[Previous by date - Nomenclatural Freedom IS the issue]
[Next by date - Apomorphy-based clades; was Re: Panstems]
[Previous by subject - Re: Panstems]
[Next by subject - Re: Panstems]
Date: Tue, 14 Sep 2004 13:06:01 +0200
From: [unknown]
To: phylocode@ouvaxa.cats.ohiou.edu
Subject: Re: Panstems
--Boundary_(ID_rbSxaOZziNe5KqqlC7weJA)
Content-type: text/plain; charset=3D"iso-8859-1" ; format=3D"flowed"
Content-transfer-encoding: quoted-printable
Here are my responses to a few of Kevin's comments on my letter (belo=
w).
> >Of course, the reason that Jacques and I=3D20
>defined the name Synapsida phylogenetically as=3D20
>referring to the total clade of Mammalia is that=3D20
>this use "approximates" the way the name had=3D20
>been used for many years before. However, it's=3D20
>misleading to imply, as Michel does, that the=3D20
>name was uniquely associated with the stem,=3D20
>rather than with an apomorphy (or a node) for=3D20
>the approximatly one hundred years before=3D20
>Jacques and I formulated a phylogenetic=3D20
>definition for it.
=09I didn't meant to imply that the name=3D20
Synapsida was UNIQUELY associated with the=3D20
mammalian stem, but it WAS associated with that=3D20
stem, more than any other name. For me, that's=3D20
enough to retain this name in that sense. I=3D20
gather that for Kevin, that reason is not a good=3D20
enough. My point, like Chris emphasized well, is=3D20
that specialists of early amniotes should make=3D20
that decision, not a few of us, even when they=3D20
are (or especially when they are) authors of the=3D20
PhyloCode. That would immediately trigger=3D20
negative reactions from our colleagues vertebrate=3D20
paleontologists. Perhaps Kevin does not care=3D20
about this, but I do; perhaps he does not agree=3D20
with me on this, but I think that I know my=3D20
colleagues (vertebrate paleontologists) at least=3D20
as well as he does, so I stand by my opinion on=3D20
this.
>The name has been associated with both, and in=3D20
>this particular case the exact association is=3D20
>ambiguous because the known composition of the=3D20
>two clades has been identical (i.e., there are=3D20
>no clear examples of taxa that lack the=3D20
>character but are clearly more closely related=3D20
>to mammals than to birds and turtles). In=3D20
>addition, potential homoplasy in the character=3D20
>is NOT a good reason for avoiding an=3D20
>apomorphy-based definition (and in this case,=3D20
>Michel seems to be disregarding comments that=3D20
>were made at the Paris meeting). If other taxa=3D20
>have evolved the character convegently, those=3D20
>taxa can be excluded by specifying that the=3D20
>condition has to be homologous with that in=3D20
>mammals (or Varanops). If instead the character=3D20
>arose earlier, then we would simply have to=3D20
>conclude that some other early amniotes that=3D20
>have not traditionally been considered synapsids=3D20
>are, in fact, synapsids (changes in hypothesized=3D20
>composition happen all the time).
=09Here, I disagree. If the synapsid=3D20
condition were primitive for amniotes, Synapsida=3D20
could become redundant with Amniota or even=3D20
include Amniota (suppose we find a sister-group=3D20
of Amniota with the synapsid fenestra, that would=3D20
be homologous or synapomorphic with the mammalian=3D20
fenestra). I think that few vertebrate=3D20
paleontologists would think that this is=3D20
irrelevant to how we define the name Synapsida.=3D20
I don't like this possibility (Synapsida more=3D20
inclusive than Amniota) because it would be=3D20
disruptive in that it would represent a drastic=3D20
change in inclusion of Synapsida, at odds with=3D20
all previous usages of this term. I believe that=3D20
the nomenclatural decisions should be made by=3D20
specialists of each group, for this very reason.=3D20
Need I recall Kevin the consequences of his=3D20
informally defining Temnospondyli (de Queiroz &=3D20
Gauthier, 1992: 474) on the basis of extant=3D20
amphibians, with the consequence that under most=3D20
recent phylogenies, this name excludes ALL taxa=3D20
that have traditionally been considered=3D20
temnospondyls? At that time, many thought that=3D20
it was best to define all names on the basis of=3D20
extant taxa; with hindsight, that does not appear=3D20
to be a good idea, and this conclusion has been=3D20
incorporated into the PhyloCode. My point is=3D20
that the PhyloCode is not really about how to=3D20
name clades, but how to define their names. The=3D20
choice of names (and definitions) is best left to=3D20
specialists and I think, for several reasons=3D20
(including those I just gave), the PhyloCode=3D20
should include as few restrictions on how to form=3D20
names as possible, as it now stands. The Pan=3D20
convention seems like a good idea now to some of=3D20
us (like Kevin), but that could turn out to be=3D20
wrong too.
=09I must also point out that many (most?)=3D20
mistakes that were made in the early history of=3D20
PN were made because some people proposed names=3D20
and definitions in groups with which they were=3D20
not very familiar. The result, predictibly=3D20
enough, was that the specifiers were not always=3D20
well chosen, and that the composition of taxa=3D20
thus delimited has fluctuated widely, with dire=3D20
consequences for the reputation and acceptance of=3D20
PN and the PhyloCode. The PhyloCode now=3D20
stipulates that definitions should be proposed by=3D20
scientists familiar with the relevent group, so=3D20
hopefully, such problems will be rare in the=3D20
future, and acceptance of PN and the PhyloCode=3D20
should benefit from all this, IF we leave=3D20
sufficient nomenclatural freedom in the PhyloCode.
>Moreover, even if one doesn't like=3D20
>apomorphy-based definitions, one could also make=3D20
>the argument that the name Synapsida has been no=3D20
>more clearly associated (at least prior to the=3D20
>first phylogenetic definition) with the stem of=3D20
>Mammalia than with the node representing the=3D20
>last common ancestor of Varanops, Eothyris, and=3D20
>Cynognathus. In this context, it should be=3D20
>clear that I was not simply disregarding=3D20
>Michel's point made at the Paris meeting.=3D20
>Instead, I consider that argument incorrect. If=3D20
>one is going to argue that the name Synapsida=3D20
>was unambiguously associated with a particular=3D20
>clade, then one has to cite an explicit phy
>and de Queiroz). Conversely, if one is going to=3D20
>use the work of earlier authors to establish=3D20
>usage, then one is not going to be able to=3D20
>associate the name unambiguously with a=3D20
>particular stem as opposed to a node or=3D20
>apomorphy. The use of this name by earlier=3D20
>authors is highly ambiguous with regard to these=3D20
>three different possible conceptualizations,=3D20
>because the precise distinctions embodied in=3D20
>them generally were not made at that time.
=09Indeed.
>Consequently, the ambiguous association of the=3D20
>name Synapsida with the stem in question by=3D20
>earlier authors is not a good reason for=3D20
>rejecting PanMammalia as the name for the total=3D20
>clade originating from that stem, as an equally=3D20
>good case can be made for a historical=3D20
>association of the name Synapsida with either an=3D20
>apomorphy or a node.
=09I disagree. Few people will want to have=3D20
redundant taxa (apomorphy-based Synapsida and=3D20
stem-based PanMammalia). Previous messages=3D20
emphasized that we should keep things simple=3D20
whenever possible, and I agree. Many people will=3D20
not understand that two names refer to EXACTLY=3D20
the same set of terminal taxa, even though people=3D20
on this list understand the difference. But more=3D20
than anything, as I emphasized above, this is for=3D20
me more an issue of public relations than logics.=3D20
I appreciate the use of a Pan convention for=3D20
linking a crown to a total group, but I believe=3D20
that it will do more harm than good for=3D20
acceptance of the PhyloCode. Most vertebrate=3D20
paleontologists will not like PanMammalia and=3D20
will not like us telling them how to name "their"=3D20
clades. A code, no matter how great it may be,=3D20
is useless if the scientific community does not=3D20
adopt it.
>Indeed, if one is going to follow the usage of=3D20
>earlier authors, Theropsida would seem to be a=3D20
>better choice than Synapsida, given that there=3D20
>is less reason to think that the former name=3D20
>does not describe an apomorphy and was=3D20
>originally associated unambiguously with thel=3D20
>clade in question (rather than with a=3D20
>paraphyletic group as in the case of=3D20
>Synapsida).< =20
>
>=09One of the most common critique of the
>PhyloCode is that it will generate many new names
>and disrupt continuity with the literature; that
>is certainly not true of the PhyloCode as it now
>stands, but it COULD be true if the use of the
>Pan- prefix for total clades is made mandatory
>for most such clades. Let's not give critiques
>of the PhyloCode such an obvious problem to point
>out.
>
>>For the reasons given above, the argument about=3D20
>>disrupting continuity is false.
=09I can't see that. Recommendation 10A of the PhyloCode states:
Recommendation 10A. Clade names should be=3D20
selected in such a way as to minimize disruption=3D20
of current and/or historical usage (with regard=3D20
to composition, diagnostic characters, or both)=3D20
and to maximize access to the literature.=3D20
Therefore, when establishing the name of a clade,=3D20
a preexisting name that has been applied to that=3D20
clade, or to a paraphyletic group stemming from=3D20
the same ancestor, should generally be selected=3D20
if such a name exists.
=09Replacing (at least partly) Synapsida by=3D20
PanMammalia could disrupt nomenclature by=3D20
severing the connection with older literature.=3D20
Suppose that paleontologists use=3D20
(apomorphy-based) Synapsida and neontologists=3D20
start using (stem-based) PanMammalia. It would=3D20
not be long before many authors fail to realize=3D20
that these taxa are nearly the same things.
> Regarding new names, this compaint has been=3D20
>raised repeatedly in the history of taxonomy,=3D20
>yet the proposal of new names never ceases, nor=3D20
>does it cause any new problems beyond the=3D20
>general one of dealing wth ever increading=3D20
>information that characterizes all of human=3D20
>history since the invention of language.=3D20
>Neither of these concerns is well founded, and=3D20
>we should not therefore allow them to constrain=3D20
>us.
=09This seems to say that we have always had=3D20
this problem (of new, unneeded names being=3D20
erected), so let's keep it that way. I disagree;=3D20
we can do better, and we should.
=09Sincerely,
=09Michel
--=20
Michel Laurin
=3D46RE 2696, CNRS
Universit=3DE9 Paris 7 - Denis Diderot
2, place Jussieu
case 7077
75005 Paris
=3D46RANCE
tel. (33 1) 44 27 36 92
http://tolweb.org/tree/laurin/Laurin_Home_page.html
Secretary of the International Society for Phylogenetic Nomenclature=
=3D
--Boundary_(ID_rbSxaOZziNe5KqqlC7weJA)
Content-type: text/html; charset=3D"iso-8859-1"
Content-transfer-encoding: quoted-printable
<!doctype html public "-//W3C//DTD W3 HTML//EN">
<html><head><style type=3D3D"text/css"><!--
blockquote, dl, ul, ol, li { padding-top: 0 ; padding-bottom: 0 }
--></style><title>Re: Panstems</title></head><body>
<div>Here are my responses to a few of Kevin's comments on my letter
(below).</div>
<div><br></div>
<blockquote type=3D3D"cite" cite>>Of course, the reason that Jacqu=
es
and I defined the name Synapsida phylogenetically as referring to the
total clade of Mammalia is that this use "approximates" the
way the name had been used for many years before. However, it's
misleading to imply, as Michel does, that the name was uniquely
associated with the stem, rather than with an apomorphy (or a node)
for the approximatly one hundred years before Jacques and I formulate=
d
a phylogenetic definition for it. </blockquote>
<div><br></div>
<div><x-tab> </x-tab>I
didn't meant to imply that the name Synapsida was UNIQUELY associated
with the mammalian stem, but it WAS associated with that stem, more
than any other name. For me, that's enough to retain this name
in that sense. I gather that for Kevin, that reason is not a
good enough. My point, like Chris emphasized well, is that
specialists of early amniotes should make that decision, not a few of
us, even when they are (or especially when they are) authors of the
PhyloCode. That would immediately trigger negative reactions
=66rom our colleagues vertebrate paleontologists. Perhaps Kevin
does not care about this, but I do; perhaps he does not agree with me
on this, but I think that I know my colleagues (vertebrate
paleontologists) at least as well as he does, so I stand by my opinio=
n
on this.</div>
<div><br></div>
<blockquote type=3D3D"cite" cite>The name has been associated with bo=
th,
and in this particular case the exact association is ambiguous becaus=
e
the known composition of the two clades has been identical (i.e.,
there are no clear examples of taxa that lack the character but are
clearly more closely related to mammals than to birds and turtles).&n=
bsp;
In addition, potential homoplasy in the character is NOT a good reaso=
n
for avoiding an apomorphy-based definition (and in this case, Michel
seems to be disregarding comments that were made at the Paris
meeting). If other taxa have evolved the character convegently,
those taxa can be excluded by specifying that the condition has to be
homologous with that in mammals (or Varanops). If instead the
character arose earlier, then we would simply have to conclude that
some other early amniotes that have not traditionally been considered
synapsids are, in fact, synapsids (changes in hypothesized compositio=
n
happen all the time). </blockquote>
<div><br></div>
<div><x-tab> </x-tab>Here, =
I
disagree. If the synapsid condition were primitive for amniotes=
,
Synapsida could become redundant with Amniota or even include Amniota
(suppose we find a sister-group of Amniota with the synapsid fenestra=
,
that would be homologous or synapomorphic with the mammalian
fenestra). I think that few vertebrate paleontologists would
think that this is irrelevant to how we define the name Synapsida.&nb=
sp;
I don't like this possibility (Synapsida more inclusive than Amniota)
because it would be disruptive in that it would represent a drastic
change in inclusion of Synapsida, at odds with all previous usages of
this term. I believe that the nomenclatural decisions should be
made by specialists of each group, for this very reason. Need I
recall Kevin the consequences of his informally defining Temnospondyl=
i
(de Queiroz & Gauthier, 1992: 474) on the basis of extant
amphibians, with the consequence that under most recent phylogenies,
this name excludes ALL taxa that have traditionally been considered
temnospondyls? At that time, many thought that it was best to
define all names on the basis of extant taxa; with hindsight, that
does not appear to be a good idea, and this conclusion has been
incorporated into the PhyloCode. My point is that the PhyloCode
is not really about how to name clades, but how to define their
names. The choice of names (and definitions) is best left to
specialists and I think, for several reasons (including those I just
gave), the PhyloCode should include as few restrictions on how to for=
m
names as possible, as it now stands. The Pan convention seems
like a good idea now to some of us (like Kevin), but that could turn
out to be wrong too.</div>
<div><br></div>
<div><x-tab> </x-tab>I must
also point out that many (most?) mistakes that were made in the early
history of PN were made because some people proposed names and
definitions in groups with which they were not very familiar.
The result, predictibly enough, was that the specifiers were not
always well chosen, and that the composition of taxa thus delimited
has fluctuated widely, with dire consequences for the reputation and
acceptance of PN and the PhyloCode. The PhyloCode now stipulate=
s
that definitions should be proposed by scientists familiar with the
relevent group, so hopefully, such problems will be rare in the
future, and acceptance of PN and the PhyloCode should benefit from al=
l
this, IF we leave sufficient nomenclatural freedom in the
PhyloCode.</div>
<div><br></div>
<blockquote type=3D3D"cite" cite>Moreover, even if one doesn't like
apomorphy-based definitions, one could also make the argument that th=
e
name Synapsida has been no more clearly associated (at least prior to
the first phylogenetic definition) with the stem of Mammalia than wit=
h
the node representing the last common ancestor of Varanops, Eothyris,
and Cynognathus. In this context, it should be clear that I was
not simply disregarding Michel's point made at the Paris meeting.
Instead, I consider that argument incorrect. If one is going to
argue that the name Synapsida was unambiguously associated with a
particular clade, then one has to cite an explicit phy<br>
and de Queiroz). Conversely, if one is going to use the work of
earlier authors to establish usage, then one is not going to be able
to associate the name unambiguously with a particular stem as opposed
to a node or apomorphy. The use of this name by earlier authors
is highly ambiguous with regard to these three different possible
conceptualizations, because the precise distinctions embodied in them
generally were not made at that time. </blockquote>
<div><br>
<x-tab> </x-tab>Indeed.<br>
</div>
<blockquote type=3D3D"cite" cite>Consequently, the ambiguous associat=
ion
of the name Synapsida with the stem in question by earlier authors is
not a good reason for rejecting PanMammalia as the name for the total
clade originating from that stem, as an equally good case can be made
for a historical association of the name Synapsida with either an
apomorphy or a node. </blockquote>
<div><br></div>
<div><x-tab> </x-tab>I
disagree. Few people will want to have redundant taxa
(apomorphy-based Synapsida and stem-based PanMammalia). Previou=
s
messages emphasized that we should keep things simple whenever
possible, and I agree. Many people will not understand that two
names refer to EXACTLY the same set of terminal taxa, even though
people on this list understand the difference. But more than
anything, as I emphasized above, this is for me more an issue of
public relations than logics. I appreciate the use of a Pan
convention for linking a crown to a total group, but I believe that i=
t
will do more harm than good for acceptance of the PhyloCode.
Most vertebrate paleontologists will not like PanMammalia and will no=
t
like us telling them how to name "their" clades. A
code, no matter how great it may be, is useless if the scientific
community does not adopt it.</div>
<div><br></div>
<blockquote type=3D3D"cite" cite>Indeed, if one is going to follow th=
e
usage of earlier authors, Theropsida would seem to be a better choice
than Synapsida, given that there is less reason to think that the
former name does not describe an apomorphy and was originally
associated unambiguously with thel clade in question (rather than wit=
h
a paraphyletic group as in the case of
Synapsida).< <br>
<br>
<x-tab> </x-tab>One of the most
common critique of the<br>
PhyloCode is that it will generate many new names<br>
and disrupt continuity with the literature; that<br>
is certainly not true of the PhyloCode as it now<br>
stands, but it COULD be true if the use of the<br>
Pan- prefix for total clades is made mandatory<br>
for most such clades. Let's not give critiques<br>
of the PhyloCode such an obvious problem to point<br>
out.<br>
<br>
>For the reasons given above, the argument about disrupting
continuity is false.</blockquote>
<div><br></div>
<div><x-tab> </x-tab>I can'=
t
see that. Recommendation 10A of the PhyloCode states:</div>
<div><br></div>
<div><font face=3D3D"Times New Roman" size=3D3D"-1"
color=3D3D"#000000">Recommendation 10A. Clade names should be selecte=
d in
such a way as to minimize disruption of current and/or historical
usage (with regard to composition, diagnostic characters, or both) an=
d
to maximize access to the literature. Therefore, when establishing th=
e
name of a clade, a preexisting name that has been applied to that
clade, or to a paraphyletic group stemming from the same ancestor,
should generally be selected if such a name exists.</font></div>
<div><br></div>
<div><x-tab>
</x-tab>Replacing (at least partly) Synapsida by PanMammalia could
disrupt nomenclature by severing the connection with older
literature. Suppose that paleontologists use (apomorphy-based)
Synapsida and neontologists start using (stem-based) PanMammalia.&nbs=
p;
It would not be long before many authors fail to realize that these
taxa are nearly the same things. </div>
<div><br></div>
<blockquote type=3D3D"cite" cite> Regarding new names, this comp=
aint
has been raised repeatedly in the history of taxonomy, yet the
proposal of new names never ceases, nor does it cause any new problem=
s
beyond the general one of dealing wth ever increading information tha=
t
characterizes all of human history since the invention of language.&n=
bsp;
Neither of these concerns is well founded, and we should not therefor=
e
allow them to constrain us.</blockquote>
<div><br></div>
<div><x-tab> </x-tab>This
seems to say that we have always had this problem (of new, unneeded&n=
bsp;
names being erected), so let's keep it that way. I disagree; we
can do better, and we should.</div>
<div><br></div>
<div><x-tab>
</x-tab>Sincerely,</div>
<div><br></div>
<div><x-tab>
</x-tab>Michel</div>
<x-sigsep><pre>--=20
</pre></x-sigsep>
<div><font color=3D3D"#000000">Michel Laurin<br>
=3D46RE 2696, CNRS<br>
Universit=3DE9 Paris 7 - Denis Diderot<br>
2, place Jussieu<br>
case 7077<br>
75005 Paris<br>
=3D46RANCE<br>
<br>
tel. (33 1) 44 27 36 92<br>
http://tolweb.org/tree/laurin/Laurin_Home_page.html</font><br>
<font color=3D3D"#000000"></font></div>
<div><font color=3D3D"#000000">Secretary of the<i> International Soci=
ety
for Phylogenetic Nomenclature</i></font></div>
</body>
</html>=3D
--Boundary_(ID_rbSxaOZziNe5KqqlC7weJA)--