Comments to Cladistics/Eclecticism

[SKÁLA Zdenek]

Tom DiBenedetto:
>Perhaps it is more understandable if I were to
>focus on the defintion of taxa. You say "named" clade, with an emphasis on
>_named_ (i.e to actually put a name on it). I would say taxa are
>"recognized" clades.

I do not understand why do you redefine the term "taxon" in such a manner; taxa were (and still are, I believe) generally understood as named groups (of taxa of lower rank except terminal taxa). By the way, what do you mean when do you write "paraphyletic taxon"? It can be completely unacceptable to you but still it is a taxon (in your own words) but is not a clade. Further, can you kindly tell me why do you need to have two terms for a clade? In your definition, clade and taxon are the same thing.
However, OK. Let's redefine the original terms in this manner, if you prefer so. Then:
clade = taxon
"taxon" (in the original meaning) = named taxon

> There is nothing arbitrary
>about cladistic recognition of clades.

Well, be sure that it is - but it is not a matter of the current discussion. Now we are not discussing the cladogram construction/recognition of clades but rules for making "named taxa" (= taxa in usual sense).

>Actually labeling them [clades] with a name is, at that point, a trivial matter.
>You seem to claim repeatedly that this is somehow arbitrary. But what is the point 
>of calling it arbitrary, ....

You are pointing that placing limits on the cladogram to split it into (paraphyletic) taxa is an arbitrary operation. I am responding that such placement of limits is equally arbitrary when we are splitting cladogram into a system of holophyletic "named taxa" - it seems that even you do not intend to actually label _all_ clades with a name. Hence, named taxa are only a selection of all clades - the rules how to select them is the matter.
That is (almost) all, but two notes:
(1) For many purposes - including purely scientific ones - we need a system of not many taxa (say orders, families, genera, species). We need to have them because (a) the main purpose of the system (unlike a cladogram) is to summarize information (into less higher taxa than are the terminal taxa) and (b) many clades can (and usually are) weakly supported and so have low information value.
(2) When I am writing that the cladogram split into named taxa is always equally arbitrary, it does not mean that this arbitrariness must be very high. I believe that we can develop algorithms for finding the most important gaps in character change that will work similarly well like the algorithms for cladogram construction.
Best!
Zdenek Skala