, all that phylogenetic 
>nomenclature will take away is the illusion that some taxa are somehow 
>at some kind of same level and therefore somehow comparable (for 
>example, that it's possible to count families). That is not a bug, it's

>a feature. :-)

* For a long rant about the nomenclatural practices of Frost et al., or
rather the lack thereof, see Appendix 6 (which will be online at
http://www.systematicbiology.org/) of: David Marjanovic & Michel Laurin,
next month. Fossils, Molecules, Divergence Times, and the Origin of
Lissampibians. Systematic Biology 56(3): 20 pp.