The two ways of organizing evolutionary information: Linnean and cladistic

[Peter Nyikos]

The long-running thread, "Did we come from monkeys?" has morphed into
a comparison between the two systems of classification that have come
down to us: the traditional ("Linnean") and the cladistic ("phylogenetic").

I thought it a good idea to present, first, a "primer" on what these
two systems are, and then to get down to the nitty-gritty of their
largely complementary nature. First, the basic concepts:

clade - a group of organisms that consist of a single species and all
its descendants, extant or extinct

paraphyletic - consisting of a clade with one or more clades removed.

Two examples: 1. the old *Amphibia* (as in "The Age of Amphibians")
of the Linnean system was a paraphyletic group which consisted of a
clade of primitive terrestrial animals, minus the clade that is now
called *Amniota*.

2. the old *Reptilia* (as in "The Age of Reptiles") consisted of
that same *Amniota* with both *Mammalia* and *Aves* (birds) removed,
and so it too was paraphyletic.

An important feature of both kinds of groups is that they all
include descendants of one species. In the case of paraphyletic
groups, the Linnean ones were only missing clades, and not haphazard
groups of creatures.

The Linnean system was the dominant one until the last quarter
or so of the 20th century, the only one with which people were
familiar when most participants to that "...monkeys?" thread
were growing up. Then came what is known as "The Cladist
Wars" and the system that uses clades and only clades has
won, and the Linnean is being slowly driven to extinction.

Which is a shame, because the two systems are by nature
complementary and each makes up for certain disadvantages of
the other. My next post to this thread will review
some of what has already transpired on this theme.

Peter Nyikos
Professor, Dept. of Mathematics -- standard disclaimer--
University of South Carolina
http://people.math.sc.edu/nyikos/

[John Harshman]

On 11/17/15 5:46 PM, Peter Nyikos wrote:
> The long-running thread, "Did we come from monkeys?" has morphed into
> a comparison between the two systems of classification that have come
> down to us: the traditional ("Linnean") and the cladistic ("phylogenetic").

No, you have once again conflated two questions: the value of ranked
groups and the value of clades vs. paraphyetic groups. "Linnean" refers
to the former question, while below you discuss only the latter.

> I thought it a good idea to present, first, a "primer" on what these
> two systems are, and then to get down to the nitty-gritty of their
> largely complementary nature. First, the basic concepts:
>
> clade - a group of organisms that consist of a single species and all
> its descendants, extant or extinct
>
> paraphyletic - consisting of a clade with one or more clades removed.
>
> Two examples: 1. the old *Amphibia* (as in "The Age of Amphibians")
> of the Linnean system was a paraphyletic group which consisted of a
> clade of primitive terrestrial animals, minus the clade that is now
> called *Amniota*.
>
> 2. the old *Reptilia* (as in "The Age of Reptiles") consisted of
> that same *Amniota* with both *Mammalia* and *Aves* (birds) removed,
> and so it too was paraphyletic.
>
> An important feature of both kinds of groups is that they all
> include descendants of one species. In the case of paraphyletic
> groups, the Linnean ones were only missing clades, and not haphazard
> groups of creatures.
>
> The Linnean system was the dominant one until the last quarter
> or so of the 20th century, the only one with which people were
> familiar when most participants to that "...monkeys?" thread
> were growing up. Then came what is known as "The Cladist
> Wars" and the system that uses clades and only clades has
> won, and the Linnean is being slowly driven to extinction.

The Cladist Wars were about two things: the best method for
classification and what classifications ought to represent. There were
three sides: cladists, traditional "evolutionary" systematists, and
pheneticists. The latter group denied that we should try to recover
evolutionary relationships at all and just classify based on overall
similarity.

> Which is a shame, because the two systems are by nature
> complementary and each makes up for certain disadvantages of
> the other. My next post to this thread will review
> some of what has already transpired on this theme.

I hope you will be a bit more accurate next time.

[Peter Nyikos]

On Tuesday, November 17, 2015 at 8:49:13 PM UTC-5, Peter Nyikos wrote:
> The long-running thread, "Did we come from monkeys?" has morphed into
> a comparison between the two systems of classification that have come
> down to us: the traditional ("Linnean") and the cladistic ("phylogenetic").
>
> I thought it a good idea to present, first, a "primer" on what these
> two systems are, and then to get down to the nitty-gritty of their
> largely complementary nature.

The Linnean system's great strength lies in the way it classifies all
organisms, both extant and extinct, in a nested hierarchy starting
with the species, and then in ever-expanding circles we have its
genus, family, order, class, phylum, kingdom, and domain. [There are
intermediate ranks, like subfamily, superfamily, infraclass, subclass,
etc.] To keep things simple, I will confine myself mainly to animals
and indeed to vertebrates.

Each rank comes with a rough measure of disparity -- the amount
of variation allowed between members of the same taxon. Two living members
of the family *Hominidae*, for instance, are less disparate from each
other than the two most disparate members of the old Linnean class
*Amphibia* were from each other -- or even than the two most disparate
living amphibians are from each other.

These rankings are never better than ballpark estimates of disparity,
but they are reasonably good for gauging e.g. how close we are to the
true ancestral species of a clade.

Cladistic ("phylogenetic") classification could also give us this
same nested hierarchy where LIVING creatures are concerned, but
it gets more and more sketchy the further back one goes into
the eons. As I told Richard Norman on the "...monkeys?" thread,

The nested hierarchy for Tiktaalik stops with the clade determined
by it and yourself and doesn't resume until you get all the way
down to Tiktaalik itself. [Cladists banished its family,
order and even class to oblivion. Not to mention subfamily, superfamily,
infraorder, suborder, etc. [By the way, I doubt that there
is an appropriate "rank" for its BIG clade to be-- "superclass"
probably deserves to go down only as far as the LCA of Amniota,
or at best the LCA of Tetrapoda.]

Now, a Linnean could put Tiktaalik in the same order, and perhaps
also the same family, as *Acathostega* and *Ichthyostega*, and
also in the same family as the last common ancestor (LCA) of
the old Linnean *Amphibia*, just as the family *Archaeopterygidae*
was widely considered to be the birthplace of the class *Aves*.


In fairness, I should add that the Linnean system has two big
drawbacks. One, it is useless for reconstructing the Tree of Life
without lots of supplementary information. The phylogenetic
system, on the other hand, is tailor-made for this task, provided
the information we use is accurate.

Two, it arbitrarily cuts off a taxon from its parent taxon, and
it gets worse the farther out one goes. Thus, in the above example,
some fish awfully close to Tiktaalik in their anatomy are arbitrarily
put into a different paraphyletic class, *Osteichthyes* ("bony fish")
from the one into which I've put Tiktaalik.

Here, however, cladistic systematics suffers from the same drawback,
only more so. It simply declares that, IF Tiktaalik happens to
be our direct ancestor, then it is more distantly related to its
immediate ancestor than it is to us. And, in general, every
creature that ever lived was more distantly related to its
immediate ancestor than it is to its remotest descendants.

Peter Nyikos
Professor, Dept. of Mathematics -- standard disclaimer--
University of South Carolina
http://people.math.sc.edu/nyikos/

nyikos "at" math.sc.edu

[Peter Nyikos]

I see Harshman has already weighed in on this thread, but
I have some unfinished business with him from his last post
on the "...monkeys?" thread which segues well with what I've
posted so far on this one.

On Monday, November 16, 2015 at 11:54:16 AM UTC-5, John Harshman wrote:
https://groups.google.com/forum/#!original/talk.origins/rehTL2kf_jU/2wbrz-5WBAAJ
> On 11/13/15 5:02 PM, Peter Nyikos wrote:
> > On Thursday, November 12, 2015 at 11:44:29 AM UTC-5, John Harshman wrote:
> >> On 11/12/15 7:26 AM, Peter Nyikos wrote:
> >>
> >>> Once ancestry is resolved down to the subfamily level, one
> >>> cannot expect more refinement unless many times more fossils
> >>> are found as we now have. The family *Equidae* is almost unique
> >>> in being able to have very strong hypotheses at the genus level.
> >>>
> >>> And, as anyone reading Popper knows, very strong hypotheses are
> >>> the best one can hope for in phylogenetic reconstruction. Those
> >>> 15 (fifteen) rival trees on Ornithodira show just how far the
> >>> triumphantly victorious cladophiles are from getting even
> >>> that degree of "certainty".
> >>
> >> The prior certainly was illusory,
> >
> > There WAS no prior certainty, no more than attends any phylogenetic
> > tree published nowadays. How could you have missed that point,
> > when I said "...cannot expect more refinement..."? It's just that
> > they didn't venture into the region of uncertainty.

<snip unreflective tit-for-tat by Harshman>

> >> as a great many possible trees were
> >> hidden in that "subfamily" classification. The cladistic equivalent,
> >> which would be much more explicit, would be a polytomy. And this would
> >> have actual, visible support from data.
> >
> > Not equivalent at all. Polytomy gives a very different category
> > of information than the use of a subfamily.
>
> So you may claim.

Herr Harschmann ist ein geborener Zweifler.

> What category of information?

Polytomy is for a case where there is too much uncertainty as to
*which* of two different speciation events came first. The use of
a subfamily is for a case where there is too much uncertainty as
to whether any of the fossils in our possession is a good
candidate for the ancestral genus of a clade-- or indeed whether
any of them are.

And it gives some idea of how disparate the true ancestral genus
could be from the ones in the subfamily -- too much and you don't
narrow down to a subfamily at all, but to the family that contains
it, or the superfamily that contains it, etc.

Once upon a time you told me that there are measures of disparity.
But all through the thread where you posted this,
you have acted as though no such things ever existed.
You did that far below in this post, so far down that it
will be dealt with only in my second (and penultimate) reply.

<snip for focus>

> > Cladistic classification does a perfect job of letting you get the
> > topology of the tree [1], but that is about all that it is good for.
> > It gives you no idea of how disparate [2] the organisms in a given
> > clade are. A huge number of branches, as in that (sub?)order of
> > Cambrian trilobites I mentioned another day, measures diversity
> > [i.e, sheer number of species] and not disparity. That order
> > is probably a lot less disparate than the class of placental mammals.
>
> I suspect your views are biased by being a placental mammal yourself,
> but never mind.

Yes, you never did use your mind for that suspicion. :-)

Now, if you want to claim that the group of Cambrian trilobites
I mentioned was at least as disparate as the (infraclass, actually)
of placental mammals, I'm willing to listen to any explanation.

> I agree that cladistic classification deals only with
> cladistic relationships, not with similarity. But while you think that's
> a bad thing, I think it's a good thing.

I don't think it's a bad thing at all; in fact, it has lots to recommend it.
But then, so does the traditional system--but the two lots are very
different, and I say, "Vive la difference!"

Continued in next reply, to be done as soon as I've seen this one
has posted.

Peter Nyikos
Professor, Dept. of Math. -- standard disclaimer--
U. of S. Carolina, Columbia SC
http://people.math.sc.edu/nyikos/

[Peter Nyikos]

On Monday, November 16, 2015 at 11:54:16 AM UTC-5, John Harshman wrote:
https://groups.google.com/forum/#!original/talk.origins/rehTL2kf_jU/2wbrz-5WBAAJ
> On 11/13/15 5:02 PM, Peter Nyikos wrote:

> > You have
> > simply got it into your head that letting people know about
> > the disparity in groups -- be they paraphyletic or clades --
> > is something a classification system SHOULD NOT be concerned with.
>
> Correct.
>
> > But I don't think even a genius with an IQ of 180 and all the knowledge
> > Romer or Carroll could come up with a method anywhere near as good
> > as the traditional system for conveying comprehensive information.
> > While comparison of disparities in different families must always
> > be imprecise, nesting goes a long way towards getting rough and ready
> > estimates. For instance...
>
> That's certainly your opinion.

You are a polemicist first, a propagandist second, and a reasoner
a distant third, so of course you didn't try to come up with
any alternative method of organizing the information we have
on the disparity of organisms in a manner useful to people.

Instead, you reached into your polemical bag labeled "opinion"
and pulled out the best sentence that fit.

> > *Equinae* is less disparate than *Equidae* which in turn is less disparate
> > than *Equioidea* which in turn is less disparate than *Perissodactyla*
> > which in turn is MORE disparate than *Tapiroidea which in turn is
> > MORE disparate than *Tapiridae*, etc. etc. And someone familiar with the
> > whole order might even venture to compare the disparity in *Equidae* with
> > that in *Tapiridae* enough to be able to say that neither is as low in
> > disparity as a subfamily of the other.
>
> All you did above, until the last comparison, is note that groups are
> nested within groups.

All you are doing here is using a literally false sentence for effect;
and next, a sentence that refers to the false picture you painted:

> Cladistic classification does that too.

> Your final
> comparison, of non-nested groups at the same rank, is I think
> fallacious; such taxa have no clear comparability.

So much for alleged measures of disparity, eh?

> Further, you are once
> again confusing two issues: ranking and paraphyly.

"confusing" is false, and "once again" also, unless you can point to
a *valid* prior example.

Paraphyly is an indispensible concomitant to any attempt to give
any ranking of animals as far back as Tiktaalik. I've started
to give people a feel for that in my second post to this thread,
but I expect you and at least one other active participant
on the old thread to find it very unclear, and to request that
I stop beating around the bush and to state my point.

Remainder deleted, to be replied to tomorrow, with the reply depending
on feedback to this one and the preceding one.

Peter Nyikos
Professor, Dept. of Math. -- standard disclaimer--

U. of South Carolina, Columbia SC
http://people.math.sc.edu/nyikos/

[John Harshman]

On 11/17/15 6:42 PM, Peter Nyikos wrote:
> On Tuesday, November 17, 2015 at 8:49:13 PM UTC-5, Peter Nyikos wrote:
>> The long-running thread, "Did we come from monkeys?" has morphed into
>> a comparison between the two systems of classification that have come
>> down to us: the traditional ("Linnean") and the cladistic ("phylogenetic").
>>
>> I thought it a good idea to present, first, a "primer" on what these
>> two systems are, and then to get down to the nitty-gritty of their
>> largely complementary nature.
>
> The Linnean system's great strength lies in the way it classifies all
> organisms, both extant and extinct, in a nested hierarchy starting
> with the species, and then in ever-expanding circles we have its
> genus, family, order, class, phylum, kingdom, and domain. [There are
> intermediate ranks, like subfamily, superfamily, infraclass, subclass,
> etc.] To keep things simple, I will confine myself mainly to animals
> and indeed to vertebrates.

So far, so good. Except that the Linnean system could be cladistic, as I
mentioned before, and that nested hierarchy is not a difference between
systems. So that great strength isn't in fact a point of difference.

> Each rank comes with a rough measure of disparity -- the amount
> of variation allowed between members of the same taxon. Two living members
> of the family *Hominidae*, for instance, are less disparate from each
> other than the two most disparate members of the old Linnean class
> *Amphibia* were from each other -- or even than the two most disparate
> living amphibians are from each other.

Now this just isn't true. There is no rough measure of disparity worth
considering.

> These rankings are never better than ballpark estimates of disparity,
> but they are reasonably good for gauging e.g. how close we are to the
> true ancestral species of a clade.

It isn't even clear what that would mean.

> Cladistic ("phylogenetic") classification could also give us this
> same nested hierarchy where LIVING creatures are concerned, but
> it gets more and more sketchy the further back one goes into
> the eons. As I told Richard Norman on the "...monkeys?" thread,
>
> The nested hierarchy for Tiktaalik stops with the clade determined
> by it and yourself and doesn't resume until you get all the way
> down to Tiktaalik itself. [Cladists banished its family,
> order and even class to oblivion. Not to mention subfamily, superfamily,
> infraorder, suborder, etc. [By the way, I doubt that there
> is an appropriate "rank" for its BIG clade to be-- "superclass"
> probably deserves to go down only as far as the LCA of Amniota,
> or at best the LCA of Tetrapoda.]

I also have no idea what that means. Stops? Resume?

> Now, a Linnean could put Tiktaalik in the same order, and perhaps
> also the same family, as *Acathostega* and *Ichthyostega*, and
> also in the same family as the last common ancestor (LCA) of
> the old Linnean *Amphibia*, just as the family *Archaeopterygidae*
> was widely considered to be the birthplace of the class *Aves*.

Again you confuse Linnean (ranked) classification with paraphyly. These
are orthogonal matters.

> In fairness, I should add that the Linnean system has two big
> drawbacks. One, it is useless for reconstructing the Tree of Life
> without lots of supplementary information. The phylogenetic
> system, on the other hand, is tailor-made for this task, provided
> the information we use is accurate.
>
> Two, it arbitrarily cuts off a taxon from its parent taxon, and
> it gets worse the farther out one goes. Thus, in the above example,
> some fish awfully close to Tiktaalik in their anatomy are arbitrarily
> put into a different paraphyletic class, *Osteichthyes* ("bony fish")
> from the one into which I've put Tiktaalik.

Correct.

> Here, however, cladistic systematics suffers from the same drawback,
> only more so. It simply declares that, IF Tiktaalik happens to
> be our direct ancestor, then it is more distantly related to its
> immediate ancestor than it is to us. And, in general, every
> creature that ever lived was more distantly related to its
> immediate ancestor than it is to its remotest descendants.

You consider that a problem. I don't. We accurately represent cladistic
relationships, but as you point out, it's impossible to accurately
represent simple similarity, as we will always arbitrarily divide
similar species. Classification should pick the things we can do over
those we can't.

[John Harshman]

There was no tit for tat. Nor was it unreflective. Your argument by
adjective annoys me.

>>>> as a great many possible trees were
>>>> hidden in that "subfamily" classification. The cladistic equivalent,
>>>> which would be much more explicit, would be a polytomy. And this would
>>>> have actual, visible support from data.
>>>
>>> Not equivalent at all. Polytomy gives a very different category
>>> of information than the use of a subfamily.
>>
>> So you may claim.
>
> Herr Harschmann ist ein geborener Zweifler.

That would be "Hirschmann".

>> What category of information?
>
> Polytomy is for a case where there is too much uncertainty as to
> *which* of two different speciation events came first. The use of
> a subfamily is for a case where there is too much uncertainty as
> to whether any of the fossils in our possession is a good
> candidate for the ancestral genus of a clade-- or indeed whether
> any of them are.

Same result, really. It's the same uncertainty.

> And it gives some idea of how disparate the true ancestral genus
> could be from the ones in the subfamily -- too much and you don't
> narrow down to a subfamily at all, but to the family that contains
> it, or the superfamily that contains it, etc.

I deny that it gives any such idea.

> Once upon a time you told me that there are measures of disparity.
> But all through the thread where you posted this,
> you have acted as though no such things ever existed.
> You did that far below in this post, so far down that it
> will be dealt with only in my second (and penultimate) reply.

Of course there are such measures. They suffer from the same weaknesses
as measures of similarity: they can vary radically depending on choice
of data.

> <snip for focus>
>
>>> Cladistic classification does a perfect job of letting you get the
>>> topology of the tree [1], but that is about all that it is good for.
>>> It gives you no idea of how disparate [2] the organisms in a given
>>> clade are. A huge number of branches, as in that (sub?)order of
>>> Cambrian trilobites I mentioned another day, measures diversity
>>> [i.e, sheer number of species] and not disparity. That order
>>> is probably a lot less disparate than the class of placental mammals.
>>
>> I suspect your views are biased by being a placental mammal yourself,
>> but never mind.
>
> Yes, you never did use your mind for that suspicion. :-)
>
> Now, if you want to claim that the group of Cambrian trilobites
> I mentioned was at least as disparate as the (infraclass, actually)
> of placental mammals, I'm willing to listen to any explanation.

I merely say that you haven't supported your claim and that your
subjective impression can be colored by your proximity to one of the taxa.

>> I agree that cladistic classification deals only with
>> cladistic relationships, not with similarity. But while you think that's
>> a bad thing, I think it's a good thing.
>
> I don't think it's a bad thing at all; in fact, it has lots to recommend it.
> But then, so does the traditional system--but the two lots are very
> different, and I say, "Vive la difference!"

Maintaining two systems in parallel is impractical. And the virtues of
the traditional system are either maintained in the modern system or
were illusory.

[John Harshman]

On 11/17/15 7:36 PM, Peter Nyikos wrote:
> On Monday, November 16, 2015 at 11:54:16 AM UTC-5, John Harshman wrote:
> https://groups.google.com/forum/#!original/talk.origins/rehTL2kf_jU/2wbrz-5WBAAJ
>> On 11/13/15 5:02 PM, Peter Nyikos wrote:
>
>>> You have
>>> simply got it into your head that letting people know about
>>> the disparity in groups -- be they paraphyletic or clades --
>>> is something a classification system SHOULD NOT be concerned with.
>>
>> Correct.
>>
>>> But I don't think even a genius with an IQ of 180 and all the knowledge
>>> Romer or Carroll could come up with a method anywhere near as good
>>> as the traditional system for conveying comprehensive information.
>>> While comparison of disparities in different families must always
>>> be imprecise, nesting goes a long way towards getting rough and ready
>>> estimates. For instance...
>>
>> That's certainly your opinion.
>
> You are a polemicist first, a propagandist second, and a reasoner
> a distant third, so of course you didn't try to come up with
> any alternative method of organizing the information we have
> on the disparity of organisms in a manner useful to people.

And you like your accusations.

> Instead, you reached into your polemical bag labeled "opinion"
> and pulled out the best sentence that fit.

If you want information on disparity, some kind of morphospace plot
would seem to be the most useful, or perhaps an index of disparity
abstracted from the plot. As I have mentioned, trying to capture both
phylogeny and disparity at once fails in both missions.

>>> *Equinae* is less disparate than *Equidae* which in turn is less disparate
>>> than *Equioidea* which in turn is less disparate than *Perissodactyla*
>>> which in turn is MORE disparate than *Tapiroidea which in turn is
>>> MORE disparate than *Tapiridae*, etc. etc. And someone familiar with the
>>> whole order might even venture to compare the disparity in *Equidae* with
>>> that in *Tapiridae* enough to be able to say that neither is as low in
>>> disparity as a subfamily of the other.
>>
>> All you did above, until the last comparison, is note that groups are
>> nested within groups.
>
> All you are doing here is using a literally false sentence for effect;
> and next, a sentence that refers to the false picture you painted:

More accusations. You should stop that. That sentence is true. By
definition a group Y that includes X must be more disparate than X, and
that's all you show.

>> Cladistic classification does that too.
>
>> Your final
>> comparison, of non-nested groups at the same rank, is I think
>> fallacious; such taxa have no clear comparability.
>
> So much for alleged measures of disparity, eh?

What measures of disparity? You have mentioned none.

>> Further, you are once
>> again confusing two issues: ranking and paraphyly.
>
> "confusing" is false, and "once again" also, unless you can point to
> a *valid* prior example.

You have done it many times in this thread and the one in
sci.bio.paleontology. I have tried to point it out each time.

> Paraphyly is an indispensible concomitant to any attempt to give
> any ranking of animals as far back as Tiktaalik.

No, it is not. But was that your point at last? What you might mean is
that it's indispensible if you want to make all groups of the same rank
equally disparate. But nobody has ever done that before, so I don't know
how you can take that as your model.

> I've started
> to give people a feel for that in my second post to this thread,
> but I expect you and at least one other active participant
> on the old thread to find it very unclear, and to request that
> I stop beating around the bush and to state my point.

Which you didn't actually do, did you?

[jillery]

On Tue, 17 Nov 2015 17:46:06 -0800 (PST), Peter Nyikos
<nyi...@bellsouth.net> wrote:

>The long-running thread, "Did we come from monkeys?" has morphed into
>a comparison between the two systems of classification that have come
>down to us: the traditional ("Linnean") and the cladistic ("phylogenetic").

Actually that's how it first started, along with their relationship to
vernacular terms.

--
This space is intentionally not blank.

[Peter Nyikos]

On Wednesday, November 18, 2015 at 1:04:12 AM UTC-5, John Harshman wrote:
> On 11/17/15 6:42 PM, Peter Nyikos wrote:
> > On Tuesday, November 17, 2015 at 8:49:13 PM UTC-5, Peter Nyikos wrote:
> >> The long-running thread, "Did we come from monkeys?" has morphed into
> >> a comparison between the two systems of classification that have come
> >> down to us: the traditional ("Linnean") and the cladistic ("phylogenetic").
> >>
> >> I thought it a good idea to present, first, a "primer" on what these
> >> two systems are, and then to get down to the nitty-gritty of their
> >> largely complementary nature.
> >
> > The Linnean system's great strength lies in the way it classifies all
> > organisms, both extant and extinct, in a nested hierarchy starting
> > with the species, and then in ever-expanding circles we have its
> > genus, family, order, class, phylum, kingdom, and domain. [There are
> > intermediate ranks, like subfamily, superfamily, infraclass, subclass,
> > etc.] To keep things simple, I will confine myself mainly to animals
> > and indeed to vertebrates.
>
> So far, so good. Except that the Linnean system could be cladistic,

Not without becoming a weak version of the phylogenetic.

> as I mentioned before, and that nested hierarchy is not a difference between
> systems.

Its sheer existence is not, but the two take on radically different shapes.
See below where I quote what I wrote in reply to Richard Norman.

> So that great strength isn't in fact a point of difference.

That's like saying,

the concept of color is not a point of difference between black and white,
so the great strength of white over black isn't in fact a point of
difference.

That great strength, by the way, comes from the way white can
be resolved into a spectrum of colors.

> > Each rank comes with a rough measure of disparity -- the amount
> > of variation allowed between members of the same taxon. Two living members
> > of the family *Hominidae*, for instance, are less disparate from each
> > other than the two most disparate members of the old Linnean class
> > *Amphibia* were from each other -- or even than the two most disparate
> > living amphibians are from each other.
>
> Now this just isn't true. There is no rough measure of disparity worth
> considering.

Yes, there is. Romer himself used some rough and ready rules of thumb
when he wrote,

[Birds] are divided into many orders; but the differences, for
example, between a humming bird and an albatross are much less
than the differences between a seal and a cat, or between a
stegosaurus and a duck-billed dinosaur, forms which are commonly
placed in a single order. The different [bird] orders have,
in general, no more differences between them than exist between
families in other classes of vertebrates, and, anatomically,
generic differences are so slight that fossils are hard to place.
--_Vertebrate Paleontology_, 1945 ed., pp. 264-5

> > These rankings are never better than ballpark estimates of disparity,
> > but they are reasonably good for gauging e.g. how close we are to the
> > true ancestral species of a clade.
>
> It isn't even clear what that would mean.

Too bad Romer isn't alive so you can take up the issue with him. But
Carroll is still alive, and if you wish, I can run this comment by
Romer by him and see whether he agrees with it.

But hey, you specialize in neornithine birds like the hummingbird
and the albatross; perhaps you can actually *reason* against what
Romer wrote instead of falling back on your polemical tricks. You
know, the tricks you used when I cited John Hawks, hominin paleontology
specialist, in support of the statement that we should not,
under present day circumstances, say "humans are apes" in everyday
conversation.

> > Cladistic ("phylogenetic") classification could also give us this
> > same nested hierarchy where LIVING creatures are concerned, but
> > it gets more and more sketchy the further back one goes into
> > the eons. As I told Richard Norman on the "...monkeys?" thread,
> >
> > The nested hierarchy for Tiktaalik stops with the clade determined
> > by it and yourself and doesn't resume until you get all the way
> > down to Tiktaalik itself. [Cladists banished its family,
> > order and even class to oblivion. Not to mention subfamily, superfamily,
> > infraorder, suborder, etc. [By the way, I doubt that there
> > is an appropriate "rank" for its BIG clade to be-- "superclass"
> > probably deserves to go down only as far as the LCA of Amniota,
> > or at best the LCA of Tetrapoda.]
>
> I also have no idea what that means. Stops? Resume?

Try to think for yourself instead of throwing in the towel so soon.

> > Now, a Linnean could put Tiktaalik in the same order, and perhaps
> > also the same family, as *Acathostega* and *Ichthyostega*, and
> > also in the same family as the last common ancestor (LCA) of
> > the old Linnean *Amphibia*, just as the family *Archaeopterygidae*
> > was widely considered to be the birthplace of the class *Aves*.
>
> Again you confuse Linnean (ranked) classification with paraphyly. These
> are orthogonal matters.

You are repeating something I started to rebut in another reply, and
I'll continue the rebuttal there.

> > In fairness, I should add that the Linnean system has two big
> > drawbacks. One, it is useless for reconstructing the Tree of Life
> > without lots of supplementary information. The phylogenetic
> > system, on the other hand, is tailor-made for this task, provided
> > the information we use is accurate.
> >
> > Two, it arbitrarily cuts off a taxon from its parent taxon, and
> > it gets worse the farther out one goes. Thus, in the above example,
> > some fish awfully close to Tiktaalik in their anatomy are arbitrarily
> > put into a different paraphyletic class, *Osteichthyes* ("bony fish")
> > from the one into which I've put Tiktaalik.
>
> Correct.
>
> > Here, however, cladistic systematics suffers from the same drawback,
> > only more so. It simply declares that, IF Tiktaalik happens to
> > be our direct ancestor, then it is more distantly related to its
> > immediate ancestor than it is to us. And, in general, every
> > creature that ever lived was more distantly related to its
> > immediate ancestor than it is to its remotest descendants.
>
> You consider that a problem. I don't. We accurately represent cladistic
> relationships, but as you point out, it's impossible to accurately
> represent simple similarity, as we will always arbitrarily divide
> similar species.

Only *some* similar species, and darn few at our present state of knowledge
about fossils.

> Classification should pick the things we can do over
> those we can't.

You're starting to devolve into a broken record routine on this issue.

Peter Nyikos
Professor, Dept. of Mathematics -- standard disclaimer--
University of South Carolina

http://www.math.sc.edu/~nyikos/

[John Harshman]

On 11/18/15 6:57 AM, Peter Nyikos wrote:
> On Wednesday, November 18, 2015 at 1:04:12 AM UTC-5, John Harshman wrote:
>> On 11/17/15 6:42 PM, Peter Nyikos wrote:
>>> On Tuesday, November 17, 2015 at 8:49:13 PM UTC-5, Peter Nyikos wrote:
>>>> The long-running thread, "Did we come from monkeys?" has morphed into
>>>> a comparison between the two systems of classification that have come
>>>> down to us: the traditional ("Linnean") and the cladistic ("phylogenetic").
>>>>
>>>> I thought it a good idea to present, first, a "primer" on what these
>>>> two systems are, and then to get down to the nitty-gritty of their
>>>> largely complementary nature.
>>>
>>> The Linnean system's great strength lies in the way it classifies all
>>> organisms, both extant and extinct, in a nested hierarchy starting
>>> with the species, and then in ever-expanding circles we have its
>>> genus, family, order, class, phylum, kingdom, and domain. [There are
>>> intermediate ranks, like subfamily, superfamily, infraclass, subclass,
>>> etc.] To keep things simple, I will confine myself mainly to animals
>>> and indeed to vertebrates.
>>
>> So far, so good. Except that the Linnean system could be cladistic,
>
> Not without becoming a weak version of the phylogenetic.

Again, the point of the Linnean system is ranks. Ranked taxa could be
clades as easily as not.

>> as I mentioned before, and that nested hierarchy is not a difference between
>> systems.
>
> Its sheer existence is not, but the two take on radically different shapes.
> See below where I quote what I wrote in reply to Richard Norman.

You are confusing, again, ranked groups with attempts to represent
overall similarity.

>> So that great strength isn't in fact a point of difference.
>
> That's like saying,
>
> the concept of color is not a point of difference between black and white,
> so the great strength of white over black isn't in fact a point of
> difference.
>
> That great strength, by the way, comes from the way white can
> be resolved into a spectrum of colors.

I don't see the analogy here.

>>> Each rank comes with a rough measure of disparity -- the amount
>>> of variation allowed between members of the same taxon. Two living members
>>> of the family *Hominidae*, for instance, are less disparate from each
>>> other than the two most disparate members of the old Linnean class
>>> *Amphibia* were from each other -- or even than the two most disparate
>>> living amphibians are from each other.
>>
>> Now this just isn't true. There is no rough measure of disparity worth
>> considering.
>
> Yes, there is. Romer himself used some rough and ready rules of thumb
> when he wrote,
>
> [Birds] are divided into many orders; but the differences, for
> example, between a humming bird and an albatross are much less
> than the differences between a seal and a cat, or between a
> stegosaurus and a duck-billed dinosaur, forms which are commonly
> placed in a single order. The different [bird] orders have,
> in general, no more differences between them than exist between
> families in other classes of vertebrates, and, anatomically,
> generic differences are so slight that fossils are hard to place.
> --_Vertebrate Paleontology_, 1945 ed., pp. 264-5

So Romer is in fact saying that the traditional system is a bad way of
representing disparity. And you use this in your defense?

>>> These rankings are never better than ballpark estimates of disparity,
>>> but they are reasonably good for gauging e.g. how close we are to the
>>> true ancestral species of a clade.
>>
>> It isn't even clear what that would mean.
>
> Too bad Romer isn't alive so you can take up the issue with him. But
> Carroll is still alive, and if you wish, I can run this comment by
> Romer by him and see whether he agrees with it.

Sure. Don't I recall that Carroll converted to cladistic classification?

> But hey, you specialize in neornithine birds like the hummingbird
> and the albatross; perhaps you can actually *reason* against what
> Romer wrote instead of falling back on your polemical tricks. You
> know, the tricks you used when I cited John Hawks, hominin paleontology
> specialist, in support of the statement that we should not,
> under present day circumstances, say "humans are apes" in everyday
> conversation.

No, I don't recall any tricks. You should stop with the accusations.

As for Romer, I will agree that my subjective impression is that the
range of disparity in mammals is greater than that within birds, though
this is mostly due to cetaceans and bats, not the averages of orders.
And accent on "subjective".

>>> Cladistic ("phylogenetic") classification could also give us this
>>> same nested hierarchy where LIVING creatures are concerned, but
>>> it gets more and more sketchy the further back one goes into
>>> the eons. As I told Richard Norman on the "...monkeys?" thread,
>>>
>>> The nested hierarchy for Tiktaalik stops with the clade determined
>>> by it and yourself and doesn't resume until you get all the way
>>> down to Tiktaalik itself. [Cladists banished its family,
>>> order and even class to oblivion. Not to mention subfamily, superfamily,
>>> infraorder, suborder, etc. [By the way, I doubt that there
>>> is an appropriate "rank" for its BIG clade to be-- "superclass"
>>> probably deserves to go down only as far as the LCA of Amniota,
>>> or at best the LCA of Tetrapoda.]
>>
>> I also have no idea what that means. Stops? Resume?
>
> Try to think for yourself instead of throwing in the towel so soon.

Try to explain more clearly.

Not the case. Where, for example, would you divide birds from dinosaurs
without leaving similar species on opposite sides?

>> Classification should pick the things we can do over
>> those we can't.
>
> You're starting to devolve into a broken record routine on this issue.

That's certainly an easier thing to say than actually addressing my point.

[Peter Nyikos]

You had a chance to *reason* for these allegations. Now you will
get what I originally intended to post, but cut out of my final
draft. Here is your tit-for-tat:

> I was talking about "even that degree of 'certainty'". How could you
> have missed that point?

I didn't, as my "no more than" should have made it clear to
anyone who isn't biased against me, as you have been since
the first day I returned to talk.origins in December 2010.

> Calling a subfamily ancestral provides the
> illusion of having nailed down an ancestor.

"nailed down an ancestor" creates the impression of talking
about an ancestral *species*. If you did NOT mean to create
that impression, then you are ignoring the clear meaning
of "even that degree of certainty."

And don't try to wiggle out of this by continuing to play
games with that word "certainty." You made some pretty absolutist
claims about the placement of gavials in Crocodylia
in sci.bio.paleontology, and I even asked you "What ever happened
to the concept of falsifiablity?"

Yes, you made some claims to the effect that it has been "falsified"
[by which you obviously meant "passed tests for falsification"]
enough times to be like what federal officials call "settled law"
but I don't recall any references.

Remainder deleted, to be replied to later. Duty calls.

Peter Nyikos

[John Harshman]

"I tried to run a taught ship but they fought me at every turn."

>> Calling a subfamily ancestral provides the
>> illusion of having nailed down an ancestor.
>
> "nailed down an ancestor" creates the impression of talking
> about an ancestral *species*. If you did NOT mean to create
> that impression, then you are ignoring the clear meaning
> of "even that degree of certainty."

I'm trying to say that it creates a false impression of what you know by
giving a name to ambiguity.

> And don't try to wiggle out of this by continuing to play
> games with that word "certainty." You made some pretty absolutist
> claims about the placement of gavials in Crocodylia
> in sci.bio.paleontology, and I even asked you "What ever happened
> to the concept of falsifiablity?"

> Yes, you made some claims to the effect that it has been "falsified"
> [by which you obviously meant "passed tests for falsification"]
> enough times to be like what federal officials call "settled law"
> but I don't recall any references.

I gave you at least one reference. Not sure what you're trying to say
here. What does the placement of gharials have to do with what we're
talking about here? And we do know, with as close as science gets to
certainty, what the living sister group of Gavialis is. Some fossils are
in doubt, which is incidentally another good argument for crown groups.

[Peter Nyikos]

On Wednesday, November 18, 2015 at 1:19:11 AM UTC-5, John Harshman wrote:
> On 11/17/15 7:36 PM, Peter Nyikos wrote:
> > On Monday, November 16, 2015 at 11:54:16 AM UTC-5, John Harshman wrote:
> > https://groups.google.com/forum/#!original/talk.origins/rehTL2kf_jU/2wbrz-5WBAAJ
> >> On 11/13/15 5:02 PM, Peter Nyikos wrote:
> >
> >>> You have
> >>> simply got it into your head that letting people know about
> >>> the disparity in groups -- be they paraphyletic or clades --
> >>> is something a classification system SHOULD NOT be concerned with.
> >>
> >> Correct.
> >>
> >>> But I don't think even a genius with an IQ of 180 and all the knowledge
> >>> Romer or Carroll could come up with a method anywhere near as good
> >>> as the traditional system for conveying comprehensive information.
> >>> While comparison of disparities in different families must always
> >>> be imprecise, nesting goes a long way towards getting rough and ready
> >>> estimates. For instance...
> >>
> >> That's certainly your opinion.
> >
> > You are a polemicist first, a propagandist second, and a reasoner
> > a distant third, so of course you didn't try to come up with
> > any alternative method of organizing the information we have
> > on the disparity of organisms in a manner useful to people.
>
> And you like your accusations.

So do you, and you've been a lot more nasty with them as well
as starting a lot earlier. There is a dandy illustration of both
of those aspects from just ONE week after I resumed posting
to talk.origins after almost a decade of absence.

It was something you said to "el cid," and it only turned out
a long time later that it was a flagrant case of wishful thinking
on your part, with absolutely no basis in reality as far as anything
you had seen at the time. Keywords: protein ribosome.

> > Instead, you reached into your polemical bag labeled "opinion"
> > and pulled out the best sentence that fit.
>
> If you want information on disparity, some kind of morphospace plot
> would seem to be the most useful, or perhaps an index of disparity
> abstracted from the plot.

Assuming such a thing is even possible. Non-mathematicians have very
naive ideas on what can and what cannot be quantified. Usually one
can only hope for comparisons of "more disparate than" as in Romer's
examples that I gave you in another reply earlier today.

> As I have mentioned, trying to capture both
> phylogeny and disparity at once fails in both missions.

And as I have said, there is no attempt of the Linnean system to
capture phylogeny. On the other hand, it remains faithful to
phylogeny with its very careful use of taxa that may either
be clades or paraphyletic (the latter in the very precise sense
defined in my first post).

However, I sort of succeeded where you didn't even dare try. I worked out
one system in the late 1990's in sci.bio.paleontology that does do a
much better job of capturing disparity than the Linnean system AND of
getting much closer to recapturing phylogenetic trees. The only reason
I said what I did earlier about "an IQ of 180" is that it is much harder to
teach than the Linnean system was. Hence also my "in a manner useful
to people."

The main complicating feature is that it allows orders to overlap with
ancestral orders, families to overlap with ancestral families, and so
forth. [The amount of overlap would have been at least one rank lower;
for instance, in my system, the class *Aviremigia* could overlap the
class *Archosauria* in the order *Maniraptora* but not in the bigger clade Coelurosauria.]

Thus I avoid the the abrupt cutoffs of the Linnean system and allow
each species to be its own center of the "taxonomic universe."
I'll gladly explain this further if you wish. Just don't expect
me to make posts like this 300+ lines long by explaining everything
on the spot in terms even a middle school kid could understand if
it didn't exceed his attention span.

In fact, I should probably stop here before I exceed YOUR attention
span. But I'll finish dealing with this post of yours before this
evening is over.

Peter Nyikos
Professor, Dept. of Math -- standard disclaimer--
University of S. Carolina

[Mark Isaak]

On 11/17/15 7:36 PM, Peter Nyikos wrote:

> [snip less problematic parts]

>
> Paraphyly is an indispensible concomitant to any attempt to give
> any ranking of animals as far back as Tiktaalik.

No, that is not true. Paraphyly is a property of tree topology only.
(You do know something about topology, do you not?) Ranking is, more or
less, a matter of applying scaling to the tree. The scaling can be
independent of the shape.

--
Mark Isaak eciton (at) curioustaxonomy (dot) net
"Keep the company of those who seek the truth; run from those who have
found it." - Vaclav Havel

[John Harshman]

You also like your obsessions with wrongs done to you many years ago,
which you must be turning over and over in your memory until the sum
total of them fills your head. Let it go.

>>> Instead, you reached into your polemical bag labeled "opinion"
>>> and pulled out the best sentence that fit.
>>
>> If you want information on disparity, some kind of morphospace plot
>> would seem to be the most useful, or perhaps an index of disparity
>> abstracted from the plot.
>
> Assuming such a thing is even possible. Non-mathematicians have very
> naive ideas on what can and what cannot be quantified. Usually one
> can only hope for comparisons of "more disparate than" as in Romer's
> examples that I gave you in another reply earlier today.

There are such things as morphospace plots, you know. They have the
disadvantages of any such measures, that they can change quite a bit
based on chosen data. Your condescension is unpleasant.

>> As I have mentioned, trying to capture both
>> phylogeny and disparity at once fails in both missions.
>
> And as I have said, there is no attempt of the Linnean system to
> capture phylogeny. On the other hand, it remains faithful to
> phylogeny with its very careful use of taxa that may either
> be clades or paraphyletic (the latter in the very precise sense
> defined in my first post).

Again, you confuse ranks (the Linnean system) with paraphyly (attempts
to capture disparity). Constantly. Your distinction between capturing
phylogeny and remaining faithful to phylogeny is not useful.

> However, I sort of succeeded where you didn't even dare try.

....by your own judgment.

> I worked out
> one system in the late 1990's in sci.bio.paleontology that does do a
> much better job of capturing disparity than the Linnean system AND of
> getting much closer to recapturing phylogenetic trees. The only reason
> I said what I did earlier about "an IQ of 180" is that it is much harder to
> teach than the Linnean system was. Hence also my "in a manner useful
> to people."

I deny that your system did a good job of anything. We disagree.

> The main complicating feature is that it allows orders to overlap with
> ancestral orders, families to overlap with ancestral families, and so
> forth. [The amount of overlap would have been at least one rank lower;
> for instance, in my system, the class *Aviremigia* could overlap the
> class *Archosauria* in the order *Maniraptora* but not in the bigger clade Coelurosauria.]

Yes. I consider this a hideous monster. Less said the better.

> Thus I avoid the the abrupt cutoffs of the Linnean system and allow
> each species to be its own center of the "taxonomic universe."
> I'll gladly explain this further if you wish. Just don't expect
> me to make posts like this 300+ lines long by explaining everything
> on the spot in terms even a middle school kid could understand if
> it didn't exceed his attention span.

No need. I have some memory of your system, which I seem to recall I
hated at the time just as much as I do now. It combines the worst of
multiple worlds, while the traditional classification combines only the
worst of two worlds.

[Peter Nyikos]

On Tuesday, November 17, 2015 at 8:59:14 PM UTC-5, John Harshman wrote:
> On 11/17/15 5:46 PM, Peter Nyikos wrote:
> > The long-running thread, "Did we come from monkeys?" has morphed into
> > a comparison between the two systems of classification that have come
> > down to us: the traditional ("Linnean") and the cladistic ("phylogenetic").
>
> No, you have once again conflated two questions: the value of ranked
> groups and the value of clades vs. paraphyetic groups. "Linnean" refers
> to the former question, while below you discuss only the latter.

The Linnean system in its modern form (including the classification
of Vertebrata used by Romer) was rife with paraphyletic groups.
Surely you cannot deny this. Just look at all the orders to which the
old *Condylartha* gave rise before it was decimated to make the thing
now bearing the same name into a clade.

I gave you a very detailed hypothesis by Carroll according to which
a baker's dozen orders of mammals descended from various families and
subfamilies of Condylartha. And below, I gave two other examples
that were once known to almost all high school biology students.

Thanks for adding more details to what I wrote. You told Richard Norman
quite a lot about those wars in the "...monkeys?" thread, and I was
just telling the relevant part of the outcome.

> There were
> three sides: cladists, traditional "evolutionary" systematists, and
> pheneticists. The latter group denied that we should try to recover
> evolutionary relationships at all and just classify based on overall
> similarity.

How "far out" were they? Did they group *Metamynodon* with hippos, or
worse, golden moles with marsupial moles, or worse yet, ichthyosaurs
with dolphins or snakes with caecilians?

> > Which is a shame, because the two systems are by nature
> > complementary and each makes up for certain disadvantages of
> > the other. My next post to this thread will review
> > some of what has already transpired on this theme.
>
> I hope you will be a bit more accurate next time.

The word "accurate" should be replaced with "comprehensive."
That is, if you are referring to the Cladist Wars.

Peter Nyikos
Professor, Dept. of Math -- standard disclaimer--
U. of S. Carolina

[John Harshman]

On 11/18/15 5:09 PM, Peter Nyikos wrote:
> On Tuesday, November 17, 2015 at 8:59:14 PM UTC-5, John Harshman wrote:
>> On 11/17/15 5:46 PM, Peter Nyikos wrote:
>>> The long-running thread, "Did we come from monkeys?" has morphed into
>>> a comparison between the two systems of classification that have come
>>> down to us: the traditional ("Linnean") and the cladistic ("phylogenetic").
>>
>> No, you have once again conflated two questions: the value of ranked
>> groups and the value of clades vs. paraphyetic groups. "Linnean" refers
>> to the former question, while below you discuss only the latter.
>
> The Linnean system in its modern form (including the classification
> of Vertebrata used by Romer) was rife with paraphyletic groups.
> Surely you cannot deny this. Just look at all the orders to which the
> old *Condylartha* gave rise before it was decimated to make the thing
> now bearing the same name into a clade.

What you have there is not the Linnean system but a set of particular
classifications using the Linnean system; again, the Linnean system
itself is just a set of nested ranks.

> I gave you a very detailed hypothesis by Carroll according to which
> a baker's dozen orders of mammals descended from various families and
> subfamilies of Condylartha. And below, I gave two other examples
> that were once known to almost all high school biology students.

What you have given me is a set of paraphyletic groups. As I said, you
have conflated ranks with paraphyly. The former is the Linnean system,
the latter is a feature of some classifications within the Linnean system.

I don't recall any phenetic classifications of any of those groups. But
similar things may have happened, depending on the data sets chosen.

>>> Which is a shame, because the two systems are by nature
>>> complementary and each makes up for certain disadvantages of
>>> the other. My next post to this thread will review
>>> some of what has already transpired on this theme.
>>
>> I hope you will be a bit more accurate next time.
>
> The word "accurate" should be replaced with "comprehensive."
> That is, if you are referring to the Cladist Wars.

I am referring to the entire post, but mostly to your continuing
conflation of Linnean system and paraphyly.

[erik simpson]

I have a well-intentioned suggestion: take this discussion to SBP. You have
a tendency in this group to say things that sound to others condescending and/or
insulting. You also often bring up unpleasant references to past wrangles.
Your better behavior in SBP would result in less perceived hostility from
others.

[Peter Nyikos]

On Monday, November 16, 2015 at 11:54:16 AM UTC-5, John Harshman wrote:
> On 11/13/15 5:02 PM, Peter Nyikos wrote:

> > On Thursday, November 12, 2015 at 11:44:29 AM UTC-5, John Harshman wrote:
> >> On 11/12/15 7:26 AM, Peter Nyikos wrote:

This is my third and final reply to a post on the "...monkeys?" thread.
Repeating a little from my first reply for continuity:

> > Cladistic classification does a perfect job of letting you get the
> > topology of the tree [1], but that is about all that it is good for.
> > It gives you no idea of how disparate [2] the organisms in a given
> > clade are. A huge number of branches, as in that (sub?)order of
> > Cambrian trilobites I mentioned another day, measures diversity
> > [i.e, sheer number of species] and not disparity. That order
> > is probably a lot less disparate than the class of placental mammals.

> > [1] This is only of use if you are one of those unfortunate people
> > who either can't find the tree online or are reading a paper
> > that doesn't provide them.
>
> Beg to differ. It's just a matter of having convenient names by which to
> refer to clades. And a classification also summarizes a tree in
> frequently handy form.

Not any old classification, just the phylogenetic [cladistic]. Are you
trying to perform an Orwellian feat of defining the Linnean classification
out of existence?

But otherwise, yes, phylogenetic classification gives convenient names
just like any reasonable classification would.

> > [2] I almost wrote "diverse" instead of "disparate" because that is the
> > layman's term for how different and contrasting things are from each other,
> > while the layman's "disparate" has connotations that the official biological
> > term does not.
> >
> >> Once again you have merely displayed a set of claims without any attempt
> >> to show why those claims are better science than a cladistic attempt at
> >> similar questions.
> >
> > I never claimed *better* science, only *supplementary* science, which
> > is why I keep opting for parallel systems, sort of like Dewey Decimal
> > and Library of Congress but, as you can see, serving far more
> > disparate purposes than is the case with those two systems.
>
> Parallel systems would be cumbersome and confusing, especially if you
> re-used names between systems, which it seems you may intend.

Finally, you get a glimmer of why I dislike the commandeering of names
like *Condylartha*, *Therapsida*, *Osteichthyes*, etc. to refer to other
taxa than the Linnean one did.

But you never were one for being able to put yourself in another
person's shoes, were you?

> >> It may be completely obvious to you and thus need no
> >> explanation, but you must understand that it's not obvious to other
> >> people.
> >
> > There is no "it's" of which you think you are speaking.
>
> Then we have no disagreement.

You are being smart alecky. You thought there was something I
"must understand" and now you are recanting without ever telling
what I was supposed to be understanding.

What I'm afraid of is that Erik Simpson and Richard Norman may have
also thought there was something I "must understand" and this may
be why Erik was so hostile in his last two replies to me, while
Richard said "Please please please" twice to me when asking me to
explain something I never claimed in the first place.

> >> You must justify your preference explicitly.
> >
> > My preference is for two systems, but you've fallen back into your
> > old habit of acting as though I only want one.
>
> Is this not the first time (at least in this thread) that you have
> mentioned this "parallel systems" idea? I thought you were defending the
> classical system, which is not parallel but mixed.

What's mixed about it?

Anyway "(at least in this thread)" is a cop-out. You know damn well
that I've mentioned the analogy with Dewey Decimal and Library of Congress
MANY times, always in connection with my wish to see two parallel
systems. You've confused quite a number of people with your innuendo
to the contrary, including Mark Isaak.

> You should provide a few details of what you mean by "parallel systems"
> if that's what you are advocating.

I don't think there are any grade school or middle school children
following this thread, so I'll ignore your "should provide."

Peter Nyikos
Professor, Dept. of Math -- standard disclaimer--
University of S. Carolina at Columbia

[John Harshman]

On 11/18/15 5:49 PM, Peter Nyikos wrote:
> On Monday, November 16, 2015 at 11:54:16 AM UTC-5, John Harshman wrote:
>> On 11/13/15 5:02 PM, Peter Nyikos wrote:
>>> On Thursday, November 12, 2015 at 11:44:29 AM UTC-5, John Harshman wrote:
>>>> On 11/12/15 7:26 AM, Peter Nyikos wrote:
>
> This is my third and final reply to a post on the "...monkeys?" thread.
> Repeating a little from my first reply for continuity:
>
>>> Cladistic classification does a perfect job of letting you get the
>>> topology of the tree [1], but that is about all that it is good for.
>>> It gives you no idea of how disparate [2] the organisms in a given
>>> clade are. A huge number of branches, as in that (sub?)order of
>>> Cambrian trilobites I mentioned another day, measures diversity
>>> [i.e, sheer number of species] and not disparity. That order
>>> is probably a lot less disparate than the class of placental mammals.
>
>>> [1] This is only of use if you are one of those unfortunate people
>>> who either can't find the tree online or are reading a paper
>>> that doesn't provide them.
>>
>> Beg to differ. It's just a matter of having convenient names by which to
>> refer to clades. And a classification also summarizes a tree in
>> frequently handy form.
>
> Not any old classification, just the phylogenetic [cladistic]. Are you
> trying to perform an Orwellian feat of defining the Linnean classification
> out of existence?

As is your habit, you conflate Linnean classifications with paraphyly.
Yes, only a classification limited to monophyly can summarize a tree.

> But otherwise, yes, phylogenetic classification gives convenient names
> just like any reasonable classification would.
>
>>> [2] I almost wrote "diverse" instead of "disparate" because that is the
>>> layman's term for how different and contrasting things are from each other,
>>> while the layman's "disparate" has connotations that the official biological
>>> term does not.
>>>
>>>> Once again you have merely displayed a set of claims without any attempt
>>>> to show why those claims are better science than a cladistic attempt at
>>>> similar questions.
>>>
>>> I never claimed *better* science, only *supplementary* science, which
>>> is why I keep opting for parallel systems, sort of like Dewey Decimal
>>> and Library of Congress but, as you can see, serving far more
>>> disparate purposes than is the case with those two systems.
>>
>> Parallel systems would be cumbersome and confusing, especially if you
>> re-used names between systems, which it seems you may intend.
>
> Finally, you get a glimmer of why I dislike the commandeering of names
> like *Condylartha*, *Therapsida*, *Osteichthyes*, etc. to refer to other
> taxa than the Linnean one did.

I think I know why, but it isn't contained in what I said there. And
once again you confuse "Linnean" with paraphyly. Please stop.

>>>> It may be completely obvious to you and thus need no
>>>> explanation, but you must understand that it's not obvious to other
>>>> people.
>>>
>>> There is no "it's" of which you think you are speaking.
>>
>> Then we have no disagreement.
>
> You are being smart alecky. You thought there was something I
> "must understand" and now you are recanting without ever telling
> what I was supposed to be understanding.

I believe I did tell you. What you were supposed to understand is that
your point wasn't clear, and what was obvious to you was what your point
may have been. If you deny having a point, which seems unlikely, then
the lack of "it" would make sense. But not otherwise.

> What I'm afraid of is that Erik Simpson and Richard Norman may have
> also thought there was something I "must understand" and this may
> be why Erik was so hostile in his last two replies to me, while
> Richard said "Please please please" twice to me when asking me to
> explain something I never claimed in the first place.

They are both probably experiencing frustration at your failure to make
your points (which I still assume you have) clearly and explicitly.

>>>> You must justify your preference explicitly.
>>>
>>> My preference is for two systems, but you've fallen back into your
>>> old habit of acting as though I only want one.
>>
>> Is this not the first time (at least in this thread) that you have
>> mentioned this "parallel systems" idea? I thought you were defending the
>> classical system, which is not parallel but mixed.
>
> What's mixed about it?

It's mixed in that it allows for both monophyly and paraphyly and tries
to compromise between representing a tree and representing similarity.

> Anyway "(at least in this thread)" is a cop-out. You know damn well
> that I've mentioned the analogy with Dewey Decimal and Library of Congress
> MANY times, always in connection with my wish to see two parallel
> systems. You've confused quite a number of people with your innuendo
> to the contrary, including Mark Isaak.

Yes, I clearly must be cause of everyone's confusion.

>> You should provide a few details of what you mean by "parallel systems"
>> if that's what you are advocating.
>
> I don't think there are any grade school or middle school children
> following this thread, so I'll ignore your "should provide."

I don't know what that was intended to mean, though I suppose it was
some kind of insult. If you want anyone to know what you're talking
about, you need to explain. I remember the last time you brought this
up, but I don't think anyone else does.

[Peter Nyikos]

On Wednesday, November 18, 2015 at 1:19:11 AM UTC-5, John Harshman wrote:
> On 11/17/15 7:36 PM, Peter Nyikos wrote:
> > On Monday, November 16, 2015 at 11:54:16 AM UTC-5, John Harshman wrote:
> > https://groups.google.com/forum/#!original/talk.origins/rehTL2kf_jU/2wbrz-5WBAAJ
> >> On 11/13/15 5:02 PM, Peter Nyikos wrote:

This is my second and final reply to this post of Harshman's, and
picks up where the other left off.

> >>> *Equinae* is less disparate than *Equidae* which in turn is less disparate
> >>> than *Equioidea* which in turn is less disparate than *Perissodactyla*
> >>> which in turn is MORE disparate than *Tapiroidea which in turn is
> >>> MORE disparate than *Tapiridae*, etc. etc. And someone familiar with the
> >>> whole order might even venture to compare the disparity in *Equidae* with
> >>> that in *Tapiridae* enough to be able to say that neither is as low in
> >>> disparity as a subfamily of the other.
> >>
> >> All you did above, until the last comparison, is note that groups are
> >> nested within groups.
> >
> > All you are doing here is using a literally false sentence for effect;
> > and next, a sentence that refers to the false picture you painted:
>
> More accusations. You should stop that.

Stating facts is not something you can get me to stop. Not even the
lovers of censorship over in Amazon.com could get me to do that,
and unlike one of your fellow ornithologists over there, you cannot
summon people to sink my words out of plain sight -- talk.origins
doesn't work like that.

> That sentence is true. By
> definition a group Y that includes X must be more disparate than X, and
> that's all you show.

How simple the world would be if that were true! But I have an interesting
counterexample: the disparity between the two most different members
of the species *Ursus arctos* ("brown bears") is far greater than the
disparity between that paraphyletic species and *Ursus maritimus*, the
polar bear, which is descended from it.


https://en.wikipedia.org/wiki/Polar_bear

Adding in the polar bear, as many now advocate, would do nothing to
increase the overall disparity of the group, only its diversity.

[By the way, I remember an old book on genetics according to which
the polar bear was classed back then as a separate *genus* from
the brown bear. But that has no bearing on the reasoning used above.]

> >> Cladistic classification does that too.

It does nothing in the direction of disparity, and you even think
of that as a virtue.

> >> Your final
> >> comparison, of non-nested groups at the same rank, is I think
> >> fallacious; such taxa have no clear comparability.
> >
> > So much for alleged measures of disparity, eh?
>
> What measures of disparity? You have mentioned none.

You claimed a year or so ago that there were some in general use.
Do you wish to retract that statement?

> >> Further, you are once
> >> again confusing two issues: ranking and paraphyly.
> >
> > "confusing" is false, and "once again" also, unless you can point to
> > a *valid* prior example.
>
> You have done it many times in this thread and the one in
> sci.bio.paleontology. I have tried to point it out each time.

Well, if you ever did before these two threads, it must have
been in a context where I decided it wasn't worth making an
issue of it. Unlike you, I have had a full time job for the last
four decades, and plan to continue with mine for at least another

> > Paraphyly is an indispensible concomitant to any attempt to give
> > any ranking of animals as far back as Tiktaalik.
>
> No, it is not. But was that your point at last? What you might mean is
> that it's indispensible if you want to make all groups of the same rank
> equally disparate.

You are SO stuck in an "extant species" frame of mind! Don't you realize how
you cannot have e.g. all Eocene mammals being in genera that are all
clades, subfamilies that are all clades, families that are all clades,
suborders that are all clades, orders that are all clades, etc. and
still have a nested hierarchy that has any basis in reality???

Or just try doing that for Equidae, starting out from whatever
tree you like and including all the genera that Kathleen Hunt use
in her tree:

http://www.talkorigins.org/faqs/horses/horse_evol.html

I'd love to see the surrealistic example you would come up with
if you tried.

Another example: re-read what I wrote about Tiktaalik in my
second post until you see how naturally paraphyletic taxa
come into play.

The nested hierarchy for Tiktaalik stops with the clade determined
by it and yourself and doesn't resume until you get all the way
down to Tiktaalik itself. [Cladists] banished its family,
order and even class to oblivion. Not to mention subfamily, superfamily,
infraorder, suborder, etc. [By the way, I doubt that there
is an appropriate "rank" for its BIG clade to be-- "superclass"
probably deserves to go down only as far as the LCA of Amniota,
or at best the LCA of Tetrapoda.]

Now, a Linnean could put Tiktaalik in the same order, and perhaps
also the same family, as *Acathostega* and *Ichthyostega*, and
also in the same family as the last common ancestor (LCA) of
the old Linnean *Amphibia*, just as the family *Archaeopterygidae*
was widely considered to be the birthplace of the class *Aves*.

And all of these intermedidate taxa are paraphyletic, not to
mention the subclass of *Amphibia* as well as the class itself.
We cannot get back to a clade containing Tiktaalik until we get
to the clade of all terrestrial vertebrates.

<snip second half of GIGO>

> > I've started
> > to give people a feel for that in my second post to this thread,
> > but I expect you and at least one other active participant
> > on the old thread to find it very unclear, and to request that
> > I stop beating around the bush and to state my point.
>
> Which you didn't actually do, did you?

I did it this time around, but I expect two or three of you
to be enormously creative in finding things you don't
understand about it.

Peter Nyikos

Professor, Dept. of Mathematics -- standard disclaimer--
University of South Carolina

http://www.math.sc.edu/~nyikos/
nyikos @ math.sc.edu

[RSNorman]

My "please, please, please" was an attempt, still unanswered, to get
you to explain simply and clearly just what bothers you about
cladistics.

That "node-base vs. stem-based tree" paper cited says clearly in the
first sentence of the introduction:

“Treethinking”, using phylogenies to understand evolutionary
relationships, name clades, and understand evolutionary
transformations and biogeography, is now ubiquitous in systematics and
evolutionary biology and is making its way quickly into the
educational and public realms.

Those phylogenies and names are monophyletic. Frankly everyone always
knew that those earliest tetrapods who hauled themselves out of the
water were not really "Amphibia" nor were the jawed cartilaginous
aquatic vertebrates that gave rise to bony fish were not really
"Sharks". Everyone always knew that the diversity of one phylum of
animals might be more like that of some order of a different phylum
and that there was no rules whatsoever for deciding which branching
point was best called an "order" and which a "family". The crustacea
went through some contortions trying to name subgroups and give them a
rank. Everyone always knew that the "reptiles' included a rather
motley group mainly being tetrapods that lacked endothermal
homeothermy and completely separated pulmonary and systemic
circulation but had no evolutionary unity.

Did you complain when the rotifers were separated from the nematodes,
formerly one phylum in Linnean classification? Did you complain when
rabbits were moved out of the rodents in Linnean classification?
Things have always changed and you simply had to keep up.

My impression, which is probably completely wrong but you have not
done anything to clarify it, is that you simply object to the words
used to name clades. There is nothing "non-Linnean" about requiring
that all Linnean ranks be clades and there is nothing "non-Linnean"
about giving names to occasion subtrees (monophyletic, of course) in
the entire tree of some major organismal grouping. It seems that you
only object to the fact that some of your favorite names have either
been simply redefined or gone defunct, only to be replaced by other
names to reflect phylogeny.

The purpose of classification is two-fold: to name and identy and to
classify according to some important schema or foundation. A
dichotomous key does the first and every intro student in classes that
I have taught learns about those and how they are rather arbitrary in
the choice of keys. The second is phylogeny because the schema is, in
biology, the evolutionary tree of life, the directed tree of ancestry.
The naming system should holdfully reflect the important
relationships, the phylogeny, and not appearances, the phenetics.

[John Harshman]

Classic, right down to the pointless dig at a third party.

>> That sentence is true. By
>> definition a group Y that includes X must be more disparate than X, and
>> that's all you show.
>
> How simple the world would be if that were true! But I have an interesting
> counterexample: the disparity between the two most different members
> of the species *Ursus arctos* ("brown bears") is far greater than the
> disparity between that paraphyletic species and *Ursus maritimus*, the
> polar bear, which is descended from it.

That works only because you explicitly remove group X from group Y, so
that group Y doesn't include group X.

> https://en.wikipedia.org/wiki/Polar_bear
>
> Adding in the polar bear, as many now advocate, would do nothing to
> increase the overall disparity of the group, only its diversity.

I don't think that's true, but it's such a stupid argument that I feel
disinclined to continue.

> [By the way, I remember an old book on genetics according to which
> the polar bear was classed back then as a separate *genus* from
> the brown bear. But that has no bearing on the reasoning used above.]

Correct. No bearing. Ha.

>>>> Cladistic classification does that too.
>
> It does nothing in the direction of disparity, and you even think
> of that as a virtue.

Agreed. It does exactly as much or as little as your example did. And
yes, it's a virtue.

>>>> Your final
>>>> comparison, of non-nested groups at the same rank, is I think
>>>> fallacious; such taxa have no clear comparability.
>>>
>>> So much for alleged measures of disparity, eh?
>>
>> What measures of disparity? You have mentioned none.
>
> You claimed a year or so ago that there were some in general use.
> Do you wish to retract that statement?

No.

>>>> Further, you are once
>>>> again confusing two issues: ranking and paraphyly.
>>>
>>> "confusing" is false, and "once again" also, unless you can point to
>>> a *valid* prior example.
>>
>> You have done it many times in this thread and the one in
>> sci.bio.paleontology. I have tried to point it out each time.
>
> Well, if you ever did before these two threads, it must have
> been in a context where I decided it wasn't worth making an
> issue of it. Unlike you, I have had a full time job for the last
> four decades, and plan to continue with mine for at least another

It's these two threads that I was talking about. And you really need to
cool down.

>>> Paraphyly is an indispensible concomitant to any attempt to give
>>> any ranking of animals as far back as Tiktaalik.
>>
>> No, it is not. But was that your point at last? What you might mean is
>> that it's indispensible if you want to make all groups of the same rank
>> equally disparate.
>
> You are SO stuck in an "extant species" frame of mind! Don't you realize how
> you cannot have e.g. all Eocene mammals being in genera that are all
> clades, subfamilies that are all clades, families that are all clades,
> suborders that are all clades, orders that are all clades, etc. and
> still have a nested hierarchy that has any basis in reality???

No, I don't.

> Or just try doing that for Equidae, starting out from whatever
> tree you like and including all the genera that Kathleen Hunt use
> in her tree:
>
> http://www.talkorigins.org/faqs/horses/horse_evol.html
>
> I'd love to see the surrealistic example you would come up with
> if you tried.

Easily done. Just get yourself a cladogram and name the clades. What
lesson are you trying to teach me?

> Another example: re-read what I wrote about Tiktaalik in my
> second post until you see how naturally paraphyletic taxa
> come into play.

Sorry, I don't see anything there that shows me the naturalness of
paraphyletic taxa. You will have to explain.

> The nested hierarchy for Tiktaalik stops with the clade determined
> by it and yourself and doesn't resume until you get all the way
> down to Tiktaalik itself. [Cladists] banished its family,
> order and even class to oblivion. Not to mention subfamily, superfamily,
> infraorder, suborder, etc. [By the way, I doubt that there
> is an appropriate "rank" for its BIG clade to be-- "superclass"
> probably deserves to go down only as far as the LCA of Amniota,
> or at best the LCA of Tetrapoda.]
>
> Now, a Linnean could put Tiktaalik in the same order, and perhaps
> also the same family, as *Acathostega* and *Ichthyostega*, and
> also in the same family as the last common ancestor (LCA) of
> the old Linnean *Amphibia*, just as the family *Archaeopterygidae*
> was widely considered to be the birthplace of the class *Aves*.
>
> And all of these intermedidate taxa are paraphyletic, not to
> mention the subclass of *Amphibia* as well as the class itself.
> We cannot get back to a clade containing Tiktaalik until we get
> to the clade of all terrestrial vertebrates.

Surely this depends on whether there are any relatives of Tiktaalik. You
seem to be assuming a tree in which Tiktaalik is the sister group of all
other tetrapod-like organisms (there may not be a name for that). In
which case, it would be all by itself. But the Linnean system allows for
a monotypic family to be synonymous with a monotypic order, etc. There
are many examples in the typical Linnean classification. But you do
present a good argument for not using ranks.

> <snip second half of GIGO>
>
>>> I've started
>>> to give people a feel for that in my second post to this thread,
>>> but I expect you and at least one other active participant
>>> on the old thread to find it very unclear, and to request that
>>> I stop beating around the bush and to state my point.
>>
>> Which you didn't actually do, did you?
>
> I did it this time around, but I expect two or three of you
> to be enormously creative in finding things you don't
> understand about it.

So you're saying that because nobody understand your point, it must be
everyone else's fault.

[Öö Tiib]

It may be all of well-intentioned, reasonable and helpful suggestion,
only problem is that it is incomprehensible.

What is SBP? It is seemingly not Spontaneous Bacterial Peritonitis,
Streptavidin Binding Peptide, State Bank of Pakistan, Serial Bus
Protocol, Scannerless Boolean Parser and so on.

Human languages are too ambiguous without those annoying 3-letter
abbreviations.

[jillery]

On Thu, 19 Nov 2015 05:31:59 -0800 (PST), 嘱 Tiib <oot...@hot.ee>
wrote:

It's almost certain that "SBP" refers to another newsgroup,
sci.bio.paleontology. It's been irregularly mentioned in T.O. posts.

In fairness to Erik, his reply was directed to one person, who almost
certainly recognized the abbreviation, and not to the group at large.

>> You have a tendency in this group to say things that sound to
>> others condescending and/or insulting. You also often bring up
>> unpleasant references to past wrangles.
>> Your better behavior in SBP would result in less perceived hostility from
>> others.

[erik simpson]

Apologies. SBP = sci.bio.paleontology is the intended meaning. The post
really was intended for Peter specifically, as he has promised (and so far in
the last several months delivered) to discuss things in a civil manner. In this
thread and some others in this group he's become pretty agitated, and I think
a change in venue might improve the climate.

[Mark Isaak]

On 11/18/15 4:41 PM, Peter Nyikos wrote:

> [...]

> And as I have said, there is no attempt of the Linnean system to
> capture phylogeny. On the other hand, it remains faithful to
> phylogeny with its very careful use of taxa that may either
> be clades or paraphyletic (the latter in the very precise sense
> defined in my first post).

I disagree. Linnaeus was explicitly seeking a natural system, meaning
one which reflected the order of nature. He recognized that the order
was hierarchical, and although he did not know it, that order was a
consequence of common descent. Lack of perfect knowledge prevented it
from being so initially, but the Linnaean system is, in spirit, an
inherently cladistic system.

[Peter Nyikos]

So does everyone else posting to this thread, including you. Just
look at how hostile you were in your last post to the "...monkeys?"
thread:

https://groups.google.com/forum/#!original/talk.origins/rehTL2kf_jU/wuFzJ3TAAwAJ
Message-ID: <596fe1c8-1b39-404e...@googlegroups.com>

Part of your hostility was evidently due to your failure to distinguish
between comments aimed at Harshman and the ONE comment that I let stand
as referring to Richard Norman, albeit in modified form.

It might have also been affected by my not having replied to your
testy comments in your next to last post to that thread:

https://groups.google.com/forum/#!original/talk.origins/rehTL2kf_jU/fFYPPaRJAwAJ
Message-ID: <fda34dc5-eb54-4d5b...@googlegroups.com>

I had written:

Anyway, I hope my second reply, where I reply to one thing
after another that you wrote, will make things more clear to you.

Your testy reply ran:

No, I didm't see much clarity in that, or much in any of this
discussion. It's not worth my effort to struggle through any
more of the endless digressions you can't seem to avoid introducing.

If you could just state, succinctly and without disparaging remarks
about my and others' retardation, just what is compellingly wrong
with 'cladism' that is remedied by whatever you think people did
in 'olden times', I would read it. Otherwise, I'm folding my hand
here. There are better things to do.

Well, I have been posting to this thread about the thing to which
I really object, namely the rejection of a centuries-old approach
to something that phylogenetic systematics (what you call 'cladism')
cannot, by its very nature, address at all.

And so far, you haven't let out a peep about this, or any other
comments by me of a scientific nature, but instead are talking
about personal comments in a reply to a post that Harshman had made
on the other thread.

But I'm willing to meet you halfway. After I see that this has
posted, I will post an abridged version of the post that was
the subject of your testy comments -- a post to which no one has
yet replied -- where I cut out all personal references.

I hope you will be willing, in return, to tell me what, if anything,
you still find unclear about it.

Peter Nyikos

[erik simpson]

On Thursday, November 19, 2015 at 9:24:09 AM UTC-8, Peter Nyikos wrote:
> On Wednesday, November 18, 2015 at 8:44:11 PM UTC-5, erik simpson wrote:

> <...>

Fair enough. But take it over to sci.bio.paleontology, please. If I insulted
you, I apologize, but I will confes to exasperation. It's very hard to see
what point you're trying to make when you introduce digressions or other
distracting 'sub-points'. I still don't see what your larger point is in all
this quarreling about what Romer and Carrol had to say in the 70's or earlier.
Like it or not, the field has moved on, and obsolete 'groups' such as
Condylarthra aren't regarded as having the same evolutionary significance that
was previously thought. (Not that they aren't interesting critters.)

[Peter Nyikos]

On Thursday, November 19, 2015 at 12:44:08 PM UTC-5, erik simpson wrote:
> On Thursday, November 19, 2015 at 9:24:09 AM UTC-8, Peter Nyikos wrote:
> > On Wednesday, November 18, 2015 at 8:44:11 PM UTC-5, erik simpson wrote:
> > <...>
>
> > > I have a well-intentioned suggestion: take this discussion to SBP. You have
> > > a tendency in this group to say things that sound to others condescending and/or
> > > insulting.
> >
> > So does everyone else posting to this thread, including you.

I made this comment before I saw Oo Tiib's post. I cannot recall
anything insulting or condescending written by him.

<snip for focus>

> > But I'm willing to meet you halfway. After I see that this has
> > posted, I will post an abridged version of the post that was
> > the subject of your testy comments -- a post to which no one has
> > yet replied -- where I cut out all personal references.
> >
> > I hope you will be willing, in return, to tell me what, if anything,
> > you still find unclear about it.
> >
> > Peter Nyikos
>
> Fair enough. But take it over to sci.bio.paleontology, please.

I respectfully decline to do so. There are people posting to this
thread whom I have never seen over there. More importantly, some of
what I have to say will, I hope, help clarify some issues that
some people still find confusing.

So, here I go with a rewrite which takes this thread into account.

On Saturday, November 7, 2015 at 1:44:46 AM UTC-5, erik simpson wrote:

> My understanding is that you agree that the cladistic
> analysis works to establish accurate phylogenetic trees, but object that it
> is a 'juggernaut' that crushes the old 'horizontal' classification scheme?

Wrong referent for "it". The juggernaut I was talking about is the
one that crushes any official way of classifying the results of any
phylogenetic analysis besides besides the one that uses clades
and only clades.

> Why not? If modern analysis is an improvement to deducing relationships,
> why bother with outdated procedures?

As I've been trying to make clear on this new thread, it isn't the
modern *analysis* to which I object, nor the phylogenetic trees
that result, but a lack of toleration about how the trees are
split up into taxa.

> Your objections sound querulous to me.

Which objections? Could you quote at least one of the ones you had
in mind back at me?

> Another aspect of your thinking also raises alarms: You speak of
> 'our more immediate ancestors, like the South African "Homo naledi"'

Haste makes waste. I should have said "candidates for our more
immediate ancestors..."

> It's
> my impression that along with phylogenetic analysis, an appreciation of the
> unlikelyhood of ever determining 'ancestors' definitively is one of the better
> developments in paleontology.

Not just in paleontology, but in all of science, thanks to the
methodological emphasis on falsifiability.

But when a reasonably complete fossil passes all tests that could
falsify the hypothesis that the species x to which it belongs is
directly ancestral to the species of Clade Y, then it seems eminently
reasonable to designate it as "the ancestral candidate" [shorthand for
"the best available candidate for an ancestor"].

Lacking such clear candidates, a naturalist with the expertise of
Romer or Carroll could still draw "bubble diagrams" of ancestry using
paraphyletic taxa organized in tree-like fashion.

Kathleen Hunt's diagram of of hypothesized ancestor-descendant
relationships among genera in *Equioidea* [1] is one of the few trees
of the old school that goes down to the genus level.

http://www.talkorigins.org/faqs/horses/horse_evol.html

[1] Kathleen uses "Equidae" but her inclusion of *Palaeotherium*
makes her diagram more conformal to the superfamily *Equioidea*.

Peter Nyikos
Professor, Dept. of Mathematics
U. of S. Carolina, Columbia -- standard disclaimer--
http://people.math.sc.edu/nyikos/

[erik simpson]

This assertion in uncomfortably similar to "Bill"'s claim that the earth must
be unique, since we have no evidence that it isn't. Why do we need "the
ancestral candidate"? A plausible statistical case could be made that
Charlemagne was an ancestor of mine, and he might be, but so what?

> Lacking such clear candidates, a naturalist with the expertise of
> Romer or Carroll could still draw "bubble diagrams" of ancestry using
> paraphyletic taxa organized in tree-like fashion.
>
> Kathleen Hunt's diagram of of hypothesized ancestor-descendant
> relationships among genera in *Equioidea* [1] is one of the few trees
> of the old school that goes down to the genus level.
>
> http://www.talkorigins.org/faqs/horses/horse_evol.html
>
> [1] Kathleen uses "Equidae" but her inclusion of *Palaeotherium*
> makes her diagram more conformal to the superfamily *Equioidea*.
>

Romer and Carroll could certainly draw bubble diagrams, as might Ms. Hunt,
but as has been pointed out (many times), it matters what considerations
were used to draw them. "I heard from Van Valen" at last year's GSA meeting
isn't very falsifiable, or even believable.

[John Harshman]

On 11/19/15 10:18 AM, Peter Nyikos wrote:
> On Thursday, November 19, 2015 at 12:44:08 PM UTC-5, erik simpson wrote:
>> On Thursday, November 19, 2015 at 9:24:09 AM UTC-8, Peter Nyikos wrote:
>>> On Wednesday, November 18, 2015 at 8:44:11 PM UTC-5, erik simpson wrote:
>>> <...>
>>
>>>> I have a well-intentioned suggestion: take this discussion to SBP. You have
>>>> a tendency in this group to say things that sound to others condescending and/or
>>>> insulting.
>>>
>>> So does everyone else posting to this thread, including you.
>
> I made this comment before I saw Oo Tiib's post. I cannot recall
> anything insulting or condescending written by him.
>
> <snip for focus>
>
>>> But I'm willing to meet you halfway. After I see that this has
>>> posted, I will post an abridged version of the post that was
>>> the subject of your testy comments -- a post to which no one has
>>> yet replied -- where I cut out all personal references.
>>>
>>> I hope you will be willing, in return, to tell me what, if anything,
>>> you still find unclear about it.
>>>
>>> Peter Nyikos
>>
>> Fair enough. But take it over to sci.bio.paleontology, please.
>
> I respectfully decline to do so.

Then perhaps you would consider following your sbp posting standards
here, at least in this thread. That's what Erik is really asking for.

> There are people posting to this
> thread whom I have never seen over there. More importantly, some of
> what I have to say will, I hope, help clarify some issues that
> some people still find confusing.

> So, here I go with a rewrite which takes this thread into account.
>
> On Saturday, November 7, 2015 at 1:44:46 AM UTC-5, erik simpson wrote:
>
>> My understanding is that you agree that the cladistic
>> analysis works to establish accurate phylogenetic trees, but object that it
>> is a 'juggernaut' that crushes the old 'horizontal' classification scheme?
>
> Wrong referent for "it". The juggernaut I was talking about is the
> one that crushes any official way of classifying the results of any
> phylogenetic analysis besides besides the one that uses clades
> and only clades.

In other words, Peter is distinguishing between cladistic methods for
inferring phylogeny and cladistic classification. They are different
things, though elsewhere he also complains about cladistic methods. He
complains about the practice of equally weighting characters, for example.

The question here is what the value would be in such exercises, either
declaring candidate ancestors, constructing bubble diagrams, or putting
genera as ancestors in phylogenetic trees. I would maintain that you
learn nothing from them, communicate no useful information, and in the
latter two cases obfuscate the hypotheses you are advancing.

[John Harshman]

I will note that the bubble diagram and the Hunt phylogeny go beyond
Peter's expressed main concern with classification and enter into the
province of phylogenetic trees. So he has more against cladism than its
philosophy of classification. This comes close to objecting to methods,
or at least the closely related way of displaying results.

[Peter Nyikos]

On Thursday, November 19, 2015 at 1:34:08 PM UTC-5, erik simpson wrote:
> On Thursday, November 19, 2015 at 10:19:07 AM UTC-8, Peter Nyikos wrote:
> > On Thursday, November 19, 2015 at 12:44:08 PM UTC-5, erik simpson wrote:

Thank you for being willing to meet me halfway, Erik.

<snip for focus>

> > > It's
> > > my impression that along with phylogenetic analysis, an appreciation of the
> > > unlikelyhood of ever determining 'ancestors' definitively is one of the better
> > > developments in paleontology.
> >
> > Not just in paleontology, but in all of science, thanks to the
> > methodological emphasis on falsifiability.
> >
> > But when a reasonably complete fossil passes all tests that could
> > falsify the hypothesis that the species x to which it belongs is
> > directly ancestral to the species of Clade Y, then it seems eminently
> > reasonable to designate it as "the ancestral candidate" [shorthand for
> > "the best available candidate for an ancestor"].
> >
>
> This assertion in uncomfortably similar to "Bill"'s claim that the earth must
> be unique, since we have no evidence that it isn't.

I have no idea why it would be even remotely similar. There is
no claim as to uniqueness (as in "LCA candidate for Clade Y") which
*would* be pretentious. We aren't stifling discussion on anything --
note the word "available".

In fact, just the opposite is true. A reasonably complete fossil
that has passed all falsifiability tests that take all its
characters into account makes a good starting point for discussion.

One topic: if an even more complete fossil were found, what should
one especially look for to either falsify the ancestral role
of the species, or to strengthen the case for it?

> Why do we need "the
> ancestral candidate"?

1. To make it possible for researchers and students and interested laymen
like myself to form some concept as to what a real ancestor might have
been like. Discussions such as the one above would contribute to that.

2. To help people in the field understand the significance of a find
that would shed more light on whether we have a true ancestral species
on our hands.

3. To generate public interest in special cases. Just look at all the
excitement over the discovery of *Tiktaalik*. Who would have cared
about it if it were not at least close to our own ancestry, and
have filled a morphological gap between panderichthyan fish and
such early amphibians as *Acanthostega* and *Ichthyostega*?

Phylogenetic trees of the sort that we are treated to nowadays
say NOTHING about this aspect of Tiktaalik. For all anyone
unfamiliar with it and *Acanthostega* could tell, they could be just
as far off our ancestry as the platypus and the echidnas,
for all that the tree is telling them.

> A plausible statistical case could be made that
> Charlemagne was an ancestor of mine, and he might be, but so what?

Substitute *Tiktaalik* for Charlemagne, and maybe you'll see my point.

> > Lacking such clear candidates, a naturalist with the expertise of
> > Romer or Carroll could still draw "bubble diagrams" of ancestry using
> > paraphyletic taxa organized in tree-like fashion.
> >
> > Kathleen Hunt's diagram of of hypothesized ancestor-descendant
> > relationships among genera in *Equioidea* [1] is one of the few trees
> > of the old school that goes down to the genus level.
> >
> > http://www.talkorigins.org/faqs/horses/horse_evol.html
> >
> > [1] Kathleen uses "Equidae" but her inclusion of *Palaeotherium*
> > makes her diagram more conformal to the superfamily *Equioidea*.
> >
>
> Romer and Carroll could certainly draw bubble diagrams, as might Ms. Hunt,
> but as has been pointed out (many times), it matters what considerations
> were used to draw them.

Of course it does, but that is a completely different issue. I tried
to make that clear in a post I did to the "...monkeys?" thread this
morning.

By the way, what you write below reads like a merciless lampoon of
the second half of what I posted there. Is that just a coincidence?

> "I heard from Van Valen" at last year's GSA meeting
> isn't very falsifiable, or even believable.

How about "Van Valen was kind enough to hand me a preprint,
which he hasn't published up to now (1988), in which he showed the
detailed analysis that led to the narrowing down you see
in Fig.21.8 (b) of the ancestry of three distinct orders of
ungulates and the order Cetacea to various subfamilies of
Condylartha, and of nine other orders to two of its families."

Why else would Carroll go out on a limb the way he did
in his book?

[RSNorman]

On Thu, 19 Nov 2015 11:40:00 -0800 (PST), Peter Nyikos
<nyi...@bellsouth.net> wrote:

>On Thursday, November 19, 2015 at 1:34:08 PM UTC-5, erik simpson wrote:
>
>> Why do we need "the
>> ancestral candidate"?
>
>1. To make it possible for researchers and students and interested laymen
>like myself to form some concept as to what a real ancestor might have
>been like. Discussions such as the one above would contribute to that.
>
>2. To help people in the field understand the significance of a find
>that would shed more light on whether we have a true ancestral species
>on our hands.
>
>3. To generate public interest in special cases. Just look at all the
>excitement over the discovery of *Tiktaalik*. Who would have cared
>about it if it were not at least close to our own ancestry, and
>have filled a morphological gap between panderichthyan fish and
>such early amphibians as *Acanthostega* and *Ichthyostega*?
>
>Phylogenetic trees of the sort that we are treated to nowadays
>say NOTHING about this aspect of Tiktaalik. For all anyone
>unfamiliar with it and *Acanthostega* could tell, they could be just
>as far off our ancestry as the platypus and the echidnas,
>for all that the tree is telling them.
>

Once again let me admit to being totally ignorant of just what the
problem is.

I found Tiktaalik on Wikipedia and it referred me to the clade
Stegocephilia. I looked that up on Wikipedia
https://en.wikipedia.org/wiki/Stegocephalia
and it showed me a very pretty phylogenetic tree, illustrated even,
that shows just how Tiktaalik, Acanthostega, and Ichthyostega are
related to Tetrapoda.

If you look at Wikipedia on Tetrapoda
https://en.wikipedia.org/wiki/Tetrapod
it has a tree in the section "Stem Tetrapoda" that shows the same
Stegocephalia but extended backwards in time to the separation of
lungfish (Dipnomorpha). It then shows several different hypotheses ab
out the relationships that ended up producing the modern amphibians,
the Lissamphibia, and the Amniota.

It all looks very clear and easy to see to me. I don't see any
problem with students learning and understanding the relationship
between the fossil groups and modern tetrapod groups.

I can see no reason to declare or even suggest that Tiktaalik is THE
ancestor of one group, that Acanthostega is THE ancestor of a subgroup
and the Ichthyostega THE ancestor of a sub-subgroup which includes the
two sub-sub-subgroups Amniota and Lissamphibia.

And for more than a century we have known that it really is not
possible to identify the specific ancestral form. Go back in time to
the small village in eastern Belarus where my mother's family came
from. Dig up a grave and there is no way of telling whether that was
my great-great-...-great grandfather or only a
great-great-great-....-great uncle who was not my ancestor at all.

[John Harshman]

What's needed for this is reconstruction of the ancestral characters,
not identification of an ancestor candidate.

> 2. To help people in the field understand the significance of a find
> that would shed more light on whether we have a true ancestral species
> on our hands.

That falls into the category of press-release hype. It's commonly done,
and it's reprehensible.

> 3. To generate public interest in special cases. Just look at all the
> excitement over the discovery of *Tiktaalik*. Who would have cared
> about it if it were not at least close to our own ancestry, and
> have filled a morphological gap between panderichthyan fish and
> such early amphibians as *Acanthostega* and *Ichthyostega*?

It's the filling of the gap that's interesting, not any sort of ancestor
candidacy. You just don't need this.

I will also note that all your appeals are to popular imagination, not
to what's needed for science. While popular appeal is not a bad thing, I
thought we were talking about science here. No?

> Phylogenetic trees of the sort that we are treated to nowadays
> say NOTHING about this aspect of Tiktaalik. For all anyone
> unfamiliar with it and *Acanthostega* could tell, they could be just
> as far off our ancestry as the platypus and the echidnas,
> for all that the tree is telling them.

I suppose that depends on what you mean by "as far off our ancestry".
What do you mean? Are you referring to the length of the terminal
branch? Why should that be important?

>> A plausible statistical case could be made that
>> Charlemagne was an ancestor of mine, and he might be, but so what?
>
> Substitute *Tiktaalik* for Charlemagne, and maybe you'll see my point.

Different case entirely, and the question "so what" still applies.

>>> Lacking such clear candidates, a naturalist with the expertise of
>>> Romer or Carroll could still draw "bubble diagrams" of ancestry using
>>> paraphyletic taxa organized in tree-like fashion.
>>>
>>> Kathleen Hunt's diagram of of hypothesized ancestor-descendant
>>> relationships among genera in *Equioidea* [1] is one of the few trees
>>> of the old school that goes down to the genus level.
>>>
>>> http://www.talkorigins.org/faqs/horses/horse_evol.html
>>>
>>> [1] Kathleen uses "Equidae" but her inclusion of *Palaeotherium*
>>> makes her diagram more conformal to the superfamily *Equioidea*.
>>
>> Romer and Carroll could certainly draw bubble diagrams, as might Ms. Hunt,
>> but as has been pointed out (many times), it matters what considerations
>> were used to draw them.
>
> Of course it does, but that is a completely different issue. I tried
> to make that clear in a post I did to the "...monkeys?" thread this
> morning.

I'm afraid you did not make that clear as far as I can tell.

> By the way, what you write below reads like a merciless lampoon of
> the second half of what I posted there. Is that just a coincidence?
>
>> "I heard from Van Valen" at last year's GSA meeting
>> isn't very falsifiable, or even believable.
>
> How about "Van Valen was kind enough to hand me a preprint,
> which he hasn't published up to now (1988), in which he showed the
> detailed analysis that led to the narrowing down you see
> in Fig.21.8 (b) of the ancestry of three distinct orders of
> ungulates and the order Cetacea to various subfamilies of
> Condylartha, and of nine other orders to two of its families."
>
> Why else would Carroll go out on a limb the way he did
> in his book?

This is all pointless speculation. We don't know what data Van Valen
had, how he analyzed it, what he communicated to Carroll, or what
Carroll did and didn't take from it. Supposedly, this was an example of
how old-time analyses worked; or that's what I thought. What was this an
example of?

[Peter Nyikos]

Substitute "Elpistostege" for "Acanthostega" in that pretty tree
and the analogy with the platypus and echidnas might become clear,
especially if for Acanthostega you now substitute a cow and for
Ichthyostega you substitute a rabbit in MY tree with the platypus
and the echidna.

Just how close to your direct line of ancestry do you think a platypus,
echidna, cow, and rabbit are?

Don't get me wrong -- I don't think those very early terrestrial animals
are that far from our ancestry. I just point out that the phylogenetic
tree tells nothing about that one way or another.

> If you look at Wikipedia on Tetrapoda
> https://en.wikipedia.org/wiki/Tetrapod
> it has a tree in the section "Stem Tetrapoda" that shows the same
> Stegocephalia but extended backwards in time to the separation of
> lungfish (Dipnomorpha).

And I could extend my analogous tree back to *Cynognathus*
and *Dimetrodon* etc. -- but what's the point?

> It then shows several different hypotheses ab
> out the relationships that ended up producing the modern amphibians,
> the Lissamphibia, and the Amniota.
>
> It all looks very clear and easy to see to me. I don't see any
> problem with students learning and understanding the relationship
> between the fossil groups and modern tetrapod groups.

All fine and dandy, if this one-dimensional concept of "relationship"
is all you are interested in having students learn.

> I can see no reason to declare or even suggest that Tiktaalik is THE
> ancestor of one group, that Acanthostega is THE ancestor of a subgroup
> and the Ichthyostega THE ancestor of a sub-subgroup which includes the
> two sub-sub-subgroups Amniota and Lissamphibia.

Yeah, the fact that that next to last sub-sub-subgroup includes
not only ourselves but every vertebrate we are likely to see
away from the nearest body of water is of no interest to
our students, eh? Might as well treat our students as automatons
whose only purpose in class is to regurgitate information
that we feed them, eh?

> And for more than a century we have known that it really is not
> possible to identify the specific ancestral form. Go back in time to
> the small village in eastern Belarus where my mother's family came
> from. Dig up a grave and there is no way of telling whether that was
> my great-great-...-great grandfather or only a
> great-great-great-....-great uncle who was not my ancestor at all.

Better yet, I'll go to Minsk and try to find the remains of someone
who is buried in one ITS cemeteries, and who had a great-grandparent that
was one of your mother's remote ancestors.

Or would I have better luck looking around the cemeteries of Moskva?

Are you beginning to see the point of my analogies?

[jillery]

On Thu, 19 Nov 2015 10:18:19 -0800 (PST), Peter Nyikos
<nyi...@bellsouth.net> wrote:

>On Thursday, November 19, 2015 at 12:44:08 PM UTC-5, erik simpson wrote:
>> On Thursday, November 19, 2015 at 9:24:09 AM UTC-8, Peter Nyikos wrote:
>> > On Wednesday, November 18, 2015 at 8:44:11 PM UTC-5, erik simpson wrote:
>> > <...>
>>
>> > > I have a well-intentioned suggestion: take this discussion to SBP. You have
>> > > a tendency in this group to say things that sound to others condescending and/or
>> > > insulting.
>> >
>> > So does everyone else posting to this thread, including you.
>
>I made this comment before I saw Oo Tiib's post. I cannot recall
>anything insulting or condescending written by him.

Maybe I'm slipping, but I haven't posted anything insulting or
condescending to this thread, either.

><snip for focus>
>
>> > But I'm willing to meet you halfway. After I see that this has
>> > posted, I will post an abridged version of the post that was
>> > the subject of your testy comments -- a post to which no one has
>> > yet replied -- where I cut out all personal references.
>> >
>> > I hope you will be willing, in return, to tell me what, if anything,
>> > you still find unclear about it.
>> >
>> > Peter Nyikos
>>
>> Fair enough. But take it over to sci.bio.paleontology, please.
>
>I respectfully decline to do so. There are people posting to this
>thread whom I have never seen over there.

Who would that be, exactly? So far, it's mostly just you and
Harshman, with cameo appearances by myself, Oo Tiib, rnorman, Erik
Simpson, and Mark Isaak. I bet even Oo Tiib would be glad to see your
posts in a different newsgroup.

[erik simpson]

I've left all the stuff above unsnipped so I won't be accused of snipping
all the good stuff (watever it may be).

I still don't see the value in the concept of a "possible ancestral candidate".
When a new fossil is found, the first question addressed is how the
characters of the new fossil compare to similar previously known specimens.
I've never heard of anyone applying tests to falsifiy possible ancestry.

Of the two statements "Tiktaalik is a possible ancestor to tetrapods" and
"Tiktaalik is representative of the sort of lobe-finned fishes ancestral to
tetrapods", I think an interested layman wouldn't find a huge gulf of meaning,
and the second statement is actually true, while the first presents a
speculation.

When personal communication is cited in sceintific contexts, it's generally
restricted to specific points, rather than a detailed argument. I presented no
"vicious lampoon". Whether Van Valen's information was delivered in a bar or
in preprint form, if it isn't generally available it can't be discussed
intelligently.

Another request, if it isn't too presumptuous: please don't insert comments
within paragraphs. It makes it difficult to follow the argument, even if the
point were one's own.

[Paul J Gans]

Well put!

--
--- Paul J. Gans

[RSNorman]

On Thu, 19 Nov 2015 13:15:38 -0800 (PST), Peter Nyikos

Sorry, I am not beginning to see the point.

The purposel of a genealogical tree is to show relationships, not
timings. In human trees, as you go down different lines of descent
the generations become widely out of synchronization. The whole
purpose of a tree is to show the relationship between parent and
offspring, the pattern of branching. Nothing else matters. But that
relationship, the nested hierarchy inherent in the directed tree, is
everything. So to teach evolution, that is what you want. That is
what you need. That is what you get.

It is also common to draw trees with the location of specific
identified species, living or fossil, to be located on a time axis.
That will show that Tiktaalik occured some 10 million years before
Acanthostega and that both of these were some 370 million years before
us.

I do not know what tree you are talking about when you refer to
problems with platypus and echidna. If you look up a tree for the
clade containing mammals
https://en.wikipedia.org/wiki/Mammaliaformes
you see platypus and echidna on one branch with a separate group
including marsupials and placentals on another. Cladistics reproduces
traditional description.

You wrote

>Yeah, the fact that that next to last sub-sub-subgroup includes
>not only ourselves but every vertebrate we are likely to see
>away from the nearest body of water is of no interest to
>our students, eh?

That said sub-sub-subgroup includes not only ourselves but every
vertebrate we are likely to see away from water is exceptionally
important. That is a major teaching point -- that we are part of the
same group as all those others.

But frankly when I taught intro biology trying to get across the
pattern of relationships and characteristics of the major phyla and
classes, when getting into the origin of the mammals it made
absolutely no sense to differentiate bats from cows from humans. They
are all placental mammals, no more, no less.

When talking about the properties of amniotes and the truly
revolutionary differences between the amphibian/fish embryo and the
amniote embryo, all the terrerstrial amniotes are lumped together. A
bird egg with developing embryo is essentiall the same as a
"reptilian" egg with developing embryo and and placental embryo/fetus
is also essentially the same: egg sac, amnion, chorion, allantois.

My point about those cemetaries is that you seemed to want to identify
THIS SPECIFIC fossil as THE SPECIFIC ancestral form of some group.
That is not possible. It is not possible to say that Tiktaalic was an
ancestor of anything specific. It IS possible to say that it is
extremely likely that something no doubt extremely similar and very
closely related to Tiktaalik and which lived at the same time or a bit
before Tiktaalik, was the actual ancestral form and, maybe, it might
have been Tiktaalik itself.

Biology is now taught according to cladistic, phylogenetic principles.
Students are learning about the diversity of modern life and the
pattern of evolution which resulted in modern life as well as the
existence and relative placement of a variety of fossil forms. This
information about biological diversity and origins is being
disseminated to the general public. I don't see any problem that we
have now that could be helped by reverting to older systems when we
had far less information and based hypothesis on very limited data and
which was undergoing constant revision in any case.

[Peter Nyikos]

Good luck on getting anyone but one or two specialists to assimilate
the minute descriptions of a thousand or so characters in a meaningful
way, what with all the specialized jargon about cranial bones,
etc.

And where are the descriptions to be found? You don't get them from a
cladogram like you see in the American Museum of Natural History where
just one or two apomorphies are listed at each node.

And even that handful of specialists wouldn't do anything so unrewarding
(as in avoiding the second half of "Publish or perish") as paying a
forensic specialist to reconstruct what the LCA might have looked
like, when such specialists would have their hands full just working
from detailed pictures of the skeleton of an ACTUAL fossil.

But hey, if what you want is a nanny state where "experts" in one
of myriads of scientific specialties become oracles to whom everyone
must turn to get in-depth information, you are certainly arguing
in the right direction.

> > 2. To help people in the field understand the significance of a find
> > that would shed more light on whether we have a true ancestral species
> > on our hands.
>
> That falls into the category of press-release hype. It's commonly done,
> and it's reprehensible.

Hilarious misreading of "in the field". I'm talking about professional
paleontologists who may come across a fossil in their digs that
matches what they know about the remains of the ancestor candidate,
and if they can't find one of the key characters that could falsify
the status, they can search the surroundings diligently to see whether
the last rain may have washed it down the slope.

But hey, if what you want is a nanny state where the people in fossil
digs just prepare the fossils for study by the tiny handful of
specialists who know about those elusive characters, you are
certainly on the right track about it being reprehensible to
show them replicas of the fossils that we have of ancestor
candidates.

> > 3. To generate public interest in special cases. Just look at all the
> > excitement over the discovery of *Tiktaalik*. Who would have cared
> > about it if it were not at least close to our own ancestry, and
> > have filled a morphological gap between panderichthyan fish and
> > such early amphibians as *Acanthostega* and *Ichthyostega*?
>
> It's the filling of the gap that's interesting, not any sort of ancestor
> candidacy. You just don't need this.

Hey, if you want a nanny state where everyone but specialists is
in the dark about whether the way that gap was filled is any better
than the way a cow fills the gap between a platypus and the clade
of primates, you are absolutely right in saying "we don't need this"

> I will also note that all your appeals are to popular imagination, not
> to what's needed for science. While popular appeal is not a bad thing, I
> thought we were talking about science here. No?

I suppose you are too slow on the uptake to realize that what you
are writing here is GIGO in the light of what I said about
fossil digs, etc. so I will leave it up to someone besides myself who
isn't afraid to rock your boat to explain this to you.

If someone else like that exists, you might not be quite as creative
with THAT person as you can be with me in "not understanding" what
all that talk of nanny state had to do with anything you wrote.

> > Phylogenetic trees of the sort that we are treated to nowadays
> > say NOTHING about this aspect of Tiktaalik. For all anyone
> > unfamiliar with it and *Acanthostega* could tell, they could be just
> > as far off our ancestry as the platypus and the echidnas,
> > for all that the tree is telling them.
>
> I suppose that depends on what you mean by "as far off our ancestry".
> What do you mean? Are you referring to the length of the terminal
> branch? Why should that be important?

Since you "only see the posts you want to see" I'll just hang around
to see which of my posts to repeat information from. If you keep avoiding
the posts I have in mind here, I'll clue you in after Thanksgiving break
on the answers to these questions.

> >> A plausible statistical case could be made that
> >> Charlemagne was an ancestor of mine, and he might be, but so what?
> >
> > Substitute *Tiktaalik* for Charlemagne, and maybe you'll see my point.
>
> Different case entirely,

Of course it is. Without the substitution, you have a merciless lampoon
of my position.

> and the question "so what" still applies.

And you have everyone here besides me (except maybe jillery and Oo Tiib)
eating out of your hand to the extent that blatant assertions like
this by you are treated as gospel truth.

If you disagree, kindly give some reasoning behind that last remark.

Remainder deleted, to be replied to later.

Peter Nyikos

[erik simpson]

On Thursday, November 19, 2015 at 3:29:06 PM UTC-8, Peter Nyikos wrote:
> <...>

> Hey, if you want a nanny state where everyone but specialists is
> in the dark about whether the way that gap was filled is any better
> than the way a cow fills the gap between a platypus and the clade
> of primates, you are absolutely right in saying "we don't need this"
>

A cow is the missing link between platypuses and primates? We REALLY don't
need this. Why all the drama? How would the assertion that Taktaalik may
be ancestral make everything clear to everyone, when in all likelihood it
isn't?

[John Harshman]

So again, we aren't talking about science any more but about public
outreach. What you need in that case is for an artist who understands
the character descriptions to make a picture of the reconstructed
ancestor that incorporates them. Still don't need an "ancestor candidate".

> And where are the descriptions to be found? You don't get them from a
> cladogram like you see in the American Museum of Natural History where
> just one or two apomorphies are listed at each node.

Descriptions are found, if you care to, by combining data with a tree
and observing what is reconstructed for the node of interest. This of
course requires that you have a tree, a data set, and a program that
does character state optimizations at internal nodes. Mesquite is useful
for this. Find your own data set and tree.

> And even that handful of specialists wouldn't do anything so unrewarding
> (as in avoiding the second half of "Publish or perish") as paying a
> forensic specialist to reconstruct what the LCA might have looked
> like, when such specialists would have their hands full just working
> from detailed pictures of the skeleton of an ACTUAL fossil.

That could be. But if you want a picture of the ancestor, that's what
you have to do. Calling Tiktaalik an "ancestor candidate" certainly
doesn't help.

> But hey, if what you want is a nanny state where "experts" in one
> of myriads of scientific specialties become oracles to whom everyone
> must turn to get in-depth information, you are certainly arguing
> in the right direction.

!

>>> 2. To help people in the field understand the significance of a find
>>> that would shed more light on whether we have a true ancestral species
>>> on our hands.
>>
>> That falls into the category of press-release hype. It's commonly done,
>> and it's reprehensible.
>
> Hilarious misreading of "in the field". I'm talking about professional
> paleontologists who may come across a fossil in their digs that
> matches what they know about the remains of the ancestor candidate,
> and if they can't find one of the key characters that could falsify
> the status, they can search the surroundings diligently to see whether
> the last rain may have washed it down the slope.

Professional paleontologists don't need press release hype. They
understand the significance of a fossil without it. "Ancestor
candidates" would be important if indeed paleontologists were looking
for ancestors. But they aren't. You are locked into the mental state of
50 years ago. Or more.

> But hey, if what you want is a nanny state where the people in fossil
> digs just prepare the fossils for study by the tiny handful of
> specialists who know about those elusive characters, you are
> certainly on the right track about it being reprehensible to
> show them replicas of the fossils that we have of ancestor
> candidates.

So now you aren't talking about professional paleontologists any more.
You're talking, perhaps, about volunteer (or perhaps even paid) helpers
who are not paleontologists. You're saying we should lie to them to make
their work more exciting, because getting them excited by telling them
the truth is too hard.

>>> 3. To generate public interest in special cases. Just look at all the
>>> excitement over the discovery of *Tiktaalik*. Who would have cared
>>> about it if it were not at least close to our own ancestry, and
>>> have filled a morphological gap between panderichthyan fish and
>>> such early amphibians as *Acanthostega* and *Ichthyostega*?
>>
>> It's the filling of the gap that's interesting, not any sort of ancestor
>> candidacy. You just don't need this.
>
> Hey, if you want a nanny state where everyone but specialists is
> in the dark about whether the way that gap was filled is any better
> than the way a cow fills the gap between a platypus and the clade
> of primates, you are absolutely right in saying "we don't need this"

How did "nanny state" come into it? What's the state? Now what you
really do here is point out that this or that specimen has particular
primitive characters that make it interesting as a transitional fossil.
Just like we do with Tiktaalik, for example. You really are so locked
into your mindset that you seem not to be aware of how paleontologists
actually talk these days.

>> I will also note that all your appeals are to popular imagination, not
>> to what's needed for science. While popular appeal is not a bad thing, I
>> thought we were talking about science here. No?
>
> I suppose you are too slow on the uptake to realize that what you
> are writing here is GIGO in the light of what I said about
> fossil digs, etc. so I will leave it up to someone besides myself who
> isn't afraid to rock your boat to explain this to you.

I doubt any explanations will be forthcoming, but I'll wait here just in
case.

> If someone else like that exists, you might not be quite as creative
> with THAT person as you can be with me in "not understanding" what
> all that talk of nanny state had to do with anything you wrote.

It's clear you won't explain. Beneath you, I suppose.

>>> Phylogenetic trees of the sort that we are treated to nowadays
>>> say NOTHING about this aspect of Tiktaalik. For all anyone
>>> unfamiliar with it and *Acanthostega* could tell, they could be just
>>> as far off our ancestry as the platypus and the echidnas,
>>> for all that the tree is telling them.
>>
>> I suppose that depends on what you mean by "as far off our ancestry".
>> What do you mean? Are you referring to the length of the terminal
>> branch? Why should that be important?
>
> Since you "only see the posts you want to see" I'll just hang around
> to see which of my posts to repeat information from. If you keep avoiding
> the posts I have in mind here, I'll clue you in after Thanksgiving break
> on the answers to these questions.

You are an appalling human being. Or at least your internet persona is.

>>>> A plausible statistical case could be made that
>>>> Charlemagne was an ancestor of mine, and he might be, but so what?
>>>
>>> Substitute *Tiktaalik* for Charlemagne, and maybe you'll see my point.
>>
>> Different case entirely,
>
> Of course it is. Without the substitution, you have a merciless lampoon
> of my position.

>> and the question "so what" still applies.
>
> And you have everyone here besides me (except maybe jillery and Oo Tiib)
> eating out of your hand to the extent that blatant assertions like
> this by you are treated as gospel truth.
>
> If you disagree, kindly give some reasoning behind that last remark.

A plausible statistical case could be made that Tiktaalik was an
ancestor of mine, and he might be, but so what? How would you answer that?

(Now in fact one could not make such a plausible statistical case for
Tiktaalik. The reason you could for Charlemagne is that we have a single
population, and the number of ancestors doubles with each generation, so
that far enough back everyone in Europe is probably descended from
everyone in Europe at that time who has descendants. No analogous case
can be made for Tiktaalik. In fact, considering the rarity of
preservation, it's quite unlikely that we have found the actual
ancestral species that lived at the time of Tiktaalik. Still, so what?)

[erik simpson]

Attempting to put aside the acrimony that threatens to derail this whole topic,
I and many acknowledge interest is what our ancestors of the deep past were like.

Have you seen Dawkins' "The Ancestor's Tale"? It addresses this subject and is
a fascinating book. Even though it dates to 2005, there are already some
details that have changed, particularly in the area of molecular dating, but in
all, it's pretty up to date.

I learned a lot about the methods and terminology of modern evolutionary
biology, and lots of interesting details as well. Subtle and difficult topics
are not dodged, and the reader is not talked down to. Dawkins' well-known
aversion to religion is not on display here, so there is no reason to either to
read or ignore the book on those grounds.

Many of the illustrations of our possible ancestors are quite fine, particularly
in the UK version, which I highly recommend. Tiktaalik isn't there,
Acanthostega is, but it isn't presented as our ancestor or even possibly our
ancestor. The discussion is more nuanced than that and I doubt the views
expressed in the book have been influenced by anyone from TO.

[John Harshman]

I love that book.

[Peter Nyikos]

On Thursday, November 19, 2015 at 6:59:06 PM UTC-5, erik simpson wrote:
> On Thursday, November 19, 2015 at 3:29:06 PM UTC-8, Peter Nyikos wrote:
> > <...>
>
> > Hey, if you want a nanny state where everyone but specialists is
> > in the dark about whether the way that gap was filled is any better
> > than the way a cow fills the gap between a platypus and the clade
> > of primates, you are absolutely right in saying "we don't need this"
> >
>
> A cow is the missing link between platypuses and primates?
> We REALLY don't need this.

I'll give you the benefit of the doubt and assume you've missed
my reply to Richard Norman which was made BEFORE the post
to which you are replying, where the analogy was first tendered:

https://groups.google.com/forum/#!original/talk.origins/9C4Oh5U8kZo/jfCB-iBRBQAJ
Message-ID: <9a9dfaa6-b0fe-43b0...@googlegroups.com>

> Why all the drama?

Believe it or not, you are indulging in more drama here than I did
in the linked post.

And, just in case you are the type who insists on being told
things directly instead of being referred to other posts on the
same thread, I'm quoting the relevant part from my reply to Richard,
which begins with my words from the post to which Richard replied:

> > Just look at all the
> >excitement over the discovery of *Tiktaalik*. Who would have cared
> >about it if it were not at least close to our own ancestry, and
> >have filled a morphological gap between panderichthyan fish and
> >such early amphibians as *Acanthostega* and *Ichthyostega*?
> >

> >Phylogenetic trees of the sort that we are treated to nowadays
> >say NOTHING about this aspect of Tiktaalik. For all anyone
> >unfamiliar with it and *Acanthostega* could tell, they could be just
> >as far off our ancestry as the platypus and the echidnas,

> >for all that the tree is telling them.
> >
>

> Once again let me admit to being totally ignorant of just what the
> problem is.
>
> I found Tiktaalik on Wikipedia and it referred me to the clade
> Stegocephilia. I looked that up on Wikipedia
> https://en.wikipedia.org/wiki/Stegocephalia
> and it showed me a very pretty phylogenetic tree, illustrated even,
> that shows just how Tiktaalik, Acanthostega, and Ichthyostega are
> related to Tetrapoda.

Substitute "Elpistostege" for "Acanthostega" in that pretty tree
and the analogy with the platypus and echidnas might become clear,
especially if for Acanthostega you now substitute a cow and for
Ichthyostega you substitute a rabbit in MY tree with the platypus
and the echidna.

Just how close to your direct line of ancestry do you think a platypus,
echidna, cow, and rabbit are?

Don't get me wrong -- I don't think those very early terrestrial animals
are that far from our ancestry. I just point out that the phylogenetic
tree tells nothing about that one way or another.

====================== end of excerpt

> How would the assertion that Taktaalik may
> be ancestral make everything clear to everyone, when in all likelihood it
> isn't?

Since I never made such an assertion, it's fascinating to see
that Richard asked almost the same thing in his post:

_________________second excerpt__________________

> I can see no reason to declare or even suggest that Tiktaalik is THE
> ancestor of one group, that Acanthostega is THE ancestor of a subgroup
> and the Ichthyostega THE ancestor of a sub-subgroup which includes the
> two sub-sub-subgroups Amniota and Lissamphibia.

Yeah, the fact that that next to last sub-sub-subgroup includes
not only ourselves but every vertebrate we are likely to see
away from the nearest body of water is of no interest to
our students, eh? Might as well treat our students as automatons
whose only purpose in class is to regurgitate information
that we feed them, eh?

+++++++++++++++++++++++++++++ end of second excerpt

I humored Richard by going along with the same misconception
under which you are laboring [1] but the fact is, I do NOT
think Tiktaalik is ancestral to us -- only that it is much less
disparate from our ancestry than a platypus or even a rabbit [2].

[1] People here don't seem to care much one way or the other
about the abortion issue, so some of them may be unaware of
a pro-abortion [not just pro-choice] slogan: "Don't labor under
a misconception -- get an abortion!"

[2] Let me tell you something your best friends [do they include
Harshman and Norman?] won't tell you: it is quite proper to
say that platypuses are "transitional" between birds and mammals,
and in a similar vein, it is proper to say that cows are
"transitional" between platypuses and humans. Besides the
obvious facts that the phylogenetic tree of synapsids tells you,
one can also say that cows are "less placental" than we are.
Check out any good book on comparative embryology. Cattle don't have
an invasive trophoblast like ours. Their chorion has tufts of villi
scattered over its surface to draw nourishment from the mother cow.

Peter Nyikos

[Peter Nyikos]

On Thursday, November 19, 2015 at 7:09:05 PM UTC-5, John Harshman wrote:
> On 11/19/15 3:25 PM, Peter Nyikos wrote:
> > On Thursday, November 19, 2015 at 3:44:08 PM UTC-5, John Harshman wrote:
> >> On 11/19/15 11:40 AM, Peter Nyikos wrote:

<snip things to be deal with in a separate post>

> >>> 2. To help people in the field understand the significance of a find
> >>> that would shed more light on whether we have a true ancestral species
> >>> on our hands.
> >>
> >> That falls into the category of press-release hype. It's commonly done,
> >> and it's reprehensible.
> >
> > Hilarious misreading of "in the field". I'm talking about professional
> > paleontologists who may come across a fossil in their digs that
> > matches what they know about the remains of the ancestor candidate,
> > and if they can't find one of the key characters that could falsify
> > the status, they can search the surroundings diligently to see whether
> > the last rain may have washed it down the slope.
>
> Professional paleontologists don't need press release hype.

They also don't need elitists like you referring to precise
replicas of fossils as "press release hype".

> They understand the significance of a fossil without it. "Ancestor
> candidates" would be important if indeed paleontologists were looking
> for ancestors.

Ignoring of the concept of serendipity, noted.

> But they aren't. You are locked into the mental state of
> 50 years ago. Or more.

I could say what you are locked into, but I'll refrain for the nonce.

>
> > But hey, if what you want is a nanny state where the people in fossil
> > digs just prepare the fossils for study by the tiny handful of
> > specialists who know about those elusive characters, you are
> > certainly on the right track about it being reprehensible to
> > show them replicas of the fossils that we have of ancestor
> > candidates.
>
> So now you aren't talking about professional paleontologists any more.
> You're talking, perhaps, about volunteer (or perhaps even paid) helpers
> who are not paleontologists.

I dispute that, but now that we are on the subject, let's include
them too. Let's also include people who pay over $11,000 apiece to
participate on one of these digs. There's an ad I saw just the other day
putting that kind of price tag on joining a two-week expedition in
Australia, and it doesn't even include airfare to Australia.

> You're saying we should lie to them to make
> their work more exciting, because getting them excited by telling them
> the truth is too hard.

This grotesquely dishonest use of the word "lie" makes me glad that I
didn't even consider accepting your disingenuous suggestion:

Then perhaps you would consider following your sbp posting
standards here, at least in this thread. That's what Erik is
really asking for.

Funny how those standards went down your memory hole when you
maliciously insinuated on the "...monkeys?" thread that I had
a derogatory opinion of Erik and Richard.

Anyway, I *will* follow those standards here for everyone I've
seen post here so far EXCEPT for you. I don't care how little
or how much Erik's heart may bleed for you as a result.

In fact, your whole last sentence -- not just the use of "lie" --
is so outrageously propagandistic that it reminds me of the time
you suggested I apply the sbp posting standards to you all
through talk.origins.

My reply, "I'd as soon give a loaded gun to a serial killer"
was extremely dramatic, but it underscored the fact that
I'd never be stupid enough to do YOU such a favor in this politically
charged newsgroup where you wield such influence.

[erik simpson]

That Richard and I both misunderstood your comments in the same way might
suggest to you the possibility that your exposition wasn't clear, but I'm
sure you'd rather think that we're merely retarded. Allow me to quote (I hope
one last time) one of your paragraphs from above:

*******************************

"Substitute "Elpistostege" for "Acanthostega" in that pretty tree
and the analogy with the platypus and echidnas might become clear,
especially if for Acanthostega you now substitute a cow and for
Ichthyostega you substitute a rabbit in MY tree with the platypus
and the echidna. "

*******************************

As further proof of my retardation, I admit I have no idea what that means.

It's been pointed out to you (again, many times) that phylogenetic trees do not
represent disparity. Indeed, it's difficult to quantify disparity at all
unless you're dealing with reasonable closely related organisms. I will
agree that it's probable that Tiktaalik could have been less disparate wrt
our contemporaneous ancestor with it than the disparity between us and rabbits,
but again, so what?

As for whatever meaning you give to 'transitional' in your closing remarks,
it obviously has no bearing on phylogeny. I'll have no more to say, since
this has become abusive and boring.

[Peter Nyikos]

On Thursday, November 19, 2015 at 4:44:06 PM UTC-5, jillery wrote:
> On Thu, 19 Nov 2015 10:18:19 -0800 (PST), Peter Nyikos
> <nyi...@bellsouth.net> wrote:
>
> >On Thursday, November 19, 2015 at 12:44:08 PM UTC-5, erik simpson wrote:
> >> On Thursday, November 19, 2015 at 9:24:09 AM UTC-8, Peter Nyikos wrote:
> >> > On Wednesday, November 18, 2015 at 8:44:11 PM UTC-5, erik simpson wrote:
> >> > <...>
> >>
> >> > > I have a well-intentioned suggestion: take this discussion to SBP. You have
> >> > > a tendency in this group to say things that sound to others condescending and/or
> >> > > insulting.
> >> >
> >> > So does everyone else posting to this thread, including you.
> >
> >I made this comment before I saw Oo Tiib's post. I cannot recall
> >anything insulting or condescending written by him.
>
>
> Maybe I'm slipping, but I haven't posted anything insulting or
> condescending to this thread, either.

No, you've done just fine on this thread. I was speaking
of overall performance the whole time I've been here these
last five years. Granted, I've seen very little in the way
of posts by Oo Tiib, but so far, what I said up there holds.

> ><snip for focus>
> >
> >> > But I'm willing to meet you halfway. After I see that this has
> >> > posted, I will post an abridged version of the post that was
> >> > the subject of your testy comments -- a post to which no one has
> >> > yet replied -- where I cut out all personal references.
> >> >
> >> > I hope you will be willing, in return, to tell me what, if anything,
> >> > you still find unclear about it.
> >> >
> >> > Peter Nyikos
> >>
> >> Fair enough. But take it over to sci.bio.paleontology, please.
> >
> >I respectfully decline to do so. There are people posting to this
> >thread whom I have never seen over there.
>
>
> Who would that be, exactly? So far, it's mostly just you and
> Harshman, with cameo appearances by myself, Oo Tiib, rnorman, Erik
> Simpson, and Mark Isaak.

You can now add Paul Gans. I don't recall ever seeing him there, nor
Oo Tiib, nor Mark Isaak, and I don't recall seeing anything there by you
this year.

That newsgroup had fallen on very hard times five years ago, when I
returned to it after a decade of absence. Harshman and I early on
made a commitment to try and keep it alive. Aided by Erik, Richard and
(more sporadically) Inyo, we were doing a passable job of it despite
massive inundations of spam ads from time to time and frequent posts
by a crank seeing blood corpuscles etc. in Martian meteorites.

Then the newsgroup almost died when Thrinaxodon's rampage of wacko posts
with innumerable spoofed e-mail addresses and nyms, for months on end,
cluttered it up so badly that we stalwarts could hardly get a
legitimate post in edgewise. When we did, it wasn't long before it
disappeared from the top page of the "table of contents" for s.b.e.,
rendering it totally unpalatable for newcomers.

Then, *mirabile dictu*, both the spam and those wacko posts abruptly
ceased. Also some good newcomers came on board, and I suspect one
of them is actually Thrinaxodon under a new nym. [I don't care,
as long as 'e behaves him/herself.]

So I'm taking this opportunity to invite anyone reading this
and who is sincerely interested in paleontology to come join us.

Over there, for over a year now, we have been following an idea
of mine to treat that newsgroup as an "embassy" where we lay
aside our personal grievances like the best of ambassadors can.
So posting there is an experience very different from what is
going on here. We still have divergent opinions on lots of things,
but we treat them in such a way that it only adds a piquant spice
to the proceedings.

> I bet even Oo Tiib would be glad to see your
> posts in a different newsgroup.

I'll interpret that sentence in the most generous way possible,
and I hope he will join us there.

Peter Nyikos
Professor, Department of Mathematics -- standard disclaimer--
University of South Carolina
http://people.math.sc.edu/nyikos/

[Peter Nyikos]

On Thursday, November 19, 2015 at 1:44:06 PM UTC-5, John Harshman wrote:
> On 11/19/15 10:31 AM, erik simpson wrote:
> > On Thursday, November 19, 2015 at 10:19:07 AM UTC-8, Peter Nyikos wrote:

> >> On Saturday, November 7, 2015 at 1:44:46 AM UTC-5, erik simpson wrote:

> >>> If modern analysis is an improvement to deducing relationships,
> >>> why bother with outdated procedures?
> >>
> >> As I've been trying to make clear on this new thread, it isn't the
> >> modern *analysis* to which I object, nor the phylogenetic trees
> >> that result, but a lack of toleration about how the trees are
> >> split up into taxa.

<snip for focus>

> >> But when a reasonably complete fossil passes all tests that could
> >> falsify the hypothesis that the species x to which it belongs is
> >> directly ancestral to the species of Clade Y, then it seems eminently
> >> reasonable to designate it as "the ancestral candidate" [shorthand for
> >> "the best available candidate for an ancestor"].

> > This assertion in uncomfortably similar to "Bill"'s claim that the earth must
> > be unique, since we have no evidence that it isn't. Why do we need "the
> > ancestral candidate"? A plausible statistical case could be made that
> > Charlemagne was an ancestor of mine, and he might be, but so what?
> >
> >> Lacking such clear candidates, a naturalist with the expertise of
> >> Romer or Carroll could still draw "bubble diagrams" of ancestry using
> >> paraphyletic taxa organized in tree-like fashion.
> >>
> >> Kathleen Hunt's diagram of of hypothesized ancestor-descendant
> >> relationships among genera in *Equioidea* [1] is one of the few trees
> >> of the old school that goes down to the genus level.
> >>
> >> http://www.talkorigins.org/faqs/horses/horse_evol.html
> >>
> >> [1] Kathleen uses "Equidae" but her inclusion of *Palaeotherium*
> >> makes her diagram more conformal to the superfamily *Equioidea*.

<snip for focus>

> I will note that the bubble diagram and the Hunt phylogeny go beyond
> Peter's expressed main concern with classification and enter into the
> province of phylogenetic trees.

You keep trying to divide issues artificially, first with your
illogical blather about me "confusing" paraphyly with nested hierarchies,
then with your deliberate ignoring of the latter concept when pretending
to answer a question that involved it in the making of a tree
of Equidae. The question asked for a tree with subfamilies that
are clades and only clades, including ALL equids in the Hunt FAQ.
Your dishonest answer involved casting "subfamilies" to the winds.

And now you are ignoring the elementary fact that bubble diagrams
are an inevitable corollary to the very concept of "paraphyletic":
the clades that are snipped away are *necessarily* descended
from some species within the paraphyletic taxon, otherwise
it wouldn't be paraphyletic in the first place.

In this way, bubble diagrams are implicit in all the talk in the
thread about how ancestry of over a dozen orders is narrowed in Carroll's
authoritative text to families -- and, in half the cases, subfamilies
within *Condylartha*.

> So he has more against cladism than its
> philosophy of classification. This comes close to objecting to methods,
> or at least the closely related way of displaying results.

Bullshit. The methods of cladistics are tailor-made for drawing one kind
of VERY USEFUL tree -- the sort that tells where the line to Species X
forked off from the line to Species Y in relation to other lines.

The trees I was talking about *complemented* these kinds, giving
information as to where ancestral candidates are situated.
This is an aim that your ideology forbids you to treat
in any other way than with uncompromising opposition.

Your bullshit above, together with your high-handed treatment of my
own reply to this post of Erik's, help to confirm the correctness of
my decision not to even consider accepting your disingenuous suggestion:

Then perhaps you would consider following your sbp posting
standards here, at least in this thread. That's what Erik is
really asking for.

As I told you, I will follow those standards here for everyone

I've seen post here so far EXCEPT for you.

By the way, how is it that you presumed to speak for Erik?

Peter Nyikos
Professor, Dept. of Mathematics -- standard disclaimer--
University of South Carolina
nyikos @ math.sc.edu

[John Harshman]

On 11/19/15 8:08 PM, Peter Nyikos wrote:
> On Thursday, November 19, 2015 at 7:09:05 PM UTC-5, John Harshman wrote:
>> On 11/19/15 3:25 PM, Peter Nyikos wrote:
>>> On Thursday, November 19, 2015 at 3:44:08 PM UTC-5, John Harshman wrote:
>>>> On 11/19/15 11:40 AM, Peter Nyikos wrote:
>
> <snip things to be deal with in a separate post>
>
>>>>> 2. To help people in the field understand the significance of a find
>>>>> that would shed more light on whether we have a true ancestral species
>>>>> on our hands.
>>>>
>>>> That falls into the category of press-release hype. It's commonly done,
>>>> and it's reprehensible.
>>>
>>> Hilarious misreading of "in the field". I'm talking about professional
>>> paleontologists who may come across a fossil in their digs that
>>> matches what they know about the remains of the ancestor candidate,
>>> and if they can't find one of the key characters that could falsify
>>> the status, they can search the surroundings diligently to see whether
>>> the last rain may have washed it down the slope.
>>
>> Professional paleontologists don't need press release hype.
>
> They also don't need elitists like you referring to precise
> replicas of fossils as "press release hype".

It isn't the replicas I was referring to, it was calling fossils
"ancestor candidates" in an attempt to hype their significance. Now, we
all enjoy illustrations of reconstructions, and to some degree they can
be helpful in understanding the organism. Nothing wrong with that,
though it isn't precisely science. It's just of no use to claim you're
looking at a picture of the ancestor.

>> They understand the significance of a fossil without it. "Ancestor
>> candidates" would be important if indeed paleontologists were looking
>> for ancestors.
>
> Ignoring of the concept of serendipity, noted.

You sure are hostile for no apparent reason. You're saying that
paleontologists might look more diligently for material if they thought
it might be an ancestor. I'm saying they wouldn't.

>> But they aren't. You are locked into the mental state of
>> 50 years ago. Or more.
>
> I could say what you are locked into, but I'll refrain for the nonce.

Thank you for your restraint.

>>> But hey, if what you want is a nanny state where the people in fossil
>>> digs just prepare the fossils for study by the tiny handful of
>>> specialists who know about those elusive characters, you are
>>> certainly on the right track about it being reprehensible to
>>> show them replicas of the fossils that we have of ancestor
>>> candidates.
>>
>> So now you aren't talking about professional paleontologists any more.
>> You're talking, perhaps, about volunteer (or perhaps even paid) helpers
>> who are not paleontologists.
>
> I dispute that, but now that we are on the subject, let's include
> them too. Let's also include people who pay over $11,000 apiece to
> participate on one of these digs. There's an ad I saw just the other day
> putting that kind of price tag on joining a two-week expedition in
> Australia, and it doesn't even include airfare to Australia.

No problem. The more the merrier.

>> You're saying we should lie to them to make
>> their work more exciting, because getting them excited by telling them
>> the truth is too hard.
>
> This grotesquely dishonest use of the word "lie" makes me glad that I
> didn't even consider accepting your disingenuous suggestion:

I define "lie" as "say something you know to be untrue". What you
describe is precisely that. Well, it would be lying for a professional
paleontologist. It wouldn't be lying for you, but you aren't the one in
the scenario.

> Then perhaps you would consider following your sbp posting
> standards here, at least in this thread. That's what Erik is
> really asking for.
>
> Funny how those standards went down your memory hole when you
> maliciously insinuated on the "...monkeys?" thread that I had
> a derogatory opinion of Erik and Richard.

I'll point out that Erik, at least, thinks you have such an opinion. You
are again arguing by adjective. I did nothing maliciously.

> Anyway, I *will* follow those standards here for everyone I've
> seen post here so far EXCEPT for you. I don't care how little
> or how much Erik's heart may bleed for you as a result.

Why make me the exception?

> In fact, your whole last sentence -- not just the use of "lie" --
> is so outrageously propagandistic that it reminds me of the time
> you suggested I apply the sbp posting standards to you all
> through talk.origins.
>
> My reply, "I'd as soon give a loaded gun to a serial killer"
> was extremely dramatic, but it underscored the fact that
> I'd never be stupid enough to do YOU such a favor in this politically
> charged newsgroup where you wield such influence.

I hesitate to bring up the word "paranoia", so I won't. I may stop
replying in hopes that you will eventually cool down. We'll see.

[John Harshman]

No, the illogical blather was about you confusing paraphyly with ranks,
not quite the same thing.

> then with your deliberate ignoring of the latter concept when pretending
> to answer a question that involved it in the making of a tree
> of Equidae. The question asked for a tree with subfamilies that
> are clades and only clades, including ALL equids in the Hunt FAQ.
> Your dishonest answer involved casting "subfamilies" to the winds.

I disagree with this characterization.

> And now you are ignoring the elementary fact that bubble diagrams
> are an inevitable corollary to the very concept of "paraphyletic":
> the clades that are snipped away are *necessarily* descended
> from some species within the paraphyletic taxon, otherwise
> it wouldn't be paraphyletic in the first place.

I don't think bubble diagrams are an inevitable corollary. One could
easily present a standard phylogenetic tree and just divide it into
paraphyletic sectors while preserving all the information about
relationships.

> In this way, bubble diagrams are implicit in all the talk in the
> thread about how ancestry of over a dozen orders is narrowed in Carroll's
> authoritative text to families -- and, in half the cases, subfamilies
> within *Condylartha*.

I don't think so. Bubble diagrams are merely one possible way of
displaying paraphyly.

>> So he has more against cladism than its
>> philosophy of classification. This comes close to objecting to methods,
>> or at least the closely related way of displaying results.
>
> Bullshit. The methods of cladistics are tailor-made for drawing one kind
> of VERY USEFUL tree -- the sort that tells where the line to Species X
> forked off from the line to Species Y in relation to other lines.
>
> The trees I was talking about *complemented* these kinds, giving
> information as to where ancestral candidates are situated.
> This is an aim that your ideology forbids you to treat
> in any other way than with uncompromising opposition.

No complement, merely an introduction of vagueness. A cladogram would
show the same thing more clearly. No ideology, merely rationality. A
bubble diagram is just the fuzzing out of a cladogram. Inside every
bubble diagram there's an unexpressed cladogram relating the species
inside the bubbles. Take a cladogram, draw blobs around various groups
of nodes, erase the cladogram, and you have a bubble diagram.

> Your bullshit above, together with your high-handed treatment of my
> own reply to this post of Erik's, help to confirm the correctness of
> my decision not to even consider accepting your disingenuous suggestion:
>
> Then perhaps you would consider following your sbp posting
> standards here, at least in this thread. That's what Erik is
> really asking for.
>
> As I told you, I will follow those standards here for everyone
> I've seen post here so far EXCEPT for you.

Then I will probably stop responding to you in this thread. We'll see.

> By the way, how is it that you presumed to speak for Erik?

Was I wrong? Erik?

[erik simpson]

No. I tried to make that clear. In fact, I expressly mentioned the fact that
Peter had promised to behave himself in that forum. He hasn't broken that
promise (in sbp), but he hasn't said much over there for some time. I think he
must enjoy the high dudgeon sessions here.

[Peter Nyikos]

n Thursday, November 19, 2015 at 6:44:06 PM UTC-5, John Harshman wrote:
> On 11/19/15 1:20 PM, Peter Nyikos wrote:
> > On Thursday, November 19, 2015 at 12:54:07 PM UTC-5, John Harshman wrote:
> >> On 11/19/15 7:44 AM, Peter Nyikos wrote:
> >>> On Tuesday, November 10, 2015 at 5:04:36 PM UTC-5, John Harshman wrote:
> >>>> On 11/10/15 1:22 PM, Peter Nyikos wrote:
> >>>>> On Monday, November 9, 2015 at 6:39:38 PM UTC-5, John Harshman wrote:

> >>>>>> Contrast "the ancestors of dinosaurs are thecodonts",
> >>>>>> with which several generations of systematists were satisfied
> >>>>>
> >>>>> I doubt that; they just lacked the information to go within
> >>>>> Thecodontia and narrow the ancestral taxon down to one of its
> >>>>> families. See my preceding post again.
> >>>>
> >>>> No, it wasn't lack of information. It was lack of method.

Here I snipped something more appropriate to the thread where
all this originally appeared, "Did we come from monkeys?"

> >>> The information consisted of showing how Carroll DID narrow things
> >>> down to families and even subfamilies of *Condylartha*, for a DOZEN
> >>> orders of ungulates, as a result of better information becoming available.
> >>
> >> No, you showed *that* Carroll did narrow things down, not *how* he
> >> narrowed things down.
> >
> > Wrong interpretation of "how". I meant the exquisite *detail* he went
> > to in presenting the results of Van Valen's narrowing down.
>
> That isn't really what "how" means, is it?

It is a legitimate meaning, when understood generically as saying
"the way in which he went about what he did."

You are doing a variation on a Humpty Dumpty saying:

"When you use a word," said John Harshman scornfully,
"it means exactly what *I* choose it to mean
-- no more and no less."

> I don't find that detail
> exquisite, but that's a matter of opinion. All we have is a figure
> showing various extant orders arising from within the nebulous
> "Condylarthra" and an assignment of a "condylarth" family or subfamily
> as ancestral to each order. Is that (sub)family even a coherent group
> (i.e. paraphyletic rather than polyphyletic)?

You have no reason to think otherwise, given the extreme caution
Carroll exercises in all his phylogenetic trees in his combining
of lines. All of them IIRC are pre-cladistic. Also, note what
I write about the person responsible for the assignments, Leigh
Van Valen.

<snip>

> And given that Artiodactyla and Cetacea are
> assigned to two different subfamilies,

...of one family, Arctocyonidae, of the order Condylartha:

Oxyclaeninae -----> Artirodactyla

Loxolophinae ----> Perissodactyla

Triisodontinae ----> Mesonychia ----> Cetacea

He also put the latter two as descended from the first, which
means that you are quibbling about fine points. But you
certainly know how to jump to conclusions based on this quibble:

> the hypothesis, such as it is,
> isn't worth much.

This hyperbole and the "reasoning" behind it just might go over
well with Erik, Richard, and most of the occasional participants
on this thread, and I doubt that any of them are keeping up to date
on the "...monkeys?" thread, so I decided to bring it over here in
the interests of a fair dissemination of <ahem> points of view.

However, they deserve to know that Leigh Van Valen was a highly
regarded polymath, maybe even close to a Renaissance Man:

https://en.wikipedia.org/wiki/Leigh_Van_Valen

The hypothesis is his baby, and in the paper by him that
I have referenced, he says the following about the two
groups which compete for whale ancestry:

Some members of the Mesonychidae ... evolved
an astragalus remarkably similar to that of
artiodactyls. ... In fact no one has suggested
a separation at even the family level between
mesonychids with and without an artiodactyl-like
astragalus.
--"Adaptive Zones and the Orders of Mammals,"
Evolution 25:420-428 (1971) at p. 425

Funny, I thought the main selling point of the claim
that whales are NOT descended from mesonychids was that
mesonychids lacked a double-pulleyed astragalus. Also
strange: I thought this claim became news long after 1971.

> But back to basics: what was this presentation intended to demonstrate
> to me?

Wait for it.

> >>>> Cladistics
> >>>> introduced a method for rigorous analysis of data, intimately tied to
> >>>> the idea that cladistic relationships are what we should be (and are
> >>>> capable of) looking for.
> >>>
> >>> This was all you ever cared about, wasn't it? That would explain why you
> >>> so completely ignored the purpose of my having gone to all that trouble.
> >>>
> >>> It would also explain why you were so keen on knowing what methods
> >>> Carroll used, even though they were completely irrelevant to
> >>> your "No, it wasn't lack of information." All you were concerned
> >>> about was their relevance to "It was lack of method."

<snip for same reason as earlier one>

> > My SOLE reason for posting as much as I did on what Carroll wrote
> > was to show what one can do with better information as far as NARROWING
> > ANCESTRY DOWN to smaller Linnean taxa than just orders like Thecodontia.
>
> Finally, the actual point.

It was made long ago, and it should be clear to you from everything
I said in post after post that it was ALL due to your
"No, it wasn't lack of information."

<snip of irrelevant digression flying in the face of "No, it wasn't...">

Peter Nyikos

Professor, Dept. of Mathematics -- standard disclaimer--
University of South Carolina

http://people.math.sc.edu/nyikos/
nyikos "at" math.sc.edu

[RSNorman]

On Fri, 20 Nov 2015 09:35:22 -0800 (PST), Peter Nyikos
<nyi...@bellsouth.net> wrote:


<cut to the chase>

>This hyperbole and the "reasoning" behind it just might go over
>well with Erik, Richard, and most of the occasional participants
>on this thread, and I doubt that any of them are keeping up to date
>on the "...monkeys?" thread, so I decided to bring it over here in
>the interests of a fair dissemination of <ahem> points of view.
>
>However, they deserve to know that Leigh Van Valen was a highly
>regarded polymath, maybe even close to a Renaissance Man:
>
>https://en.wikipedia.org/wiki/Leigh_Van_Valen
>
>The hypothesis is his baby, and in the paper by him that
>I have referenced, he says the following about the two
>groups which compete for whale ancestry:
>
> Some members of the Mesonychidae ... evolved
> an astragalus remarkably similar to that of
> artiodactyls. ... In fact no one has suggested
> a separation at even the family level between
> mesonychids with and without an artiodactyl-like
> astragalus.
> --"Adaptive Zones and the Orders of Mammals,"
> Evolution 25:420-428 (1971) at p. 425
>
>Funny, I thought the main selling point of the claim
>that whales are NOT descended from mesonychids was that
>mesonychids lacked a double-pulleyed astragalus. Also
>strange: I thought this claim became news long after 1971.
>

Perhaps the reason that John's so-called hyperbole resonates with me
is that it agrees with the perspective of all my colleagues before my
retirement.

Further I have absolutely no clue what you have in mind by presenting
the ideas of Van Valen. There are arguments and disagreements about
exactly how the evolution of many groups occured and what exactly are
the relationships between different groups and, especially about the
branching of the evolutionary tree that connects different groups.
This argument and disagreement has been continuous throught the
history of biology. Cladistics has not changed that. Reorganizations
of evolutionary trees has occured throughout the history of biology.
Cladistics has not changed that. What has drastically changed things
is the introduction of molecular biology data -- gene sequences. But
that has nothing to do with cladistics.

You still have not explained just how cladistic thinking or insistence
on naming only monophyletic groups has made things worse that
traditional "Linnean" categories helped. Incidentally I should add
that the topology of the tree, knowing the sequence of branchings, is
and never has been the sum of all paleontological knowledge. Of
course it is important to know the relative dates of different fossils
and to try to estimate the relative dates of branching events and to
make sure sense of just what characters were present in ancestral
forms just prior to a branching even though there is an aversion to
naming a specific species or fossil as that specific ancestor.
Cladistics does not prevent studying all those things or teaching
them.

[erik simpson]

That is funny. I thought whales were secondarily flightless birds descended
from some unnamed archosaur.

More seriously, the close relation of whales and hippos is established from
molecular analysis, using the same techniques that 'cladists' use to evaluate
phylogentic hypotheses. But maybe I'm not hearing you right again; are you
suggesting that whales' ancestors are to be found among the mesonychids after
all? Or is this just another digression from your main point, which is still
unclear to me?

[Peter Nyikos]

> >Yeah, the fact that [the] sub-sub-subgroup[s] include

> >not only ourselves but every vertebrate we are likely to see
> >away from the nearest body of water is of no interest to
> >our students, eh? Might as well treat our students as automatons
> >whose only purpose in class is to regurgitate information
> >that we feed them, eh?
> >
> >> And for more than a century we have known that it really is not
> >> possible to identify the specific ancestral form. Go back in time to
> >> the small village in eastern Belarus where my mother's family came
> >> from. Dig up a grave and there is no way of telling whether that was
> >> my great-great-...-great grandfather or only a
> >> great-great-great-....-great uncle who was not my ancestor at all.
> >
> >Better yet, I'll go to Minsk and try to find the remains of someone
> >who is buried in one ITS cemeteries, and who had a great-grandparent that
> >was one of your mother's remote ancestors.
> >
> >Or would I have better luck looking around the cemeteries of Moskva?
> >
> >Are you beginning to see the point of my analogies?
> >
> > Peter Nyikos
>
> Sorry, I am not beginning to see the point.
>
> The purposel of a genealogical tree is to show relationships, not
> timings.

Nor were we talking about timings above. You are going
off on an irrelevant digression.

> In human trees, as you go down different lines of descent
> the generations become widely out of synchronization.

Irrelevant. What counts is that you are far more closely
related to someone close to you in the tree than someone
farther away. Charlemagne was much more closely related
to his father Pepin than he was to remote descendants of his.
It's just a coincidence that those remote descendants are
far separated from him in time.

In contrast, the LCA of Amniota is supposed to be more distantly
related to the earliest species in its sister group than it is
to every single amniote that ever existed.

But this too is a digression from things we discussed above.
What you need to appreciate is that, due to the paucity of the
fossil record, we cannot decide how similar Tiktaalik is to
the LCA of itself and us just by looking at phylogenetic trees.

> The whole
> purpose of a tree is to show the relationship between parent and
> offspring, the pattern of branching. Nothing else matters.

You are making two things look synonymous which have nothing
to do with each other in phylogenetic trees that have Harshman's
stamp of approval. "parent and offspring" are censored from
those trees, all that is shown is the naked pattern of branchings.

To vary a famous saying of Stephen Jay Gould:

The complete lack of ancestor-descendent relationships in
phylogenetic systematics is the trade secret of cladists.
The trees that have their stamp of approval show species
at the tips of their branches, never at the nodes. All
the LCAs at the nodes are pure guesswork, and no names
are ever given to those LCAs, but the epithet "hypothetical"
is given them instead.

> But that
> relationship, the nested hierarchy inherent in the directed tree, is
> everything. So to teach evolution, that is what you want. That is
> what you need. That is what you get.

To teach evolution? Or to teach a grin without the rest of the
Cheshire cat, so to speak?

> It is also common to draw trees with the location of specific
> identified species, living or fossil, to be located on a time axis.
> That will show that Tiktaalik occured some 10 million years before
> Acanthostega and that both of these were some 370 million years before
> us.
>
> I do not know what tree you are talking about when you refer to
> problems with platypus and echidna.

I am talking about taking the tree for Stegocephalia, leaving off
dates, and substituting platypus and echidna for Tiktaalik and
Elpistostege, cow for Acanthostega, rabbit for Ichthyostega.

What you get when you do this is EXACTLY the right *relationship*
between these four extant animals, in the sense that you so emphatically
talked about with your bit that ended in a paraphrase of "What you see is
what you get." The branchings occur in the correct order.

But how silly it would sound to our students to talk about our
own evolution that way! How silly it sounded to Erik!

But my point is, Why should they think any differently
of the Stegocephalian tree?

Please actually LOOK at the tree while performing the substitution
exercise, don't just look incomprehendingly at the words I have posted.
If you wish, I will be glad to give you more extant mammals to put
in the place of the tips of branches that are intermediate between
the branches that lead to the taxa I named above.

I've snipped the rest, but will deal with it later if you request it.

Peter Nyikos

[John Harshman]

Sure. But you didn't show the way in which he went about what he did.
You are showing only the result.

>> I don't find that detail
>> exquisite, but that's a matter of opinion. All we have is a figure
>> showing various extant orders arising from within the nebulous
>> "Condylarthra" and an assignment of a "condylarth" family or subfamily
>> as ancestral to each order. Is that (sub)family even a coherent group
>> (i.e. paraphyletic rather than polyphyletic)?
>
> You have no reason to think otherwise, given the extreme caution
> Carroll exercises in all his phylogenetic trees in his combining
> of lines. All of them IIRC are pre-cladistic. Also, note what
> I write about the person responsible for the assignments, Leigh
> Van Valen.

I have every reason to think otherwise, as long as he provides no
evidence for his claims.

>> And given that Artiodactyla and Cetacea are
>> assigned to two different subfamilies,
>
> ...of one family, Arctocyonidae, of the order Condylartha:
>
> Oxyclaeninae -----> Artirodactyla
>
> Loxolophinae ----> Perissodactyla
>
> Triisodontinae ----> Mesonychia ----> Cetacea

Don't you see that as a nonsensical claim? You are putting
Perissodactyla and Artiodactyla equally closely related to Cetacea.

> He also put the latter two as descended from the first, which
> means that you are quibbling about fine points.

This is hardly a fine point. It's a major feature of phylogeny as we now
know it. Carroll was grossly mistaken about the phylogeny of mammalian
orders.

> But you
> certainly know how to jump to conclusions based on this quibble:

>> the hypothesis, such as it is,
>> isn't worth much.
>
> This hyperbole and the "reasoning" behind it just might go over
> well with Erik, Richard, and most of the occasional participants
> on this thread, and I doubt that any of them are keeping up to date
> on the "...monkeys?" thread, so I decided to bring it over here in
> the interests of a fair dissemination of <ahem> points of view.

Why is that hyperbole? What's wrong with the reasoning?

> However, they deserve to know that Leigh Van Valen was a highly
> regarded polymath, maybe even close to a Renaissance Man:

Is that an argument from authority? Yes, he was a smart guy, if a bit
strange. He had a number of blind spots, though, and a distaste for
cladistics in all its meanings was one of them.

> https://en.wikipedia.org/wiki/Leigh_Van_Valen
>
> The hypothesis is his baby, and in the paper by him that
> I have referenced, he says the following about the two
> groups which compete for whale ancestry:
>
> Some members of the Mesonychidae ... evolved
> an astragalus remarkably similar to that of
> artiodactyls. ... In fact no one has suggested
> a separation at even the family level between
> mesonychids with and without an artiodactyl-like
> astragalus.
> --"Adaptive Zones and the Orders of Mammals,"
> Evolution 25:420-428 (1971) at p. 425

Mesonychidae as traditionally described seems no to be a coherent group.
It's conceivable that some of them are fairly closely related to whales,
but if so they would be artiodactyls that lost the defining characters
of artiodactyls. Others are clearly outside Artiodactyla.

> Funny, I thought the main selling point of the claim
> that whales are NOT descended from mesonychids was that
> mesonychids lacked a double-pulleyed astragalus. Also
> strange: I thought this claim became news long after 1971.

Main selling point? No. There are other characters too. Yes, no
"mesonychid" is known to have a double-pulley astragalus. Van Valen must
have been talking about something else. Do you know what? And yes, whale
astragali were not known until 2000, when two of them appeared nearly
simultaneously.

>> But back to basics: what was this presentation intended to demonstrate
>> to me?
>
> Wait for it.

My patience is not infinite, and I don't see anything below.

>>>>>> Cladistics
>>>>>> introduced a method for rigorous analysis of data, intimately tied to
>>>>>> the idea that cladistic relationships are what we should be (and are
>>>>>> capable of) looking for.
>>>>>
>>>>> This was all you ever cared about, wasn't it? That would explain why you
>>>>> so completely ignored the purpose of my having gone to all that trouble.
>>>>>
>>>>> It would also explain why you were so keen on knowing what methods
>>>>> Carroll used, even though they were completely irrelevant to
>>>>> your "No, it wasn't lack of information." All you were concerned
>>>>> about was their relevance to "It was lack of method."
>
> <snip for same reason as earlier one>
>
>>> My SOLE reason for posting as much as I did on what Carroll wrote
>>> was to show what one can do with better information as far as NARROWING
>>> ANCESTRY DOWN to smaller Linnean taxa than just orders like Thecodontia.
>>
>> Finally, the actual point.
>
> It was made long ago, and it should be clear to you from everything
> I said in post after post that it was ALL due to your
> "No, it wasn't lack of information."

No, it wasn't at all clear. Nor do I see anything that appears to be a
response to that claim.

[Mark Isaak]

On 11/20/15 10:45 AM, Peter Nyikos wrote:
> [...]

> The complete lack of ancestor-descendent relationships in
> phylogenetic systematics is the trade secret of cladists.
> The trees that have their stamp of approval show species
> at the tips of their branches, never at the nodes. All
> the LCAs at the nodes are pure guesswork, and no names
> are ever given to those LCAs, but the epithet "hypothetical"
> is given them instead.

Which is how things should be, right? (Aside from the fact that it is
not a secret.) I, for one, see nothing wrong with hypothetical things
being called hypothetical. Do you?

--
Mark Isaak eciton (at) curioustaxonomy (dot) net
"Keep the company of those who seek the truth; run from those who have
found it." - Vaclav Havel

[erik simpson]

Congratulations! You've just joined Peter's cladophile shit list. I think
he's taking next week off to let the bile pressure build up, then you'll see...

[RSNorman]

On Fri, 20 Nov 2015 10:45:38 -0800 (PST), Peter Nyikos
<nyi...@bellsouth.net> wrote:

>On Thursday, November 19, 2015 at 6:29:06 PM UTC-5, RSNorman wrote:

<snip but I sincerely believe I am keeping the essential points and
only those points I wish to respond to>

>> >Are you beginning to see the point of my analogies?
>> >
>> > Peter Nyikos
>>
>> Sorry, I am not beginning to see the point.
>>
>> The purposel of a genealogical tree is to show relationships, not
>> timings.
>
>Nor were we talking about timings above. You are going
>off on an irrelevant digression.
>
>> In human trees, as you go down different lines of descent
>> the generations become widely out of synchronization.
>
>Irrelevant.

Also I like to put my responses all together at the end. So the above
is point (1).

>What counts is that you are far more closely
>related to someone close to you in the tree than someone
>farther away. Charlemagne was much more closely related
>to his father Pepin than he was to remote descendants of his.
>It's just a coincidence that those remote descendants are
>far separated from him in time.
>
>In contrast, the LCA of Amniota is supposed to be more distantly
>related to the earliest species in its sister group than it is
>to every single amniote that ever existed.
>
>But this too is a digression from things we discussed above.
>What you need to appreciate is that, due to the paucity of the
>fossil record, we cannot decide how similar Tiktaalik is to
>the LCA of itself and us just by looking at phylogenetic trees.

This is point (2)

>> The whole
>> purpose of a tree is to show the relationship between parent and
>> offspring, the pattern of branching. Nothing else matters.
>
>You are making two things look synonymous which have nothing
>to do with each other in phylogenetic trees that have Harshman's
>stamp of approval. "parent and offspring" are censored from
>those trees, all that is shown is the naked pattern of branchings.
>
>To vary a famous saying of Stephen Jay Gould:
>
> The complete lack of ancestor-descendent relationships in
> phylogenetic systematics is the trade secret of cladists.
> The trees that have their stamp of approval show species
> at the tips of their branches, never at the nodes. All
> the LCAs at the nodes are pure guesswork, and no names
> are ever given to those LCAs, but the epithet "hypothetical"
> is given them instead.
>

This is point (3)

> > But that
>> relationship, the nested hierarchy inherent in the directed tree, is
>> everything. So to teach evolution, that is what you want. That is
>> what you need. That is what you get.
>
>To teach evolution? Or to teach a grin without the rest of the
>Cheshire cat, so to speak?
>
>> It is also common to draw trees with the location of specific
>> identified species, living or fossil, to be located on a time axis.
>> That will show that Tiktaalik occured some 10 million years before
>> Acanthostega and that both of these were some 370 million years before
>> us.

I do not comment on these.

>> I do not know what tree you are talking about when you refer to
>> problems with platypus and echidna.
>
>I am talking about taking the tree for Stegocephalia, leaving off
>dates, and substituting platypus and echidna for Tiktaalik and
>Elpistostege, cow for Acanthostega, rabbit for Ichthyostega.
>
>What you get when you do this is EXACTLY the right *relationship*
>between these four extant animals, in the sense that you so emphatically
>talked about with your bit that ended in a paraphrase of "What you see is
>what you get." The branchings occur in the correct order.
>
>But how silly it would sound to our students to talk about our
>own evolution that way! How silly it sounded to Erik!
>
>But my point is, Why should they think any differently
>of the Stegocephalian tree?
>
>Please actually LOOK at the tree while performing the substitution
>exercise, don't just look incomprehendingly at the words I have posted.
>If you wish, I will be glad to give you more extant mammals to put
>in the place of the tips of branches that are intermediate between
>the branches that lead to the taxa I named above.

This is point (4).

>I've snipped the rest, but will deal with it later if you request it.

If I have snipped material you think essential, present that (and ony
that) so I can respond to specific points.

Herein lies my responses. Unfortunately I spent well over a half-hour
composing a response but ended up with that infamous "blue screen of
death" on my Windows system (7 Professional). It is the first time
that has happened on my four year old laptop. So I will try to
recreate the highlights as best I can remember. Shame on me for
getting careless and not bothering to save backups while working. I
did that regularly when I got money for working; now that I do stuff
for free I don't bother.

(1) My misunderstandings. I try to tell you, Peter, that I really
don't know what you are trying to get at. Obviously I still don't. I
don't say that this is your fault or mine, just that we seem to be
talking at cross purposes each with a very different mind set.

(2) Distance. I had a real problem understand your point but now,
with the Charlemagne/Pepin example I see where it is. You measure
"distance" topologically: the distance between two points on a tree is
how many nodes you have to traverse to move from one to the other
(whether forwards or backwards doesn't matter.) I see it entirely
differently. The "distance" refers to time, as in the "distant past".
If you traverse the tree from one point to the other, what is the node
closest to the root, presumably the most distant in time from the
present. It is a necessity that along any one sequence of
ancestry/descent the nodes have to increase in age from root to
terminal branches.

So, the most recent common ancestor of Charlemagne and Pepin is Pepin;
of Charlemagne and some modern descendant is Charlemagne. Pepin
predates Charlemagne so that relationship, even though it is
father-son, is more distant (in time) than all the others even though
many generations intercede.

That explanation fully accounts for your problem with "distance" in
sister groups. All the members of a sister group share descent from
one common ancestor; they are so related to each other. Somebody
outside that sister group will have a common ancestor with the sister
group but that common ancestor lies backware in time from that of the
sister group. The relationship is weaker, more distant.

As to how similar Tiktaalik is to that common ancestor we share with
it, that simply do not know but we do make some educated guesses that
it shares many of the truly important features we hypothesize for that
common ancestor. Note: there are existing life forms that adopt
multiple completely different life forms with different structures,
different organs, different morphology, different ecology during their
own lifetime. There are other life forms that appear by many respects
to be virtually identical to fossil forms many millions of years
(tens? hundreds?) old. Humans appear very different from chimps but
we are exceptionally closely related. Many species with far weaker
relatedness seem very similar.

(3) Ancestors and descendants. I previously (Nov. 3 in "What did the
Ancestral Ape Look Like") quoted this text:

"Paleontology has been criticised on the ground that she cannot
pretend to show the actual ancestors of living forms because, if in
the past genera and species were as abundant and as diverse as we find
them at present, it is very improbable that the bones of any
individual that just happened to be preserved are the bones of just
that species that took part in the evolution...but even then the
evidence is, I think, still just as valuable... It suffices to know
that there lived in the past a particular "group" of animals that had
many points in common with those that preceded them and with those
that came later. Whether these are the actual ancestors or not does
not so much matter, for the view that from such a group of species the
later species have been derived is far more probably than any other
view that has been proposed."

That seems to capture perfectly the derogatory comments of Gould about
"hypotheticals". But this argument about not being able to name a
specific ancestor but only some "group" has nothing to do with
cladistics. It is quite traditional, having been written almost
exactly one huncred years ago by Thomas Hunt Morgan

https://www.stmarys-ca.edu/sites/default/files/attachments/files/A_Critique_of_the_Theory_of_Evolution.pdf

There is no way possible to identify a particular species as being a
"real" ancestor. There is no value in giving species just above the
branching point (whether you prefer node or stem trees) a species name
or specifying that "this specific" fossil IS that ancestor. That is
all classical "Linnean" paleontology. Don't blame cladistics for
seeking better clarity and not simply lying to the public for the sake
of good publicity that comes from THE missing link.

(4) You substitute tree.

I do not see why one tree is any different in truth value or in
educational value from the other. What I see when you substitute
those mammals is that we have
Mammals
One group (Monotremes)
Platypus
Echidna
Another group (Placentals)
One subgroup (superorder Euarchontoglires)
Rabbit
Another subgroup (superorder Laurasiatheria)
Cow

I don't see anything at all silly about that arrangement of things.

Cows and rabbits are closer related to each other than to the more
distant platypus and echidna and vice versa. Isn't that what we
should be teaching? All I had time for in teaching intro was simply
to mention monotremes, marsupials, and placentals and indicate just
how the reproductive strategies of each of these three groups relates
to the structure of the amniotic egg and embryo.

Incidentally, all I had time about birds were that they were warm
dinosaurs with feathers. Oh, yes. they could fly.

If anyone has complaints about this coverage, recall that one semester
included 42 50 minute periods with time out for in-class exams and had
to cover genetics, population genetics, evolution, diversity of the
plant and animal kingdoms, plant and animal morphology and physiology,
animal behavior, and all of ecology; population, community, and
ecosystem.

[John Harshman]

I'm just going to insert some stuff here, no especially as a response to
you but simply as a contribution to the conversation.

Peter thinks cladograms are limited because they don't show disparity.
(I would argue that his preferred bubble diagrams are no better.) But
there is something sometimes called a phylogram, which is just a
cladogram whose branches are scaled based on the inferred amount of
evolution, or number of changes, along them. If the changes are to the
characters you think are important in judging disparity, a phylogram
becomes a much better guide to disparity than any bubble diagram. The
distance between two taxa along the tree branches is their disparity (or
would be if there were no homoplasy). And the distance between a
terminal taxon and its closest internal node is an index of its
similarity to the inferred ancestor. Why, if the branch length were
zero, it would be inferred as identical to that ancestor. Peter should
really love phylograms, and they would obviate any need to declare
"ancestor candidates".

However, I wouldn't equate this measure of disparity with relationship.

[erik simpson]

What is the term for the cladogram where the branch length represents time?

I would think the phylogram you describe would answer Peter's objections,
but I bet there'll be something else to object to.

[John Harshman]

Generally, "time-calibrated tree".

> I would think the phylogram you describe would answer Peter's objections,
> but I bet there'll be something else to object to.

We'll see.

[christi...@brown.edu]

"I am talking about taking the tree for Stegocephalia, leaving off
dates, and substituting platypus and echidna for Tiktaalik and
Elpistostege, cow for Acanthostega, rabbit for Ichthyostega.

What you get when you do this is EXACTLY the right *relationship*
between these four extant animals,"

Er, no. Not unless you're claiming that the echidna is more closely related to placental mammals than the platypus is.

[John Harshman]

He's trying to talk about branch lengths, though he doesn't quite know
it. He thinks it would be good to call Tiktaalik an "ancestor candidate"
and not to call a platypus that, though a cladogram would show them in
the same sort of position in a tree, and thus cladograms don't do what
he would like them to. Got that?

[RSNorman]

On Sat, 21 Nov 2015 19:02:44 -0800 (PST), christi...@brown.edu
wrote:

That is not what you get if you use the diagram in Wikipedia, which is
the one under discussion
https://en.wikipedia.org/wiki/Stegocephalia#Phylogeny

[Peter Nyikos]

I think you are applying double standards to yourself on the one
hand and me on the other. You take umbrage at personal attacks
by me on Harshman, and aren't the least bit concerned about whether
they are justified or not.

Here, you make a personal attack against me and aren't the least
bit concerned about whether it is justified or not.

Now the second part may not seems like a double standard, but
the difference is that:

1. I can (and will, if challenged) justify
any personal comment I make about Harshman -- the good [and
there are plenty of those] along with the bad.

2. You are making exactly the same sort of personal attack on me
that I criticized Harshman for earlier on this thread: a WAG about
my attitude towards someone else that not only has no basis in reality,
but even flies in the face of reality.

Bottom line: there is NO SUCH THING as a "cladophile shit list"
where I am concerned, because cladophilia could be a matter of
being exposed to education/indoctrination that doesn't give
people a look at the drawbacks of cladophilia.

Harshman's education/indoctrination obviously goes very deep, and it
may only be now, with the example of Stegocephalian tree --> weird
mammalian tree before him, that he can even begin to appreciate
certain distinctions which heretofore might have been presented
to him as nonexistent.

Now, if he keeps resorting to dirty tactics like the ones that
indirectly led to this thread, he will have gone beyond cladophilia
to cladomania. And THAT does have a shit list to go with it,
but the people who were on it haven't been seen by me in over 14 years,
and what I call "the statute of limitations" kicks in with the
passage of half that time.

And so the list is vacant and the people who were on it would get
a "let bygones be bygones" from me if they were to return. Of course,
if they were to display the same behavior that got them on it, I would
put them back on the list AND explain why they are on it and Harshman
is not [knock wood!].

Peter Nyikos

[erik simpson]

1) No 'double standard' is being applied.

2) I'm not 'taking umbrage' about any attacks on John Harshman.

3) I'm bemused and amused at this fairly pointless dustup. I wish you could lighten up and turn on some kind of humor detector.

Have you seen John's description of 'phylograms', where branch length of
cladograms represents some measure of disparity? Does it remedy the
drawbacks of cladistics?

[Peter Nyikos]

On Monday, November 23, 2015 at 12:48:54 PM UTC-5, Peter Nyikos wrote:
> On Friday, November 20, 2015 at 9:04:05 PM UTC-5, erik simpson wrote:
> > On Friday, November 20, 2015 at 5:34:04 PM UTC-8, Mark Isaak wrote:
> > > On 11/20/15 10:45 AM, Peter Nyikos wrote:

> Bottom line: there is NO SUCH THING as a "cladophile shit list"
> where I am concerned, because cladophilia could be a matter of
> being exposed to education/indoctrination that doesn't give
> people a look at the drawbacks of cladophilia.
>
> Harshman's education/indoctrination obviously goes very deep, and it
> may only be now, with the example of Stegocephalian tree --> weird
> mammalian tree before him, that he can even begin to appreciate
> certain distinctions which heretofore might have been presented
> to him as nonexistent.

I should add that if people here could give me some really sensible
reasons why cladistic classification should be the ONLY official
biological classification, while giving my counter-arguments a full, fair
hearing and discussion, I would "agree to disagree" and not bother them
with any more arguments.

Further than that I cannot go, because I am not a biologist and
so cannot enter the extra-logical aspects of a lifetime professional
biologist's world view. To me, as a mathematician who knows
the limits of quantification, the traditional Linnean
classification makes just too much sense to be cast to the
rubbish heap.

Peter Nyikos
Professor, Dept. of Mathematics -- standard disclaimer--
University of South Carolina

Specialty: set-theoretic topology

[erik simpson]

Since this post didn't actually address anything I said, this isn't actually
a 'reply' to anything that you said, but I'll make a couple of comments.

Do you really think that your objections to cladistic classification weren't
represented during the 'clasdistics wars'? Professional biologists did make
arguments similar to yours. Cladists won. Some held out, but not many.

As for 'indoctrination', I can address that. I have a few years on you, so
Linnean terminology was what I saw when I was a kid, although I remember
finding Romer pretty tough sledding. Later (much later) I read about the
'cladistic revolution', and wondered how it would turn out. I recall my
first impression was favorable, as it seemed to promise a path away from
purely subjective notions of how things went. I believe tha impression has
been confirmed, and has led to a much better knowledge of the big picture.

Science isn't like math. A proof by Euclid is as good as when he did it, but
observational sciences usually don't age as well. It's not that the grey
eminences where so much wrong, but that we've learned more and better.

[Peter Nyikos]

Thank you contributing to this discussion, Christine!

The substitution I had in mind used the Wikipedia tree for
Stegocephalia, which may be different from the tree you have
in mind. The wiki tree looks like this, topologically speaking:

|Tiktaalik
______|
| |Elpistostege
|
Stegocephalia_|
|
|______Elginerpeton
|
|_____Ventastega
|
|_____Acanthostega
|
|______Ichthyostega
|
|------------------------> to humans

My idea was to substitute platypus for Tiktaalik, echidna for
Elpistostege, cow for Acanthostega and rabbit for Ichthyostega.

In recognition of your fascinating research on *Procoptodon*, I am
substituting it for Elginerpeton, and an armadillo for Ventastega,
thereby "filling all the gaps" in my substituted tree.

NOTE TO OTHER READERS: Christine Janis not only determined that the
short-faced kangaroo *Procoptodon* was the heaviest kangaroo
of which we know, she also determined that it walked instead
of hopping! This is so interesting that I told my whole family
about it, even though some don't care much about paleontology,
because...well, that's another story.

Peter Nyikos
Professor, Department of Math. -- standard disclaimer --
U. of S. Carolina at Columbia
http://people.math.sc.edu/nyikos/

[Peter Nyikos]

I see spacings don't travel well in New Google Groups. I don't know
whether the displays in the post to which I am replying are as cockeyed
in other newsreaders, but if you hit "reply" to either that post
or this one in New Google Groups, everything gets straightened
out properly in the window where you compose your reply.

Here is the Wikipedia link where you find the tree to which
I am referring:

https://en.wikipedia.org/wiki/Stegocephalia

[RSNorman]

On Mon, 23 Nov 2015 11:25:46 -0800 (PST), Peter Nyikos
<nyi...@bellsouth.net> wrote:

>I see spacings don't travel well in New Google Groups. I don't know
>whether the displays in the post to which I am replying are as cockeyed
>in other newsreaders, but if you hit "reply" to either that post
>or this one in New Google Groups, everything gets straightened
>out properly in the window where you compose your reply.

I believe the issue about "spacings" relates to using variable or
fixed fonts. In a fixed font, also called fixed-width or
non-proportional, every character including the space and the dash
occupies the same horizontal space on the line. In a variable font,
different characters have different widths. If you try to align text
vertically carefuly adding space characters to make sure things line
up just so, when that same text is displayed on the alternative font
style things change drastically.

The font is controlled by the software you are using to display.
Usenet is supposed to be pure text so html markups, which do specify
what font to use, are no-nos.

[RSNorman]

Eric's take is very much like mine. John, too, except you and he have
a long and contentious relationship so his replies tend to be a bit
more intemperate.

There seem to be several issues at work here. First, cladistics
proper is a methodology for phylogenetic systematics that has been
pretty much universally accepted. When you rail against "cladistics"
it seems to me that you object to this methodology and the
overreaching importance of phylogeny as the basis for taxonomy, the
naming of things, and classification, the arrangment into groups.
Systematics itself has always meant classification by means of
phylogeny and has alway been the goal of biological classification.
Cladistics was a way of ensuring that a more objective set of criteria
are used to establish similarities and differences. So cladistics is
now simply a given.

Second, there is a "philosophical" or "stylistic" bent associated with
the more neutral methodological approach and this is where you take
issue. It has several components.

First, evolutionary ancestors cannot be determined with certainty so
the founders of a clade are hypothetical and should not be given a
species name nor should any particular fossil species be identified
definitively as an ancestor.

Second, there is a strong movement to assign official names to groups
only if they are a clade. Polyphyletic or paraphyletic groups are
disparaged. So, for example, traditionally the mammals and the birds
are so distinctive and so important as to be distinguished from the
"reptiles". I think perhaps a suitable analogy would be to say that
integers should not be grouped with the integral domains nor should
the complex numbers be grouped with the fields because these are so
distinctive and important that they deserve to stand on their own with
their own special name. Do non-commutative groups deserve a special
categorization because they form a natural grouping? Since
evolutionary relations form the foundation for all important
biological properties they should form the foundation for 'official'
names. Of course you can deal with "trees" or "herbacious plants" or
"invertebrates" just as you can deal with "aquatic vertebrates" or
"arctic mosses". Just don't give these groups official names.

Third, the Linnean ranking system of phylum, class, and order was long
ago shown to be completely inadequate first because there are
innumerable ranks and levels of branching and second because there is
no consistent and objective way of saying "this group deserves to be
an order that group merits class-hood."

This third point is nicely discussed at
http://www.miketaylor.org.uk/dino/faq/s-class/levels/index.html

The extreme version of completely eliminating all reference to rank,
the "Phylocode" is in fact quite controversial and is not at all
universally adopted. If your real objection to cladistics is with
Phylocode, then you really are in good company. However the examples
you give and the arguments you cite suggest that you have other
problems with modern ways of dealing with evolutionary relationships
and how to classify or organize the world of biology.

[Mark Isaak]

On 11/23/15 10:11 AM, Peter Nyikos wrote:
>
> I should add that if people here could give me some really sensible
> reasons why cladistic classification should be the ONLY official
> biological classification, while giving my counter-arguments a full, fair
> hearing and discussion, I would "agree to disagree" and not bother them
> with any more arguments.

Cladistic classification is the natural classification. I.e., it tells
how nature happens to arrange itself. NB: "the" natural system,
singular. Furthermore, adopting it has caused no great problem for
taxonomists. The pro side rests its case.

However, I fully expect that I will not give your counterarguments a
full hearing because I expect they will be overly verbose, mostly
off-point, and filled with calumny against various and sundry people.
If none of that applies, then I might pay attention to them.

> Further than that I cannot go, because I am not a biologist and
> so cannot enter the extra-logical aspects of a lifetime professional
> biologist's world view. To me, as a mathematician who knows
> the limits of quantification, the traditional Linnean
> classification makes just too much sense to be cast to the
> rubbish heap.

Who has suggested junking the Linnean system? The Linnaean system works
perfectly well with cladistics. As I mentioned in another post,
cladistics was what Linnaeus was aiming for.

[Peter Nyikos]

On Friday, November 20, 2015 at 2:34:05 PM UTC-5, John Harshman wrote:
> On 11/20/15 9:35 AM, Peter Nyikos wrote:
> > n Thursday, November 19, 2015 at 6:44:06 PM UTC-5, John Harshman wrote:
> >> On 11/19/15 1:20 PM, Peter Nyikos wrote:
> >>> On Thursday, November 19, 2015 at 12:54:07 PM UTC-5, John Harshman wrote:
> >>>> On 11/19/15 7:44 AM, Peter Nyikos wrote:
> >>>>> On Tuesday, November 10, 2015 at 5:04:36 PM UTC-5, John Harshman wrote:
> >>>>>> On 11/10/15 1:22 PM, Peter Nyikos wrote:
> >>>>>>> On Monday, November 9, 2015 at 6:39:38 PM UTC-5, John Harshman wrote:
> >

The all-important context, which Harshman evidently hopes readers
will ignore because of the many >>>>>>>> that appear in the
margin, is the following:

> >>>>>>>> Contrast "the ancestors of dinosaurs are thecodonts",
> >>>>>>>> with which several generations of systematists were satisfied
> >>>>>>>
> >>>>>>> I doubt that; they just lacked the information to go within
> >>>>>>> Thecodontia and narrow the ancestral taxon down to one of its
> >>>>>>> families. See my preceding post again.

Pay close attention to the FIRST sentence in the next line, readers,
and compare it with the preceding paragraph.

> >>>>>> No, it wasn't lack of information. It was lack of method.

Harshman has been indulging in one dirty debating tactic after another
to misdirect people into glossing over that first sentence, as though
it made no difference to anything that follows. Au contraire, it is THE
central issue that divides us in the sub-thread that began with that
one line, and continues to divide us below. All the rest consists
of side issues.

> >>>>> The information consisted of showing how Carroll DID narrow things
> >>>>> down to families and even subfamilies of *Condylartha*, for a DOZEN
> >>>>> orders of ungulates, as a result of better information becoming available.

<snip red herring which I hope the following explanation will resolve>

In my "I doubt that..." comment, I had only indicated THAT he had
narrowed things down. My "how" refers to *how* the narrowing down
actually went--the *form* it actually took.

And readers can see the form it took below, from a sample involving
three SUBfamilies of ONE family of this huge order [as it was
consitituted at the time, 1988].

<anip of things to be dealt with in separate reply, if you insist>

> >> And given that Artiodactyla and Cetacea are
> >> assigned to two different subfamilies,
> >
> > ...of one family, Arctocyonidae, of the order Condylartha:
> >
> > Oxyclaeninae -----> Artirodactyla
> >
> > Loxolophinae ----> Perissodactyla
> >
> > Triisodontinae ----> Mesonychia ----> Cetacea
>
> Don't you see that as a nonsensical claim?

It is you who are being nonsensical in a desperate effort to
divert attention from the fact that your "No, it it wasn't
lack of information" is being DISPUTED by the plain fact
that Van Valen went well inside Condylartha, which was a whole
order (just like Thecodontia, which got us started in this
discussion).

Do you wish to retract that "No, it wasn't for lack of information?"
If not, what is the explanation for your repeatedly deleting
it at first, and then repeatedly ignoring it later on, and finally
simulating incomprehension of its connection with the "nonsensical
claim" when you could not completely ignore it any more?

> You are putting
> Perissodactyla and Artiodactyla equally closely related to Cetacea.

WRONG! Once again you show that you do not see the distinction between
polytomy and the idea of narrowing down something into a subfamily,
but not to any sub-taxon of a subfamily. The three clades you mention could
have easily diverged *within* Oxyclaeninae in a way that does justice
to both the current paradigm and the one due to Van Valen. [1]

And even if I were putting them equally closely, etc, what of it? where
is the "nonsense" in a claim like that? You are relying on privileged information about molecularly based conclusions that came long after 1971,
when Van Valen apparently formed his hypothesis.

And keep in mind that it was just a *hypothesis*, subject to
correction as new data came in -- like every hypothesis in biology
that isn't just a datum due to direct observation.

> > He also put the latter two as descended from the first, which
> > means that you are quibbling about fine points.

> This is hardly a fine point. It's a major feature of phylogeny as we now know it.

You do not KNOW it. You seem to think, judging from a half-cocked
comment later on [deleted, but to be dealt with in my next reply to this
post of yours] that there is something very wrong with the idea
of *some* mesonychids losing the double-pulleyed astralagus which
you claim to be a "defining character" of artirodactyls, even though
such morphological concepts are rapidly becoming obsolete.

And what's wrong with triisodontine arctocyonids being artirodactyls,
and some of them passing on the double-pulleyed astralagus to
mesonychids? What's wrong with some of those losing them while others
retained them and gave rise to whales?

That only takes a minor adjustment in Van Valen's phylogeny: rather than
branching off from Oxyclaeninae *independently* of Triisodontinae,
Artirodactyla branched off by itself and then gave rise in turn to Triisodontinae. With our fossils of the time (1971) so fragmentary,
a few characters could easily have caused an adjustment like that
in Van Valen's analysis.

> Carroll was grossly mistaken about the phylogeny of mammalian
> orders.

If you knew you were on sound footing, you wouldn't feel the need
to indulge in flagrant hyperbole like "grossly," would you now?

[1] Artirodactyls and mesonychids both had paraxonian digits, as
opposed to the configuration of almost every condylath of the time,
which had the pattern seen in Perissodactyla: middle digit bigger than
all the rest. So it stands to reason that the paraxonian condition
was a synapomorphy even in the phylogeny favored by Van Valen and
through him, Carroll.

Continued in next reply to this post of yours.

Peter Nyikos
Professor, Dept. of Mathematics

U. of S. Carolina, Columbia -- standard disclaimer--
http://people.math.sc.edu/nyikos/

[Peter Nyikos]

On Friday, November 20, 2015 at 8:34:04 PM UTC-5, Mark Isaak wrote:
> On 11/20/15 10:45 AM, Peter Nyikos wrote:
> > [...]

You snipped the fact that the following is a paraphrasal of
a famous quotation by Stephen Jay Gould. I will return to
this aspect of it below.

> > The complete lack of ancestor-descendent relationships in
> > phylogenetic systematics is the trade secret of cladists.
> > The trees that have their stamp of approval show species
> > at the tips of their branches, never at the nodes. All
> > the LCAs at the nodes are pure guesswork, and no names
> > are ever given to those LCAs, but the epithet "hypothetical"
> > is given them instead.
>
> Which is how things should be, right?

You are only addressing the last sentence, whereas it is
what comes before that really sets the phylogenetic classification
apart from one that puts ancestor candidate species [a concept
explained below] at some nodes where such candidates exist,
and higher taxa where they don't.

Kathleen Hunt's tree in the Horse FAQ puts genera at
the nodes:

(Kalobatippus?)-----------------------------------------
25My \ | /
\ | /
|
35My |
Miohippus Mesohippus
| |
40My Mesohippus
|
|
|
45My Paleotherium |
| Epihippus
| |
Propalaeotherium | Haplohippus
| | |
50My Pachynolophus | Orohippus
| | |
| | |
------------------------------
\ | /
\ | /
55My Hyracotherium

In case the above tree gets cockeyed due to the fonts used, you
can find the original in
http://www.talkorigins.org/faqs/horses/horse_evol.html

> (Aside from the fact that it is
> not a secret.)

I think it is at least as much a "trade secret" as the one of which
Gould wrote.

> I, for one, see nothing wrong with hypothetical things
> being called hypothetical. Do you?

It depends on what you mean by "hypothetical". Do you deny the fact
that any two animals have an LCA? As far as we know, nobody has
ever found a fossil of the species that was the LCA of humans
and chimps, but only a creationist would deny that we DO have
an LCA and, more generally, that chimps and ourselves have
common ancestors going back beyond the first chordate.

Harshman assured me time and again that cladistic systematists
would never call a species an "ancestor candidate" even if
every fossil of individuals of that species passed every test for
falsifiability of its species being a common
ancestor of, say, humans and chimps AND if one had a complete
skeleton represented by those fossils. But they would be fools
to claim flat-out that it is NOT a common ancestor under those
conditions.

Look at it this way: had Gould written the above instead of
what he actually wrote, I think creationists would be quote-mining
that quote far more frequently than even is the case with
his actual quote now.

What would YOU say to a non-evolutionist who says, "I'm not
claiming that common descent is false, only that it is a hypothesis;
and as you can see, even cladists indirectly admit that the
existence of common ancestors is just a hypothesis"?

Peter Nyikos
Professor, Dept. of Mathematics

[erik simpson]

So if systematists would just say 'candidate ancestor' instead of 'hypothetical
ancestor', it'd be OK. Not that I think any would say either, since it's a
silly thing to say. What in the world be "every test for falsifiability
of [a species] being a common ancestor of, say, humans and chimps", with or
without a complete skeleton?

> What would YOU say to a non-evolutionist who says, "I'm not
> claiming that common descent is false, only that it is a hypothesis;
> and as you can see, even cladists indirectly admit that the
> existence of common ancestors is just a hypothesis"?
>

That's just silly. Cladists make no such statement. You could try to explain
what the actual understanding is, or you could make something up.

[John Harshman]

On 11/23/15 3:30 PM, Mark Isaak wrote:
> On 11/23/15 10:11 AM, Peter Nyikos wrote:
>>
>> I should add that if people here could give me some really sensible
>> reasons why cladistic classification should be the ONLY official
>> biological classification, while giving my counter-arguments a full, fair
>> hearing and discussion, I would "agree to disagree" and not bother them
>> with any more arguments.
>
> Cladistic classification is the natural classification. I.e., it tells
> how nature happens to arrange itself. NB: "the" natural system,
> singular. Furthermore, adopting it has caused no great problem for
> taxonomists. The pro side rests its case.
>
> However, I fully expect that I will not give your counterarguments a
> full hearing because I expect they will be overly verbose, mostly
> off-point, and filled with calumny against various and sundry people. If
> none of that applies, then I might pay attention to them.
>
>> Further than that I cannot go, because I am not a biologist and
>> so cannot enter the extra-logical aspects of a lifetime professional
>> biologist's world view. To me, as a mathematician who knows
>> the limits of quantification, the traditional Linnean
>> classification makes just too much sense to be cast to the
>> rubbish heap.
>
> Who has suggested junking the Linnean system?

A fair number of systematists. That's the PhyloCode's goal, for example.

[John Harshman]

Why is this tree a good thing, better than one that puts taxa (perhaps
species in that case) only at the tips?

>> (Aside from the fact that it is
>> not a secret.)
>
> I think it is at least as much a "trade secret" as the one of which
> Gould wrote.

It isn't a secret, but at least, unlike Gould's, we can be confident
that it's true.

>> I, for one, see nothing wrong with hypothetical things
>> being called hypothetical. Do you?
>
> It depends on what you mean by "hypothetical". Do you deny the fact
> that any two animals have an LCA? As far as we know, nobody has
> ever found a fossil of the species that was the LCA of humans
> and chimps, but only a creationist would deny that we DO have
> an LCA and, more generally, that chimps and ourselves have
> common ancestors going back beyond the first chordate.

That isn't what "hypothetical ancestor" means, as you well know.

> Harshman assured me time and again that cladistic systematists
> would never call a species an "ancestor candidate" even if
> every fossil of individuals of that species passed every test for
> falsifiability of its species being a common
> ancestor of, say, humans and chimps AND if one had a complete
> skeleton represented by those fossils. But they would be fools
> to claim flat-out that it is NOT a common ancestor under those
> conditions.

Agreed. We can't say it's an ancestor, and we can't say that it isn't.
We can say, usually, that it's unlikely to be an ancestor given the
ratio of known preserved species to those that must have existed.

> Look at it this way: had Gould written the above instead of
> what he actually wrote, I think creationists would be quote-mining
> that quote far more frequently than even is the case with
> his actual quote now.

I doubt that's true, but even if it were, so what?

> What would YOU say to a non-evolutionist who says, "I'm not
> claiming that common descent is false, only that it is a hypothesis;
> and as you can see, even cladists indirectly admit that the
> existence of common ancestors is just a hypothesis"?

I'd say "There is no such admission, and only a creationist would make
such a claim", except that you seem to be making such a claim.

[RSNorman]

Why do you ignore my quotation from T.H. Morgan who wrote in 1915
about the impossibility of identifying a particular fossil as an
ancestor? This has nothing whatsoever to do with cladistics. It is
just that the modern systemicist prefers to be honest and correct
rather than proclaim triumphantly "I found THE missing link!"

There are all sorts of things in science that are hypotheses. Do you
want to lie about science in order to confound the creationists? Most
definitely there really IS an last common ancestor. It is just that
we cannot put our finger on just who it was.

Incidentally, Gould had his own way of hyping his own ideas and
over-selling whatever proposal he wanted to get across. Gould is not
the be-all and end-all of evolutionary biology; he was a very
opinionated and highly partisan combatant.

[John Harshman]

On 11/23/15 6:26 PM, erik simpson wrote:

> So if systematists would just say 'candidate ancestor' instead of 'hypothetical
> ancestor', it'd be OK. Not that I think any would say either, since it's a
> silly thing to say. What in the world be "every test for falsifiability
> of [a species] being a common ancestor of, say, humans and chimps", with or
> without a complete skeleton?

The only tests I can think of are these:

1. Is it of the proper age? (Ambiguous, and requires that we be able to
define "species" in a way that determines membership over millions of
years).

2. Does it have any apomorphies? That is, is it identical in all
respects that we can tell to the reconstructed ancestral species? (Or,
if you do a phylogram, is it on a zero-length terminal branch?)
Unfortunately, both false positives and false negatives are possible.

Still, one ought to be able to falsify ancestry in a statistical sense.
The question is whether inability to falsify ancestry means anything we
should be interested in.

[D. Spencer Hines]

Euclid Romps...

Eschew Obfuscation.

DSH

Leo Tolstoy On Firmly Held Beliefs and Resultant Mental Gridlock

"I know that most men, not only those considered clever, but even those who
really are clever and capable of understanding the most difficult
scientific, mathematical or philosophic problems, can seldom discern even
the simplest and most obvious truth if it be such as obliges them to admit
the falsity of conclusions they have formed, perhaps with great difficulty -
conclusions of which they are proud, which they have taught to others, and
on which they have built their lives."

Count Lev Nikolayevich Tolstoy (1896) Source: "What Is Art?" - Leo Tolstoy,
Translated by Aylmer Maude, in Tolstoy's Collected Works, Charles Scribner's
Sons, (1902), vol. 19, p. 468

"erik simpson" wrote in message
news:23414391-5a46-495a...@googlegroups.com...

On Monday, November 23, 2015 at 10:13:54 AM UTC-8, Peter Nyikos wrote:

<...>

Science isn't like math. A proof by Euclid is as good as when he did it,
but observational sciences usually don't age as well. It's not that the
grey

eminences where [sic] so much wrong, but that we've learned more and better.

[erik simpson]

That's what I thought. Being able to say "A is a candidate ancestor to B
because we can't prove that it isn't" doesn't convey a lot of information.

[Peter Nyikos]

On Monday, November 23, 2015 at 9:28:54 PM UTC-5, erik simpson wrote:
> On Monday, November 23, 2015 at 6:03:55 PM UTC-8, Peter Nyikos wrote:
> > On Friday, November 20, 2015 at 8:34:04 PM UTC-5, Mark Isaak wrote:
> > > On 11/20/15 10:45 AM, Peter Nyikos wrote:
> > > > [...]
> >
> > You snipped the fact that the following is a paraphrasal of
> > a famous quotation by Stephen Jay Gould. I will return to
> > this aspect of it below.
> >
> > > > The complete lack of ancestor-descendent relationships in
> > > > phylogenetic systematics is the trade secret of cladists.
> > > > The trees that have their stamp of approval show species
> > > > at the tips of their branches, never at the nodes. All
> > > > the LCAs at the nodes are pure guesswork, and no names
> > > > are ever given to those LCAs, but the epithet "hypothetical"
> > > > is given them instead.
> > >
> > > Which is how things should be, right?
> >
> > You are only addressing the last sentence, whereas it is
> > what comes before that really sets the phylogenetic classification
> > apart from one that puts ancestor candidate species [a concept
> > explained below] at some nodes where such candidates exist,
> > and higher taxa where they don't.
> >
> > Kathleen Hunt's tree in the Horse FAQ puts genera at
> > the nodes:

<ASCII tree snipped; it looked good in this window, all cockeyed
in the main NGG display>

I prefer the word order "ancestor candidate" as being just a tad more
self-explanatory.

> instead of 'hypothetical
> ancestor', it'd be OK.

NO!!!! These are two totally different things. The LCA *was* (and, in
a few cases like the brown bear, *is*) an actual species of living,
breathing animals, whose status no one but a creationist would deny,
whereas an ancestor candidate is merely a CANDIDATE for that role.

Any undergrad biology major who isn't in it for the pre-med aspect
can READILY grasp the meaning of "ancestor candidate" once
it is explained, and never confuse it with "common ancestor."

It is pure cladophile ideology, and nothing else, that makes
Harshman so adamantly opposed to the "ancestor candidate" concept.
In contrast, he would be shooting himself (and all his fans)
in the foot if he were to start campaigning against the concept of
"LCA" as being a holdover from "species at the nodes" thinking.

> Not that I think any would say either, since it's a
> silly thing to say. What in the world be "every test for falsifiability
> of [a species] being a common ancestor of, say, humans and chimps", with or
> without a complete skeleton?

Looks like you have no idea what the whole concept of falsifiability
in biology is all about. Start by asking Harshman, who claims that the
gavial has passed the tests for falsifiability as far as it NOT
being basal to Crocodylia is concerned.

[Unlike so many people who say "Google is your friend" on such
occasions, I do not send people on scavenger hunts in the hope
that they will find just the right link.]

> > What would YOU say to a non-evolutionist who says, "I'm not
> > claiming that common descent is false, only that it is a hypothesis;
> > and as you can see, even cladists indirectly admit that the
> > existence of common ancestors is just a hypothesis"?
> >
>
> That's just silly. Cladists make no such statement.

You are pampered here, and are probably unaware of how the word
"admitted" is used against "outliers" like myself, in truly
*dishonest* ways, by a number of people whose antics you prefer
to ignore. [I don't fault you for ignoring them; your attitude
may be like Lincoln defending not sacking Grant over being
a drunkard: "I can't spare the man. He fights."]

> You could try to explain
> what the actual understanding is, or you could make something up.

Wrong personal pronoun. Harshman and Isaak are the ones who
need to make something up, because THEY are the ones who
champion the word "hypothetical" for LCA's.

And since you, Erik, seem to be on their side, why don't YOU
make something up? as the ancient Greeks said,

Here is Rhodes, now jump!

Peter Nyikos

Professor, Dept. of Mathematics -- standard disclaimer--
University of South Carolina

http://www.math.sc.edu/~nyikos/
nyikos @ math.sc.edu

[Peter Nyikos]

On Monday, November 23, 2015 at 10:43:55 PM UTC-5, erik simpson wrote:
> On Monday, November 23, 2015 at 7:13:55 PM UTC-8, John Harshman wrote:
> > On 11/23/15 6:26 PM, erik simpson wrote:
> >
> > > So if systematists would just say 'candidate ancestor' instead of 'hypothetical
> > > ancestor', it'd be OK. Not that I think any would say either, since it's a
> > > silly thing to say. What in the world be "every test for falsifiability
> > > of [a species] being a common ancestor of, say, humans and chimps", with or
> > > without a complete skeleton?
> >
> > The only tests I can think of are these:
> >
> > 1. Is it of the proper age? (Ambiguous, and requires that we be able to
> > define "species" in a way that determines membership over millions of
> > years).
> >
> > 2. Does it have any apomorphies? That is, is it identical in all
> > respects that we can tell to the reconstructed ancestral species? (Or,
> > if you do a phylogram, is it on a zero-length terminal branch?)
> > Unfortunately, both false positives and false negatives are possible.
> >
> > Still, one ought to be able to falsify ancestry in a statistical sense.

This piece of wishful thinking suggests that Harshman thinks of species
as disallowed paraphyletic taxa and insists on ancestry making sense only
with respect to isolated small populations. Out-Goulding Gould, so to speak.

Did you, Erik, rubber-stamp everything Harshman wrote with your
first comment below, or just his last sentence? Here it is already:

> > The question is whether inability to falsify ancestry means anything we
> > should be interested in.
>
> That's what I thought.

Really? Then why did you use the wording "What in the world...?" [See above.]
YOU, who asked me a few days ago on this very thread,
"Why all the drama?" after interpreting my "highly incomplete
tree of mammals" in the worst imaginable way?

> Being able to say "A is a candidate ancestor to B
> because we can't prove that it isn't" doesn't convey a lot of information.

This bizarre "paraphrasal" of my definition is just as irritating
as your worst imaginable misinterpretation of a tree that
was meant to illustrate a point which Norman, at least, seems
to have very belatedly grasped.

Would you have behaved very differently had I taken this whole
thread to sci.bio.paleontology as you once requested?

Peter Nyikos

[John Harshman]

On 11/24/15 5:36 AM, Peter Nyikos wrote:
> On Monday, November 23, 2015 at 10:43:55 PM UTC-5, erik simpson wrote:
>> On Monday, November 23, 2015 at 7:13:55 PM UTC-8, John Harshman wrote:
>>> On 11/23/15 6:26 PM, erik simpson wrote:
>>>
>>>> So if systematists would just say 'candidate ancestor' instead of 'hypothetical
>>>> ancestor', it'd be OK. Not that I think any would say either, since it's a
>>>> silly thing to say. What in the world be "every test for falsifiability
>>>> of [a species] being a common ancestor of, say, humans and chimps", with or
>>>> without a complete skeleton?
>>>
>>> The only tests I can think of are these:
>>>
>>> 1. Is it of the proper age? (Ambiguous, and requires that we be able to
>>> define "species" in a way that determines membership over millions of
>>> years).
>>>
>>> 2. Does it have any apomorphies? That is, is it identical in all
>>> respects that we can tell to the reconstructed ancestral species? (Or,
>>> if you do a phylogram, is it on a zero-length terminal branch?)
>>> Unfortunately, both false positives and false negatives are possible.
>>>
>>> Still, one ought to be able to falsify ancestry in a statistical sense.
>
> This piece of wishful thinking suggests that Harshman thinks of species
> as disallowed paraphyletic taxa and insists on ancestry making sense only
> with respect to isolated small populations. Out-Goulding Gould, so to speak.

It suggests no such thing, nor do I have any idea where you got that.

[Peter Nyikos]

Apply your "statistical sense" to some species for which we have
an essentially complete skeleton, like some species of *Mesohippus*.
If it passes all tests for falsifiability for ancestry to *Equus*
except a vague, unspecified "statistical test," how on earth
would your wishful thinking about it losing THAT test
actually come about, UNLESS you out-Goulded Gould and
out-Eldridged Eldridge in the way I specified?

Let's see a real alternative, because I for the life of me cannot
imagine one.

Peter Nyikos
Professor, Dept. of Math -- standard disclaimer--

[Peter Nyikos]

On Monday, November 23, 2015 at 9:48:53 PM UTC-5, RSNorman wrote:
> On Mon, 23 Nov 2015 18:01:33 -0800 (PST), Peter Nyikos
> <nyi...@bellsouth.net> wrote:
>
> >On Friday, November 20, 2015 at 8:34:04 PM UTC-5, Mark Isaak wrote:
> >> On 11/20/15 10:45 AM, Peter Nyikos wrote:
> >> > [...]
> >
> >You snipped the fact that the following is a paraphrasal of
> >a famous quotation by Stephen Jay Gould. I will return to
> >this aspect of it below.
> >
> >> > The complete lack of ancestor-descendent relationships in
> >> > phylogenetic systematics is the trade secret of cladists.
> >> > The trees that have their stamp of approval show species
> >> > at the tips of their branches, never at the nodes. All
> >> > the LCAs at the nodes are pure guesswork, and no names
> >> > are ever given to those LCAs, but the epithet "hypothetical"
> >> > is given them instead.
> >>
> >> Which is how things should be, right?
> >
> >You are only addressing the last sentence, whereas it is
> >what comes before that really sets the phylogenetic classification
> >apart from one that puts ancestor candidate species [a concept
> >explained below] at some nodes where such candidates exist,
> >and higher taxa where they don't.
> >
> >Kathleen Hunt's tree in the Horse FAQ puts genera at
> >the nodes:

<tree snipped as before, in reply to Erik>

Looks like you are "jumping in Rhodes" below, as I suggested
to Erik, but you botch the beginning so badly that I cannot
take your attempt seriously.

> Why do you ignore my quotation from T.H. Morgan who wrote in 1915
> about the impossibility of identifying a particular fossil as an
> ancestor?

Science isn't like mathematics, as Erik Simpson pointed out.

Why should I pay any attention to a quote which talks about "impossibility"
without telling the intended meaning of the word?

It is literally impossible to tell whether an
Ediacaran fossil is just a chance pattern in the rocks.

It is literally impossible to tell whether any *fossil*
of *Mesohippus* is the remains of an individual in direct
line of ancestry to all extant horses and asses.

On the other hand, it would be blatantly Harshman-serving to suggest
that ALL of the genera in the following "line" in Kathleen Hunt's FAQ
fail to be ancestral to ALL the ones that come after them.

Hyracotherium/Eohippus - Orohippus - Epihippus - Mesohippus -
Miohippus - Parahippus - Merychippus - Dinohippus - Plesippus - Equus.

Did T. H. Morgan *really* think that all the *real* ancestral
species of Equus left no fossils that we have found?

> This has nothing whatsoever to do with cladistics.

Nor does the insistence on banning all official classifications
except the cladistic. It is Harshman's package deal thinking
about the absolute victory of his kind in the "Cladist Wars"
that makes him cling for dear life to ideologies that have
nothing to do with cladistics.

Compared to that, Obama's "We Won" to silence Republican criticism
of an idea of Obama's was a mere prelude to an argument that evidently
never took place.

>It is
> just that the modern systemicist prefers to be honest and correct
> rather than proclaim triumphantly "I found THE missing link!"

How many knocked-down strawmen litter the landscape of your mind?

Remaining polemic deleted. Will address it if you insist.

Peter Nyikos

[erik simpson]

Peter, you've got to simmer down. This post is confused, and nearly incoherent.
I'm not going to reply to it, and wouldn't, no matter where you posted it.

[RSNorman]

On Tue, 24 Nov 2015 07:59:29 -0800 (PST), Peter Nyikos

Perhaps you failed to notice the text in Kathleen Hunt's tree that
"All the names on the tree are genus names, so recall that each genus
encompasses a cluster of closely related species."

Everybody knows that the ancestral "forms" on a tree represent
generalized "types" and not particular species. Even naming genera is
somewhat suspect but represents the closest information we have. That
is, something probably very similar to the fossils we have of
Epihippus were ancestral to the fossils we have of Mesohippus and
something very similar to those were ancestral to the fossils we have
of Miohippus.

What is wrong with that interpretation?

I treat the banning of all Linnean ranks separately in a different
post. That is Phylocode. If all you say is that you are opposed to
Phylocode then that is one thing. Many modern cladist systemacists
agree with you. But you never expressed that and your complaints
about cladism seem far broader. Those are what puzzle me.

[Peter Nyikos]

With the rapid-fire replies of Harshman coming thick and fast,
there are quite a number of posts to which I have not replied
and others that I've made only tiny dents in. IIRC the
one to which I am now replying belongs to the latter category.

On Wednesday, November 18, 2015 at 10:49:09 AM UTC-5, John Harshman wrote:
> On 11/18/15 6:57 AM, Peter Nyikos wrote:
> > On Wednesday, November 18, 2015 at 1:04:12 AM UTC-5, John Harshman wrote:
> >> On 11/17/15 6:42 PM, Peter Nyikos wrote:
> >>> On Tuesday, November 17, 2015 at 8:49:13 PM UTC-5, Peter Nyikos wrote:
> >>>> The long-running thread, "Did we come from monkeys?" has morphed into
> >>>> a comparison between the two systems of classification that have come
> >>>> down to us: the traditional ("Linnean") and the cladistic ("phylogenetic").
> >>>>
> >>>> I thought it a good idea to present, first, a "primer" on what these
> >>>> two systems are, and then to get down to the nitty-gritty of their
> >>>> largely complementary nature.
> >>>
> >>> The Linnean system's great strength lies in the way it classifies all
> >>> organisms, both extant and extinct, in a nested hierarchy starting
> >>> with the species, and then in ever-expanding circles we have its
> >>> genus, family, order, class, phylum, kingdom, and domain. [There are
> >>> intermediate ranks, like subfamily, superfamily, infraclass, subclass,
> >>> etc.] To keep things simple, I will confine myself mainly to animals
> >>> and indeed to vertebrates.
> >>
> >> So far, so good. Except that the Linnean system could be cladistic,
> >
> > Not without becoming a weak version of the phylogenetic.
>
> Again, the point of the Linnean system is ranks. Ranked taxa could be
> clades as easily as not.

A very ignorant comment. I challenged it a while back, and your reply to
the challenge was amusingly clueless.

But maybe that comment of yours was enough to shout down the opposition in
the Cladist Wars.

I hope to be able to find that post and show you today how clueless you were,
today, but family preparations for Thanksgiving severely limit my
time today, and Thanksgiving vacation begins for my family tomorrow.
So it may be only next Monday that I can get around to this.

> >> as I mentioned before, and that nested hierarchy is not a difference between
> >> systems.
> >
> > Its sheer existence is not, but the two take on radically different shapes.
> > See below where I quote what I wrote in reply to Richard Norman.
>
> You are confusing, again, ranked groups with attempts to represent
> overall similarity.

Stop dragging phenetics [sp?] into this discussion. Romer wouldn't
have given "overall similarity" the time of day.

> >> So that great strength isn't in fact a point of difference.
> >
> > That's like saying,
> >
> > the concept of color is not a point of difference between black and white,
> > so the great strength of white over black isn't in fact a point of
> > difference.
> >
> > That great strength, by the way, comes from the way white can
> > be resolved into a spectrum of colors.
>
> I don't see the analogy here.

Too bad. Catering to your mental laziness here will have to wait
until next week.

> >>> Each rank comes with a rough measure of disparity -- the amount
> >>> of variation allowed between members of the same taxon. Two living members
> >>> of the family *Hominidae*, for instance, are less disparate from each
> >>> other than the two most disparate members of the old Linnean class
> >>> *Amphibia* were from each other -- or even than the two most disparate
> >>> living amphibians are from each other.
> >>
> >> Now this just isn't true. There is no rough measure of disparity worth
> >> considering.
> >
> > Yes, there is. Romer himself used some rough and ready rules of thumb
> > when he wrote,
> >
> > [Birds] are divided into many orders; but the differences, for
> > example, between a humming bird and an albatross are much less
> > than the differences between a seal and a cat, or between a
> > stegosaurus and a duck-billed dinosaur, forms which are commonly
> > placed in a single order. The different [bird] orders have,
> > in general, no more differences between them than exist between
> > families in other classes of vertebrates, and, anatomically,
> > generic differences are so slight that fossils are hard to place.
> > --_Vertebrate Paleontology_, 1945 ed., pp. 264-5
>
> So Romer is in fact saying that the traditional system is a bad way of
> representing disparity.

Ah, but the very fact that he could make such comparisons with such
confidence showed that there is a way to reform, if reform were needed.

With ornithologist systematists having a whole CLASS to play around in,
it was human nature to take full advantage of the scope this gave them,
without consulting mammalian specialists on the proper limits of
the concept of "family", etc. That doesn't mean such consultation would
not have been fruitful, IF enough people on both sides really cared.

But why rock the boat? Why spend any time on this theme when it is
of dubious value where "publish or perish" is concerned?

> And you use this in your defense?

With this unreflective dig, you continue to exemplify what I wrote
about you being a polemicist first, a propagandist second, and
a reasoner a distant third.

> >>> These rankings are never better than ballpark estimates of disparity,
> >>> but they are reasonably good for gauging e.g. how close we are to the
> >>> true ancestral species of a clade.
> >>
> >> It isn't even clear what that would mean.
> >
> > Too bad Romer isn't alive so you can take up the issue with him. But
> > Carroll is still alive, and if you wish, I can run this comment by
> > Romer by him and see whether he agrees with it.
>
> Sure. Don't I recall that Carroll converted to cladistic classification?

Do you recall anything RELEVANT about Carroll, whom you blasted
as being "grossly mistaken" about ungulate evolution when it
suited your agenda to do so?

> > But hey, you specialize in neornithine birds like the hummingbird
> > and the albatross; perhaps you can actually *reason* against what
> > Romer wrote instead of falling back on your polemical tricks. You
> > know, the tricks you used when I cited John Hawks, hominin paleontology
> > specialist, in support of the statement that we should not,
> > under present day circumstances, say "humans are apes" in everyday
> > conversation.
>
> No, I don't recall any tricks.

The trick was to give every biologist an equal vote in the matter with
"So far, you have one biologist on your side" or words to that effect.
Your 'how many paleontologists do you know' is part of the same
ochlocratic thinking, isn't it?

If the practicing paleontologists had been
represented in the Cladist Wars the way even tiny nations are
represented in the U.N., would their SPECIAL insight into the unworkability
of every animal having its own genus, family, order, etc. in *cladistic*
taxa of 300 million years (or even 50 million years) ago have been given
a serious hearing?

It was a trick also because of the way you keep belittling people like
Carroll and Van Valen as though you were a far greater biologist
than either of them was.

> You should stop with the accusations.

You fully deserve them, and many more that I've refrained from making.

Remainder deleted, to be replied to later (only some time next week,
I fear).

Peter Nyikos
Professor, Dept. of Mathematics

[John Harshman]

I have no idea what your supposed explanation means. I think you have no
clear idea what my post meant either.

> Let's see a real alternative, because I for the life of me cannot
> imagine one.

Real alternative to what?

[John Harshman]

Do you agree that all that was just a series of empty characterizations,
unsupported by anything? If so, why even bother?

>>>> as I mentioned before, and that nested hierarchy is not a difference between
>>>> systems.
>>>
>>> Its sheer existence is not, but the two take on radically different shapes.
>>> See below where I quote what I wrote in reply to Richard Norman.
>>
>> You are confusing, again, ranked groups with attempts to represent
>> overall similarity.
>
> Stop dragging phenetics [sp?] into this discussion. Romer wouldn't
> have given "overall similarity" the time of day.

Really? It's merely the converse of "disparity".

>>>> So that great strength isn't in fact a point of difference.
>>>
>>> That's like saying,
>>>
>>> the concept of color is not a point of difference between black and white,
>>> so the great strength of white over black isn't in fact a point of
>>> difference.
>>>
>>> That great strength, by the way, comes from the way white can
>>> be resolved into a spectrum of colors.
>>
>> I don't see the analogy here.
>
> Too bad. Catering to your mental laziness here will have to wait
> until next week.

Once more, why even both with a "reply" like that?

What are the proper limits of the concept of "family"? Why are mammalian
systematists the guardians of these limits?

> But why rock the boat? Why spend any time on this theme when it is
> of dubious value where "publish or perish" is concerned?
>
>> And you use this in your defense?
>
> With this unreflective dig, you continue to exemplify what I wrote
> about you being a polemicist first, a propagandist second, and
> a reasoner a distant third.

Once again, an empty reply. "You're a bad person" is not an argument.

>>>>> These rankings are never better than ballpark estimates of disparity,
>>>>> but they are reasonably good for gauging e.g. how close we are to the
>>>>> true ancestral species of a clade.
>>>>
>>>> It isn't even clear what that would mean.
>>>
>>> Too bad Romer isn't alive so you can take up the issue with him. But
>>> Carroll is still alive, and if you wish, I can run this comment by
>>> Romer by him and see whether he agrees with it.
>>
>> Sure. Don't I recall that Carroll converted to cladistic classification?
>
> Do you recall anything RELEVANT about Carroll, whom you blasted
> as being "grossly mistaken" about ungulate evolution when it
> suited your agenda to do so?

He was grossly mistaken, wasn't he? He put all the ungulates together,
when modern phylogeny separates them. He separated artiodactyls from
cetaceans, when modern phylogeny puts the latter inside the former.

As for the cladistic classification bit, I think I'm recalling his later
book Patterns and Processes in Vertebrate Evolution. Have you seen it?

>>> But hey, you specialize in neornithine birds like the hummingbird
>>> and the albatross; perhaps you can actually *reason* against what
>>> Romer wrote instead of falling back on your polemical tricks. You
>>> know, the tricks you used when I cited John Hawks, hominin paleontology
>>> specialist, in support of the statement that we should not,
>>> under present day circumstances, say "humans are apes" in everyday
>>> conversation.
>>
>> No, I don't recall any tricks.
>
> The trick was to give every biologist an equal vote in the matter with
> "So far, you have one biologist on your side" or words to that effect.
> Your 'how many paleontologists do you know' is part of the same
> ochlocratic thinking, isn't it?

I never suggested giving every biologist an equal vote. I merely suggest
that your appeal to authority was cherry-picking the authority.

> If the practicing paleontologists had been
> represented in the Cladist Wars the way even tiny nations are
> represented in the U.N., would their SPECIAL insight into the unworkability
> of every animal having its own genus, family, order, etc. in *cladistic*
> taxa of 300 million years (or even 50 million years) ago have been given
> a serious hearing?

Who says they weren't represented? You are making up stories about
matters you know little about. For example, Joel Cracraft, who was
perhaps the earliest proponent of cladistics in ornithology, certainly
began as a paleontologist. E. O. Wiley, early cladist in ichthylogy and
influential proponent of Hennig, did a fair amount of paleontology. Etc.
I don't know that paleontologists actually have any special insight
anyway. Wiley did propose that extinct taxa be unranked and called
"plesions", if that makes you feel better.

> It was a trick also because of the way you keep belittling people like
> Carroll and Van Valen as though you were a far greater biologist
> than either of them was.

I make no comparisons or claims to greatness. I merely say they were
both wrong about some things. Linus Pauling was wrong about vitamin C
and the structure of DNA. So?

>> You should stop with the accusations.
>
> You fully deserve them, and many more that I've refrained from making.

Well, thanks for refraining to the extent you do.

[Mark Isaak]

Per Wikipedia, PhyloCode would only apply to clades. The Linnaean
system would still apply to naming species. Presumably, the higher
ranks would come unlabeled (i.e., nothing would be called a "class" or
"kingdom" any more), but Linnaeus himself was not all that big on
creating named ranks.

[Mark Isaak]

On 11/23/15 6:01 PM, Peter Nyikos wrote:

> On Friday, November 20, 2015 at 8:34:04 PM UTC-5, Mark Isaak wrote:
>> On 11/20/15 10:45 AM, Peter Nyikos wrote:
>>> [...]
>
> You snipped the fact that the following is a paraphrasal of
> a famous quotation by Stephen Jay Gould. I will return to
> this aspect of it below.

I left in the fact that the following is authored by Peter Nyikos, which
seemed much more relevant.

>>> The complete lack of ancestor-descendent relationships in
>>> phylogenetic systematics is the trade secret of cladists.
>>> The trees that have their stamp of approval show species
>>> at the tips of their branches, never at the nodes. All
>>> the LCAs at the nodes are pure guesswork, and no names
>>> are ever given to those LCAs, but the epithet "hypothetical"
>>> is given them instead.
>>
>> Which is how things should be, right?
>
> You are only addressing the last sentence, whereas it is
> what comes before that really sets the phylogenetic classification
> apart from one that puts ancestor candidate species [a concept
> explained below] at some nodes where such candidates exist,
> and higher taxa where they don't.

What comes before leads to the last sentence, and the last sentence is
the one most worth addressing.

Labeling a hypothetical "hypothetical" is right and proper, don't you think?

Do you deny the fact that any Last Common Ancestor from more than 10,000
years ago (and probably from more than 100 years ago) is going to be
hypothetical?

> As far as we know, nobody has
> ever found a fossil of the species that was the LCA of humans
> and chimps, but only a creationist would deny that we DO have
> an LCA and, more generally, that chimps and ourselves have
> common ancestors going back beyond the first chordate.
>
> Harshman assured me time and again that cladistic systematists
> would never call a species an "ancestor candidate" even if
> every fossil of individuals of that species passed every test for
> falsifiability of its species being a common
> ancestor of, say, humans and chimps AND if one had a complete
> skeleton represented by those fossils. But they would be fools
> to claim flat-out that it is NOT a common ancestor under those
> conditions.
>
> Look at it this way: had Gould written the above instead of
> what he actually wrote, I think creationists would be quote-mining
> that quote far more frequently than even is the case with
> his actual quote now.
>
> What would YOU say to a non-evolutionist who says, "I'm not
> claiming that common descent is false, only that it is a hypothesis;
> and as you can see, even cladists indirectly admit that the
> existence of common ancestors is just a hypothesis"?

I would say, first, that your non-evolutionist is hypothetical. So my
hypothetical response would be, "Get real. They do not. Your statement
is equivalent to saying that just because I don't know my
great-grandparents' names, maybe I did not have any."

You did not answer the question. Do you object to hypothetical
constructs being called hypothetical?

[John Harshman]

So I suppose it all depends on what you mean by "the Linnean system". If
you think it's just about naming species, presumably the binomial _Genus
species_, then some versions of the PhyloCode do away with that too. But
I would consider the Linnean system to be mostly about ranks, not just
about genus and species.