[Previous by date - Re: Possible resolution? Wishful thinking?]
[Next by date - Re: Possible resolution? Wishful thinking? [Re: Fwd: Re: Codes]]
[Previous by subject - Re: Fwd: Re: Codes]
[Next by subject - Re: Fwd: Re: Codes]
Date: Mon, 12 Feb 2001 17:11:37 -0600
From: "David M. Hillis" <dhillis@mail.utexas.edu>
To: PhyloCode@ouvaxa.cats.ohiou.edu
Subject: Re: Fwd: Re: Codes
<!doctype html public "-//W3C//DTD W3 HTML//EN">
<html><head><style type="text/css"><!--
blockquote, dl, ul, ol, li { margin-top: 0 ; margin-bottom: 0 }
--></style><title>Re: Fwd: Re: Codes</title></head><body>
<blockquote type="cite" cite><br></blockquote>
<blockquote type="cite" cite><br>
<blockquote type="cite" cite>Now I have the rest of Gerry Moore's
message. All I can say is that<br>
his experiences are very different than mine. The only objections
I<br>
hear when I explain the PhyloCode are from people who worry about<br>
the possibility of two competing, parallel systems of nomenclature<br>
(a point to which I fully agree). I cannot believe that anyone
wants<br>
that...two different names for everything, and no one able to<br>
understand the other side.</blockquote>
</blockquote>
<blockquote type="cite" cite><br>
We seem to have different understandings of what is meant by<br>
"parallel systems." I certainly don't advocate having
two different<br>
names for clades. As you know, I favor designating PhyloCode
names<br>
with a symbol (a "value-added symbol", if you will), but
the name<br>
would not differ in any other way from the name applied to that
taxon<br>
under the traditional system. Furthermore, I accept the
majority<br>
view of the advisory group that the symbol will not be
mandatory. So<br>
at this point, I don't think that we have a disagreement about
this<br>
issue. When Gerry and I refer to parallel systems, we mean that
the<br>
names are defined in a different way, not that the names
themselves</blockquote>
<blockquote type="cite" cite>differ.<br>
</blockquote>
<div><br></div>
<div>Would you like taxa be named so that new names could applied to
both sets of rules? If so, then you are correct, there is no
controversy. However, a description of Quercus-alba would not fit the
old codes, and if this form was required, a description of Quercus
alba would not fit your preferred rules. If I describe a new species,
I do not want the description to only apply to one set of rules...I
want it to fit any codes that are in place, and I want everyone to
use the same name in referring to the new species. To do that, we
have to co-opt the old system, which was my point. For clades, I want
the same names to apply to clades under both systems as well, for the
same reason. I have no objection to an optional symbol to distinguish
names that are registered from those that are not, but that doesn't
require any rules. The names themselves would be identical under both
systems.</div>
<div><br></div>
<blockquote type="cite" cite><br>
Gerry wrote:
<blockquote type="cite" cite>The only other point that I didn't
already respond to was:
<blockquote type="cite" cite>Hillis was critical of the existing
system, noting that it required<br>
"diagnoses of taxa rather than phylogenetic definitions, and
the<br>
diagnoses don't even have to be correct." However under the<br>
PhyloCode one can define taxon names under a phylogenetic<br>
hyopthesis that is also later proven incorrect.</blockquote>
</blockquote>
<blockquote type="cite" cite><br></blockquote>
</blockquote>
<blockquote type="cite" cite><br>
<br>
David wrote:<br>
<blockquote type="cite" cite>The difference is that the diagnosis
doesn't have any required<br>
connection to the name under the existing codes, so even if the<br>
diagnosis is accurate, someone else is free to apply the name to a<br>
group that doesn't fit the diagnosis. There has to be a diagnosis,<br>
but it doesn't have to be correct, and it doesn't carry any weight<br>
in assigning the name to a group of organisms. That decision is<br>
entirely subjective, as long as the group contains the type
species.<br>
Under the PhyloCode, the definitions are formulated in such a way<br>
that they unambiguously apply to a single clade in the Tree of
Life.<br>
Our understanding of phylogeny may change, but the definition
always</blockquote>
<blockquote type="cite" cite>points to a single, unambiguous clade,
no matter what the<br>
phylogenetic hypothesis may be.</blockquote>
</blockquote>
<blockquote type="cite" cite><br>
True, but the content of that clade will vary depending on
the</blockquote>
<blockquote type="cite" cite>phylogenetic hypothesis, which was
Gerry's point.</blockquote>
<div><br></div>
<div>Yes, in other words, the old codes don't define group names
(which was my point), so one can change the content or keep it same
as one wishes, and there doesn't need to be any connection to
evolutionary history whatsoever. Arguing that there is some advantage
in this is like arguing that alchemy has an advantage over modern
chemistry, because if someone wants to convert lead into gold, only
the rules of alchemy allow it.</div>
<div><br></div>
<blockquote type="cite" cite><br>
<br>
In another message, David wrote:<br>
<blockquote type="cite" cite>Group names are not defined at all in
the old codes, which is why<br>
they can be applied by anyone in any way they see fit, as long as<br>
they contain the type species for the group.</blockquote>
</blockquote>
<blockquote type="cite" cite><br>
Group names are defined in the old codes, at least operationally,
in</blockquote>
<blockquote type="cite" cite>terms of a rank and a type. For
example, Asteraceae is defined as</blockquote>
<blockquote type="cite" cite>the taxon of family rank that includes
the type of Aster.</blockquote>
<div><br></div>
<div>Exactly my point...the names aren't actually defined. Since
"family rank" is completely subjective, a systematist could
apply the name to any group that contained the type of Aster, as long
as s/he adjusted all the other ranks accordingly. There is no
definition of "family rank" except that it is a category
above genus and below order. In other words, there is no definition
of Asteraceae, and I could say that it was limited to Aster and be
correct, even if no one followed my arrangement ("Asteraceae is
a family of plants containing a single genus, Aster"). Talk
about a change in content of Asteraceae! However, such radical and
subjective changes would not be possible under the PhyloCode, because
Asteraceae would be defined phylogenetically and fixed to a single
clade.</div>
<div><br></div>
<blockquote type="cite" cite><br>
<blockquote type="cite" cite><br></blockquote>
<blockquote type="cite" cite>The new system does indeed remove this
"flexibility" (in other</blockquote>
<blockquote type="cite" cite>words, subjectivity) by adding objective
definitions to group names.<br>
I'm arguing to keep the system basically the same, but remove the<br>
subjectivity by linking the names to evolutionary history (in
other<br>
words, to real historical groups).</blockquote>
</blockquote>
<blockquote type="cite" cite><br>
I think you and I are arguing for exactly the same thing, but you<br>
view the difference between the two systems as merely an
"upgrade",<br>
whereas I agree with Gerry that the difference is more
fundamental.<br>
We are of course each free to view the difference however we want
to,<br>
but the way you describe our differences, it may sound to some as<br>
though we are actually advocating different systems, which I don't<br>
think is the case. We both basically favor the rules for clade
names<br>
embodied in the draft PhyloCode.</blockquote>
<div><br></div>
<div>As long as there is no mandatory symbol for distinguishing names
under the two systems, then I think you are correct for clade names.
I think we differ more in the naming of species. As I understand your
arguments, you want all species named under the PhyloCode to be
distinguishable and different from all species named under the old
system, even though they may look similar for awhile (Quercus-alba,
for instance).</div>
<div><br></div>
<blockquote type="cite" cite><br>
<blockquote type="cite" cite>I do not know what the eventual
convention that adopts the PhyloCode<br>
will choose to do. I do know that if the convention were to adopt
a<br>
set of rules that forced me to make separate descriptions for taxa<br>
under the old and new codes (thus creating parallel and competing<br>
taxonomies), I would no longer have any interest in
participating.</blockquote>
</blockquote>
<blockquote type="cite" cite><br>
Here and elsewhere, you refer to separate descriptions. I
don't<br>
think anyone is suggesting that there be separate descriptions for<br>
taxa under the two systems. Providing a description is a
different<br>
process than defining the name. I would hope that a
description<br>
would be provided under either system when a name is published,
and<br>
there is no reason why it shouldn't be exactly the same
description.</blockquote>
<div><br></div>
<div>If you describe a new species as Quercus-hillisi, that name will
not be valid under the old codes. Therefore, you will have to name
both Quercus-hillisi and Quercus hillisi under Method B and the ICBN,
whereas you would have to name just Quercus hillisi under
Method M and the ICBN (the clade address may be optional under Method
M, but it doesn't make the name wrong). At least for the ICZN, if you
tried this approach with an animal, the whole species description
would be invalid under the ICZN code, because authors of species
descriptions must <i>consistently</i> use binomial nomenclature in
the paper for the name to be considered valid. This is possible under
Method M, but impossible under Method B (or indeed, Methods A-L).
Once the species is described, of course, people could refer to it as
they see fit. Personally, I'd keep using Quercus hillisi unless there
was a compelling reason to do otherwise.</div>
<div><br></div>
<blockquote type="cite" cite><br>
<br>
In yet another message, David wrote:<br>
<blockquote type="cite" cite>Many (the vast majority in my
experience, although I realize that I<br>
have a biased sample) systematists reject paraphyletic groups now,<br>
and only use names that have been given to paraphyletic groups if<br>
the group concept is changed to include all the descendants of
the</blockquote>
<blockquote type="cite" cite>common ancestor. That is the same under
the old and new codes. I<br>
can't remember the last time I heard of someone purposefully<br>
supporting the naming or recognition of a paraphyletic
group.</blockquote>
</blockquote>
<blockquote type="cite" cite><br>
I think that zoologists (at least vertebrate zoologists) are ahead
of<br>
other systematists in this regard. I'm sorry to say there is
still<br>
widespread support for paraphyletic groups among plant
taxonomists.<br>
I still (frequently!) find myself having to defend my rejection of<br>
paraphyletic groups. Perhaps some of the difference in our<br>
perspectives stems from this difference in our disciplines. I
wonder<br>
how invertebrate zoologists stand on this issue. Anyone care
to</blockquote>
<blockquote type="cite" cite>comment?</blockquote>
<div><br></div>
<div>Although I have described more vertebrate taxa than
non-vertebrate taxa, I don't consider myself restricted to vertebrate
zoology. Here at Texas, most of the other systematists I work with
and see everyday are botanists, as are at least half of the students
in my systematics class and systematics discussion group. I have not
seen one of them promote the naming of a paraphyletic group since at
least the 1980s. So, there may well be a bias in the company we
interact with, but it is not a simple difference in taxonomic groups
studied. I simply don't know ANY systematists who promote
paraphyletic groups anymore. I believe you that there are still some
out there someplace, but I haven't met one in decades. I hear that
there are also systematists who are creationists, but I haven't met
any of them recently either, and I don't worry too much about what
they think of incorporating evolution into systematics.</div>
<div><br></div>
<blockquote type="cite" cite><br>
<br>
<blockquote type="cite" cite><br>
Under Method M (and only under Method M among the Cantino et al.<br>
methods), the same species description can meet the requirements
of<br>
both the old and new codes.</blockquote>
</blockquote>
<blockquote type="cite" cite><br>
"Description" again. The same species description can
and should be<br>
provided under both systems no matter which naming method is used.<br>
Where we differ is the form of the name, not whether or not
the</blockquote>
<blockquote type="cite" cite>description is the same.</blockquote>
<div><br></div>
<div>Again, if you publish a description of new species and call it
Quercus-hillisi, that description will have no standing under the old
codes. Thus, you or someone else would have to publish a second
description of Quercus hillisi if you want the name to be used under
the old code. Under Method M, one description fits both purposes. If
you list both names in one description, that is still two names, and
that approach would not be allowed under ICZN rules (I haven't
checked the ICBN rules on this point).</div>
<div><br></div>
<blockquote type="cite" cite><br>
<br>
<blockquote type="cite" cite>Under Phil's preferred system, a
description under the new rules<br>
would not fit the old rules, so one would either have to only<br>
describe species under one set of rules, or publish two<br>
descriptions for the same species (one under each set of
rules).</blockquote>
</blockquote>
<blockquote type="cite" cite><br>
Nonsense! All one would have to do is provide a single
description<br>
(mandatory under the current codes; optional under the PhyloCode),<br>
presumably define the name phylogenetically (if required under the<br>
PhyloCode--this issue has not been resolved), provide the
PhyloCode<br>
registration number, and cite the name in the two slightly
different<br>
forms required by the two codes. For example:<br>
<br>
Hypotheticus novus (ICBN)<br>
Hypotheticus-novus (PhyloCode)<br>
Phylogenetic definition:........<br>
Description:.......<br>
Type specimen:....... [this presumably will be cited as part of
the<br>
definition anyway]<br>
Registration Number:....<br>
<br>
<br>
How would this be done under David's Method M? Almost
identically:<br>
<br>
Hypotheticus novus (ICBN)<br>
novus (PhyloCode)<br>
Phylogenetic definition:........<br>
Description:.......<br>
Type specimen:.......</blockquote>
<div>Registration Number:.....</div>
<div><br></div>
<div>That is not true...under Method M, the single form of the name
Hypotheticus novus is perfectly acceptable under both codes, and
every time you mention the name of the new species in the paper you
could say Hypotheticus novus (or H. novus if you wanted to
abbreviate). Thus, the description would satisfy both codes without
making a mess of the paper or violating any rules about consistent
use of binomial nomenclature. Under Method B, you would either choose
one or the other form (and you'd have to choose the binomial form if
you wanted the description to satisfy the ICZN), or else say
"Hypotheticus novus (ICBN) or Hypotheticus-novus
(PhyloCode)" every time that you mentioned the species. I don't
see this as adding anything except confusion, and I can't imagine
trying to educate an editor on why all that extra verbiage was
needed.</div>
<div> </div>
<div>A minor point: It seems strange to me that you say that a
species description is optional under PhyloCode, since presumably the
rules have not been written for species.</div>
<div><br></div>
<div><br></div>
<div><br></div>
<blockquote type="cite" cite><br>
<br>
There are various pros and cons to both Method M and Method B, but<br>
the distinction that David made in this message is a red herring.<br>
Both methods provide a name that differs in form from a
Linnaean</blockquote>
<blockquote type="cite" cite>binomial, and neither method requires
two descriptions.</blockquote>
<div><br></div>
<div>I've now written descriptions that follow both the ICZN rules
and Method M, and it was easy. I can't see how I could have done the
same with Method B, and I do not believe that it is possible. If you
disagree, Phil, you should publish a species description that meets
the requirements of both the ICBN and Method B and prove me wrong.
Even if it can be done for a plant (which I would find surprising), I
know that it cannot be done for an animal and still satisfy all the
rules of the ICZN. Since I have consistently argued this point in
promoting Method M, it isn't fair to call this point a red herring,
which implies that I prefer Method M for some other, unstated reason.
It is, in fact, an important point with me, and the main reason I see
Method M as the only acceptable method for promoting an easy
transition from the old codes to the new one. My other reason for
preferring Method M is that everyone can continue to use the same
form of names that they are used to, which will make the shift to
PhyloCode names much more transparent and acceptable to the huge
number of users of scientific names (who are unlikely to ever change
to saying Escherichia-coli rather than E. coli, or Homo-sapiens
rather than Homo sapiens). From their standpoint, under method M the
only noticeable change will be that species names will all have
registration numbers that can optionally be used and a connected
database and searching mechanism. In other words, optional
value-added changes.</div>
<div><br></div>
<blockquote type="cite" cite><br>
David will argue that in Method M, the combination of the
PhyloCode<br>
name and a taxonomic address looks identical to the Linnaean
binomial<br>
for that species. This is true, but there is no requirement
that a<br>
taxonomic address be cited or, if one is cited, that it be the
genus<br>
name of the Linnaean binomial. If Quercus alba is converted to
alba<br>
under Method M, the name is alba. Period. Sensible people
will<br>
frequently include the taxonomic address Quercus before the name,
but<br>
there is nothing to stop people from referring to this species
solely<br>
by its name (alba) or from including some other taxonomic address<br>
such as Fagaceae--another clade to which this species belongs.
It is<br>
therefore misleading to argue that under Method M, the name is the<br>
same under both codes.</blockquote>
<div><br></div>
<div>The point is that under Method M, one can describe the species
as Quercus alba and be perfectly within both codes, and this is not
true with any of the other methods. There is nothing that keeps
others from saying just alba if they want to in other papers, but
there is nothing to stop people from doing that now. People
frequently use shorthand for species names now, and don't use the
full name in many non-taxonomic papers when it is not confusing to
abbreviate. All any code can regulate are the papers in which the
names are formed; how they are used subsequently will always depend
upon individuals and the community at large. So, all we are really
concerned about is the original paper in which the species is
described. Method M is the only of the Cantino et al. methods that
allows the use of a single form of the name under both the old and
new codes, and therefore the only method that is forward- and
reverse-compatible under the old and new codes.</div>
<div><br></div>
<div>David</div>
<div><br></div>
<div><br>
David M. Hillis<br>
Director, School of Biological Sciences<br>
Director's office: 512-232-3690 (FAX: 512-232-3699)<br>
Alfred W. Roark Centennial Professor<br>
Section of Integrative Biology<br>
University of Texas<br>
Austin, TX 78712<br>
Research Office: 512-471-5792<br>
Lab: 512-471-5661<br>
FAX: 512-471-3878<br>
E-mail: dhillis@mail.utexas.edu</div>
</body>
</html>