Man closer kin to naked mole rats than bats

[Daud Deden]

https://www-sciencealert-com.cdn.ampproject.org/v/s/www.sciencealert.com/new-dna-technology-is-shaking-up-the-branches-of-the-evolutionary-tree/amp?amp_gsa=1&amp_js_v=a9&usqp=mq331AQIKAGwASCAAgM%3D#amp_tf=From%20%251%24s&aoh=16564338853839&csi=0&referrer=https%3A%2F%2Fwww.google.com&ampshare=https%3A%2F%2Fwww.sciencealert.com%2Fnew-dna-technology-is-shaking-up-the-branches-of-the-evolutionary-tree

[John Harshman]

On 6/28/22 5:51 PM, Daud Deden wrote:
> https://www-sciencealert-com.cdn.ampproject.org/v/s/www.sciencealert.com/new-dna-technology-is-shaking-up-the-branches-of-the-evolutionary-tree/amp?amp_gsa=1&amp_js_v=a9&usqp=mq331AQIKAGwASCAAgM%3D#amp_tf=From%20%251%24s&aoh=16564338853839&csi=0&referrer=https%3A%2F%2Fwww.google.com&ampshare=https%3A%2F%2Fwww.sciencealert.com%2Fnew-dna-technology-is-shaking-up-the-branches-of-the-evolutionary-tree

Very old news. Euarchontoglires.

[Pandora]

Yet, man is closer kin to skinny wings than to naked mole rats.
Primates + Dermoptera = Primatomorpha:
https://academic.oup.com/gbe/article/9/9/2308/4095375

[John Harshman]

Couple of things:

1. Primatomorpha is of course part of Euarchontoglires and colugos as
primate relatives is also old news. To the extent that there's a
controversy, it's about the position of colugos within Euarchontoglires,
and that paper adopts what I would think of as the traditional resolution.

2. UCEs can be problematic to analyze. At least they are for birds, as
Prum et al. 2015 is in many ways an outlier vs. other large analyses.

http://animal-evolution.whu.edu.cn/PDF/2015nature.pdf

Junk DNA, e.g. introns, makes the best sequence data. Rare events, e.g.
retroelement insertions, are good too.

[Peter Nyikos]

Very old molecularly. The morphologically and molecularly produced phylogenetic trees of the various orders shown are very much at odds with each other, and it's obvious that you are a devotee of the molecular.

I've sometimes wondered whether the morphological trees take paleontology sufficiently into account, and whether the molecular ones take the insertion of viral DNA into mammalian genomes into account. The latter are a big nuisance because they have nothing to do with the ancestry of the animals in the orders and everything to do with what viruses are endemic to which land masses. Consider the fact that the indigenous people of the Americas had no immunity to smallpox, because there were no related viruses to confer it.

By the way, Cetacea is conspicuous by its absence. Paleontology would put it close to artiodactyla, though perhaps not as close as the molecular evidence does. The mesonychids are indicated as the sister group paleontologically, and their paraxonian digits argue for closeness with artiodactyls.


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

[Daud Deden]

On Wednesday, June 29, 2022 at 9:47:09 AM UTC-4, John Harshman wrote:

I figure primates are just long-limbed tree shrews. KISS.

[John Harshman]

On 6/29/22 5:38 PM, Peter Nyikos wrote:
> On Wednesday, June 29, 2022 at 9:47:09 AM UTC-4, John Harshman wrote:
>> On 6/28/22 5:51 PM, Daud Deden wrote:
>>> https://www-sciencealert-com.cdn.ampproject.org/v/s/www.sciencealert.com/new-dna-technology-is-shaking-up-the-branches-of-the-evolutionary-tree/amp?amp_gsa=1&amp_js_v=a9&usqp=mq331AQIKAGwASCAAgM%3D#amp_tf=From%20%251%24s&aoh=16564338853839&csi=0&referrer=https%3A%2F%2Fwww.google.com&ampshare=https%3A%2F%2Fwww.sciencealert.com%2Fnew-dna-technology-is-shaking-up-the-branches-of-the-evolutionary-tree
>>
>> Very old news. Euarchontoglires.
>
> Very old molecularly. The morphologically and molecularly produced phylogenetic trees of the various orders shown are very much at odds with each other, and it's obvious that you are a devotee of the molecular.

Yes. Molecular data generally work better. That should be obvious based
on the last 30 years or so. The shear available volume of data and the
comparative ease of character coding are chiefly responsible.

> I've sometimes wondered whether the morphological trees take paleontology sufficiently into account, and whether the molecular ones take the insertion of viral DNA into mammalian genomes into account. The latter are a big nuisance because they have nothing to do with the ancestry of the animals in the orders and everything to do with what viruses are endemic to which land masses. Consider the fact that the indigenous people of the Americas had no immunity to smallpox, because there were no related viruses to confer it.

All very nice, but that has nothing to do with the validity of
phylogenetic analyses. You can tell whether viral insertions are
homologous based on their points of insertion, i.e. the nature of their
flanking sequences. Viral insertions are really not a big problem for
phylogenetics.

And many morphological trees do take paleontology into account. It
helps, but not enough to make up for the essential lack of data and
problems with subjective coding.

> By the way, Cetacea is conspicuous by its absence. Paleontology would put it close to artiodactyla, though perhaps not as close as the molecular evidence does. The mesonychids are indicated as the sister group paleontologically, and their paraxonian digits argue for closeness with artiodactyls.

Cetacea as an order doesn't exist. Whales are artiodactyls.
Andrewsarchus, interestingly, is probably not a mesonychid but an
artiodactyl close to whales.

[Glenn]

Yea, yea, yea.

"Molecular phylogenetics has also been used to estimate the dates when species diverged, but there is controversy about the reliability of the molecular clock on which such estimates depend. "

Read the rest at

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

[Peter Nyikos]

On Wednesday, June 29, 2022 at 10:02:40 AM UTC-4, Pandora wrote:
> On Wed, 29 Jun 2022 06:47:02 -0700, John Harshman
> <jhar...@pacbell.net> wrote:
>
> >On 6/28/22 5:51 PM, Daud Deden wrote:
> >> https://www-sciencealert-com.cdn.ampproject.org/v/s/www.sciencealert.com/new-dna-technology-is-shaking-up-the-branches-of-the-evolutionary-tree/amp?amp_gsa=1&amp_js_v=a9&usqp=mq331AQIKAGwASCAAgM%3D#amp_tf=From%20%251%24s&aoh=16564338853839&csi=0&referrer=https%3A%2F%2Fwww.google.com&ampshare=https%3A%2F%2Fwww.sciencealert.com%2Fnew-dna-technology-is-shaking-up-the-branches-of-the-evolutionary-tree



> >Very old news. Euarchontoglires.
> Yet, man is closer kin to skinny wings than to naked mole rats.

Not according to the article Daud linked; see below.

> Primates + Dermoptera = Primatomorpha:
> https://academic.oup.com/gbe/article/9/9/2308/4095375

That's not the way it is in the article Daud linked, which groups Dermoptera
with Scandentia, then with Glires (Rodentia+Lagomorpha), and only then with Primates.

By the way, I tracked down the original article:

https://theconversation.com/evolutionary-tree-of-life-modern-science-is-showing-how-we-got-so-much-wrong-185077

I got involved in the comments section with the author, who is remarkably good at replying to reader comments
and questions. If you have a preference for the article you linked, but don't want to get into the comments
section yourself, I could pass on anything you'd like for me to pass on.


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

PS I like your literal translation of "Dermoptera", although "skinny wings" could also refer to the
patagia being "too skinny" for true flight. Better than calling them flying lemurs, but I'm
afraid we are stuck with that term.

[Peter Nyikos]

Harshman is here ignoring the evidence that mesonychids were also artiodactyls, some of which I gave him (see above)
though not yet giving the same conclusion.

I'll give him more of it today.

> Yea, yea, yea.
>
> "Molecular phylogenetics has also been used to estimate the dates when species diverged, but there is controversy about the reliability of the molecular clock on which such estimates depend. "

Very true, especially where there are no fossils to give good estimates of the divergence times using geology.

This is a major difficulty; in every case of which I have read, the molecular clock gives earlier
divergence dates than the fossil evidence in conjunction with geological dating.

Some other difficulties are mentioned here:

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

I often wonder whether there is more than one overall molecular clock in common use.
By "overall" I mean that various mutation rates of various animals are used in the calibration.
Maybe Harshman can tell us.

> Read the rest at
>
> https://en.wikipedia.org/wiki/Paleontology

Molecular phylogenetics, by itself, is reasonably reliable. But there are problems with it,
of which I'll be reminding Harshman this week. So far, though, the disagreements are not of major importance.

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

Univ. of South Carolina in Columbia
http://people.math.sc.edu/nyikos

[John Harshman]

On 7/5/22 6:26 AM, Peter Nyikos wrote:
> On Friday, July 1, 2022 at 5:34:05 PM UTC-4, Glenn wrote:
>> On Wednesday, June 29, 2022 at 8:20:49 PM UTC-7, John Harshman wrote:
>>> On 6/29/22 5:38 PM, Peter Nyikos wrote:
>
>>>> By the way, Cetacea is conspicuous by its absence. Paleontology would put it close to artiodactyla, though perhaps not as close as the molecular evidence does. The mesonychids are indicated as the sister group paleontologically, and their paraxonian digits argue for closeness with artiodactyls.
>
>>> Cetacea as an order doesn't exist. Whales are artiodactyls.
>>> Andrewsarchus, interestingly, is probably not a mesonychid but an
>>> artiodactyl close to whales.
>
> Harshman is here ignoring the evidence that mesonychids were also artiodactyls, some of which I gave him (see above)
> though not yet giving the same conclusion.

"Close to artiodactyls" and "were also artiodactyls" are not the same
thing. If you want to present evidence that mesonychids, like whales,
are nested within Artiodactyla, I would be interested. I don't recall
ever seeing an analysis that did that. Except for Andrewsarchus, that is.

> I'll give him more of it today.
>
>
>> Yea, yea, yea.
>>
>> "Molecular phylogenetics has also been used to estimate the dates when species diverged, but there is controversy about the reliability of the molecular clock on which such estimates depend. "
>
> Very true, especially where there are no fossils to give good estimates of the divergence times using geology.

Why do you think Glenn quoted this? My hypothesis is that he just found
the first thing he could locate that seemed at first glance to cast
doubt on molecular phylogeny

> This is a major difficulty; in every case of which I have read, the molecular clock gives earlier
> divergence dates than the fossil evidence in conjunction with geological dating.

Well, of course it ought to, even with perfect accuracy. It's unlikely
that the first known fossil of x is really the first x, and the average
gene will also have diverged some time before the divergence of its
containing species. One may hope to narrow the gap, but it should not be
zero.

> Some other difficulties are mentioned here:
>
> https://en.wikipedia.org/wiki/Molecular_clock
>
> I often wonder whether there is more than one overall molecular clock in common use.
> By "overall" I mean that various mutation rates of various animals are used in the calibration.
> Maybe Harshman can tell us.

These days, time calibrations don't generally use a molecular clock.
They use maximum likelihood models that allow the rate of evolution to
vary among taxa, with some parameter limiting the rate of change in
rate. The more fossil calibration points, the better. But of course,
strictly speaking, a fossil can only provide a lower limit on the age of
a node. Generally, some distribution of allowed ages is also used in the
model. A common program used for such things is called BEAST, which you
might look up if you're interested.

[Peter Nyikos]

On Tuesday, July 5, 2022 at 11:54:14 AM UTC-4, John Harshman wrote:
> On 7/5/22 6:26 AM, Peter Nyikos wrote:
> > On Friday, July 1, 2022 at 5:34:05 PM UTC-4, Glenn wrote:
> >> On Wednesday, June 29, 2022 at 8:20:49 PM UTC-7, John Harshman wrote:
> >>> On 6/29/22 5:38 PM, Peter Nyikos wrote:
> >
> >>>> By the way, Cetacea is conspicuous by its absence. Paleontology would put it close to artiodactyla, though perhaps not as close as the molecular evidence does. The mesonychids are indicated as the sister group paleontologically, and their paraxonian digits argue for closeness with artiodactyls.

I didn't want to stir things up too much at the beginning, so I talked in terms of "sister group" rather
than "ancestor, as in "mesonychids may be ancestral to whales."

> >>> Cetacea as an order doesn't exist. Whales are artiodactyls.
> >>> Andrewsarchus, interestingly, is probably not a mesonychid but an
> >>> artiodactyl close to whales.
> >
> > Harshman is here ignoring the evidence that mesonychids were also artiodactyls, some of which I gave him (see above)
> > though not yet giving the same conclusion.

> "Close to artiodactyls" and "were also artiodactyls" are not the same thing.

I take it you wrote this for Glenn's benefit. Or didn't you understand what
"though not yet giving the same conclusion" was all about?

>If you want to present evidence that mesonychids, like whales,
> are nested within Artiodactyla, I would be interested.

In my 1945 second edition of Romer's _Vertebrate_Paleontology_, p. 367, Fig. 277, on the left, there are
detailed drawings of the manus and pes of *Synoplotherium,* a mesonychid. Its astragalus looks
quite a lot like that of the Entelodont *Dinohyus,* shown on p. 452, Fig. 342 and that of the Camelid
*Poebrotherium,* p. 461, Fig. 348. All three astraguli look double-pulleyed to me.

Isn't the lack of a double-pulleyed astragalus the item that convinced most vertebrate paleontologists
to eliminate mesonychids from cetacean ancestry? AFAIK we have never found any mesonychid DNA,
so disqualification would have to use fossil morphology.

I know you dislike the language of the preceding paragraph, so I put it to you this way:
what else, if anything, disqualifies mesonychids from being artiodactyls?


Even if I am wrong in the interpretation of the astragalus of *Synoplotherium,* I have to ask:
how difficult would it have been for a double pulleyed astragalus to emerge independently in Mesonychidae?

A look at other illustrations in Romer's _Vertebrate_Paleontology_ gives me the impression that
a double-pulleyed astragalus is not all that hard to come by. Perissodactyls (p. 433, Fig. 329),
litopterns (p.402, Fig. 305) and some notoungulates (p. 397, Fig. 299) had what look to me like
double-pulleyed astragali.

> I don't recall
> ever seeing an analysis that did that. Except for Andrewsarchus, that is.

Did it have a double-pulleyed astragalus?

> > I'll give him more of it today.

Here it is above, two days late. Some very intense controversy in other forums
made me run out of time for it for both days.


At this point, the topic of your post completely changed, and I am postponing my comments on the rest
until tomorrow.

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

Univ. of South Carolina -- standard disclaimer--
http://people.math.sc.edu/nyikos

[John Harshman]

On 7/7/22 6:51 PM, Peter Nyikos wrote:
> On Tuesday, July 5, 2022 at 11:54:14 AM UTC-4, John Harshman wrote:
>> On 7/5/22 6:26 AM, Peter Nyikos wrote:
>>> On Friday, July 1, 2022 at 5:34:05 PM UTC-4, Glenn wrote:
>>>> On Wednesday, June 29, 2022 at 8:20:49 PM UTC-7, John Harshman wrote:
>>>>> On 6/29/22 5:38 PM, Peter Nyikos wrote:
>>>
>>>>>> By the way, Cetacea is conspicuous by its absence. Paleontology would put it close to artiodactyla, though perhaps not as close as the molecular evidence does. The mesonychids are indicated as the sister group paleontologically, and their paraxonian digits argue for closeness with artiodactyls.
>
> I didn't want to stir things up too much at the beginning, so I talked in terms of "sister group" rather
> than "ancestor, as in "mesonychids may be ancestral to whales."

I didn't notice that bit. Paleontology in fact puts Cetacea within
Artiodactyla, not adjacent, since Thewissen's and Gingrich's independent
papers in 2001. Mesonychids, on the other hand, have been outside
Artiodactyla in all analyses I'm aware of.

>>>>> Cetacea as an order doesn't exist. Whales are artiodactyls.
>>>>> Andrewsarchus, interestingly, is probably not a mesonychid but an
>>>>> artiodactyl close to whales.
>>>
>>> Harshman is here ignoring the evidence that mesonychids were also artiodactyls, some of which I gave him (see above)
>>> though not yet giving the same conclusion.
>
>> "Close to artiodactyls" and "were also artiodactyls" are not the same thing.
>
> I take it you wrote this for Glenn's benefit. Or didn't you understand what
> "though not yet giving the same conclusion" was all about?

It was in fact unclear what that meant. Still wondering what the
evidence is that mesonychids are artiodactyls.

>> If you want to present evidence that mesonychids, like whales,
>> are nested within Artiodactyla, I would be interested.
>
> In my 1945 second edition of Romer's _Vertebrate_Paleontology_, p. 367, Fig. 277, on the left, there are
> detailed drawings of the manus and pes of *Synoplotherium,* a mesonychid. Its astragalus looks
> quite a lot like that of the Entelodont *Dinohyus,* shown on p. 452, Fig. 342 and that of the Camelid
> *Poebrotherium,* p. 461, Fig. 348. All three astraguli look double-pulleyed to me.

I'm suspicious of this conclusion, which has never been mentioned in any
phylogenetic analysis I know about. Either you are wrong about this or
mammal paleontologists who have done the work are all incompetent.

> Isn't the lack of a double-pulleyed astragalus the item that convinced most vertebrate paleontologists
> to eliminate mesonychids from cetacean ancestry? AFAIK we have never found any mesonychid DNA,
> so disqualification would have to use fossil morphology.

It's not the lack of a double-pulleyed astragalus alone, but the
combined verdict of the many characters used in the phylogenetic
analyses of the past 30 or so years. You could look up some of those and
see how characters of mesonychid astragali (and any other relevant
characters) are scored if you want to know how that happened.

> I know you dislike the language of the preceding paragraph, so I put it to you this way:
> what else, if anything, disqualifies mesonychids from being artiodactyls?

I don't know. I would have to look up the previously mentioned
literature, which I have not done for quite a few years.

> Even if I am wrong in the interpretation of the astragalus of *Synoplotherium,* I have to ask:
> how difficult would it have been for a double pulleyed astragalus to emerge independently in Mesonychidae?

There is no way of answering that question, as far as I can tell. But of
course phylogenetics should not be based on single characters. It's the
agreement among many independent characters that gives us confidence in
any phylogeny.

> A look at other illustrations in Romer's _Vertebrate_Paleontology_ gives me the impression that
> a double-pulleyed astragalus is not all that hard to come by. Perissodactyls (p. 433, Fig. 329),
> litopterns (p.402, Fig. 305) and some notoungulates (p. 397, Fig. 299) had what look to me like
> double-pulleyed astragali.

Does Romer mention the term? It's conceivable that you are unclear on
the definition. If so, that might explain why paleontologists make so
much of the character and do generally find it diagnostic of Artiodactyla.

> > I don't recall
>> ever seeing an analysis that did that. Except for Andrewsarchus, that is.
>
> Did it have a double-pulleyed astragalus?

I don't believe its astragalus is known. It's other characters that put
it within artiodactyls (and away from mesonychids).

>>> I'll give him more of it today.
>
> Here it is above, two days late. Some very intense controversy in other forums
> made me run out of time for it for both days.

https://xkcd.com/386/

[Peter Nyikos]

This is my second and final reply to this post by John:

On Tuesday, July 5, 2022 at 11:54:14 AM UTC-4, John Harshman wrote:
> On 7/5/22 6:26 AM, Peter Nyikos wrote:
> > On Friday, July 1, 2022 at 5:34:05 PM UTC-4, Glenn wrote:

> >> Yea, yea, yea.
> >>
> >> "Molecular phylogenetics has also been used to estimate the dates when species diverged, but there is controversy about the reliability of the molecular clock on which such estimates depend. "
> >
> > Very true, especially where there are no fossils to give good estimates of the divergence times using geology.

I should have made it clear that I was only referring to the clause beginning with "but there is controversy..."

> Why do you think Glenn quoted this? My hypothesis is that he just found
> the first thing he could locate that seemed at first glance to cast

> doubt on molecular phylogeny.

You make lots of "hypotheses" about Glenn that you are unable to support. This has been amply demonstrated on the thread,
Re: Questions about BBC’s “Prehistoric Planet” Episode 1.

> > This is a major difficulty; in every case of which I have read, the molecular clock gives earlier
> > divergence dates than the fossil evidence in conjunction with geological dating.

> Well, of course it ought to, even with perfect accuracy. It's unlikely
> that the first known fossil of x is really the first x, and the average
> gene will also have diverged some time before the divergence of its
> containing species.

The main problem is not with species, but with sizable clades.
One example I read about not long ago (sorry, I can't remember where)
is that divergence of metatherians (marsupials and stem marsupials)
and eutherians (placentals and stem placentals) took place later than
the molecular clock indicates. The similarity of the representatives
of the two clades at one point in time is so close that some fossils
from that time have first been assigned to one group, and then to the other.
This point in time is well after the divergence time given by molecular clocks, IIRC.

> > Some other difficulties are mentioned here:
> >
> > https://en.wikipedia.org/wiki/Molecular_clock
> >
> > I often wonder whether there is more than one overall molecular clock in common use.
> > By "overall" I mean that various mutation rates of various animals are used in the calibration.
> > Maybe Harshman can tell us.

You did not address the above comment, but at least you indicated below how
the sentence that Glenn quoted may be outdated.

> These days, time calibrations don't generally use a molecular clock.
> They use maximum likelihood models that allow the rate of evolution to
> vary among taxa, with some parameter limiting the rate of change in
> rate.

How does ML help with that? The only place where the webpage that I linked mentions ML
is very vague about how it is used and how useful it is:

"Molecular clock users have developed workaround solutions using a number of statistical approaches including maximum likelihood techniques and later Bayesian modeling. In particular, models that take into account rate variation across lineages have been proposed in order to obtain better estimates of divergence times. These models are called relaxed molecular clocks[33] because they represent an intermediate position between the 'strict' molecular clock hypothesis and Joseph Felsenstein's many-rates model[34] and are made possible through MCMC techniques that explore a weighted range of tree topologies and simultaneously estimate parameters of the chosen substitution model."

> The more fossil calibration points, the better. But of course,
> strictly speaking, a fossil can only provide a lower limit on the age of
> a node. Generally, some distribution of allowed ages is also used in the
> model. A common program used for such things is called BEAST, which you
> might look up if you're interested.

Unlike you and "Mickey Mortimer," I don't know the first thing about computer programming.
When I needed something to be programmed when I was a "first Louie" in the Army, I gave
it to a bright high school student who worked for us in his spare time.

> >> Read the rest at
> >>
> >> https://en.wikipedia.org/wiki/Paleontology
> >
> > Molecular phylogenetics, by itself, is reasonably reliable. But there are problems with it,
> > of which I'll be reminding Harshman this week. So far, though, the disagreements are not of major importance.

I'm reminding you now of the problem of analyzing relationships within Euarchontoglires.
In the July 1 reply I did to Pandora on this thread, I detailed the way the 2022 popularization, linked by Daud in the OP,
clashes with the one in the 2017 research paper that Pandora linked:

https://academic.oup.com/gbe/article/9/9/2308/4095375?login=false

In particular, it supported the clade Primatomorpha, which the paper Daud linked did not.
It also detailed a lot of other difficulties with the molecular phylogenetics of various orders,
which you might need to look at carefully.

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

University of South Carolina
http://people.math.sc.edu/nyikos

PS You haven't replied to either Pandora's post or my reply to it yet.

[Glenn]

You only need to know that computers are programmed. Say it out loud.

ML is "guessing", but believers won't accept that.

"We now under-
stand the limitations of maximum likelihood better
than Fisher did, but far from well enough to guar-
antee safety in its application in complex situations
where it is most needed. Maximum likelihood re-
mains a truly beautiful theory, even though tragedy
may lurk around a corner."

https://arxiv.org/pdf/0804.2996.pdf

[John Harshman]

On 7/8/22 10:12 AM, Peter Nyikos wrote:
> This is my second and final reply to this post by John:
>
> On Tuesday, July 5, 2022 at 11:54:14 AM UTC-4, John Harshman wrote:
>> On 7/5/22 6:26 AM, Peter Nyikos wrote:
>>> On Friday, July 1, 2022 at 5:34:05 PM UTC-4, Glenn wrote:
>
>>>> Yea, yea, yea.
>>>>
>>>> "Molecular phylogenetics has also been used to estimate the dates when species diverged, but there is controversy about the reliability of the molecular clock on which such estimates depend. "
>>>
>>> Very true, especially where there are no fossils to give good estimates of the divergence times using geology.
>
> I should have made it clear that I was only referring to the clause beginning with "but there is controversy..."
>
>> Why do you think Glenn quoted this? My hypothesis is that he just found
>> the first thing he could locate that seemed at first glance to cast
>> doubt on molecular phylogeny.
>
> You make lots of "hypotheses" about Glenn that you are unable to support. This has been amply demonstrated on the thread,
> Re: Questions about BBC’s “Prehistoric Planet” Episode 1.

It's hard to support any hypotheses about Glenn's motivations or point,
since he never says anything that would help. One must try to
reconstruct a general picture from what he gives us. Admittedlly this is
speculative. But do you think he had a point? Do you have no hypothesis?

>>> This is a major difficulty; in every case of which I have read, the molecular clock gives earlier
>>> divergence dates than the fossil evidence in conjunction with geological dating.
>
>> Well, of course it ought to, even with perfect accuracy. It's unlikely
>> that the first known fossil of x is really the first x, and the average
>> gene will also have diverged some time before the divergence of its
>> containing species.
>
> The main problem is not with species, but with sizable clades.

That's the same problem, since clades are represented by species. The
divergence of metatherians and eutherians can only be determines by
finding fossils belonging to one of the groups or by sampling the
genomes of species on those groups.

> One example I read about not long ago (sorry, I can't remember where)
> is that divergence of metatherians (marsupials and stem marsupials)
> and eutherians (placentals and stem placentals) took place later than
> the molecular clock indicates. The similarity of the representatives
> of the two clades at one point in time is so close that some fossils
> from that time have first been assigned to one group, and then to the other.
> This point in time is well after the divergence time given by molecular clocks, IIRC.

That assumes there is a morphological clock, such that a small
morphological divergence implies a short time since divergence. That's a
problematic assumption.

>>> Some other difficulties are mentioned here:
>>>
>>> https://en.wikipedia.org/wiki/Molecular_clock
>>>
>>> I often wonder whether there is more than one overall molecular clock in common use.
>>> By "overall" I mean that various mutation rates of various animals are used in the calibration.
>>> Maybe Harshman can tell us.
>
> You did not address the above comment, but at least you indicated below how
> the sentence that Glenn quoted may be outdated.

The answer should be obvious from what I said, though. Don't you think?
Molecular clocks are not in common use.

>> These days, time calibrations don't generally use a molecular clock.
>> They use maximum likelihood models that allow the rate of evolution to
>> vary among taxa, with some parameter limiting the rate of change in
>> rate.
>
> How does ML help with that? The only place where the webpage that I linked mentions ML
> is very vague about how it is used and how useful it is:

I suggest looking up BEAST and examining the models it uses. The models
make the data more likely given some combination of rates on various
branches than given other combinations of rates.

> "Molecular clock users have developed workaround solutions using a number of statistical approaches including maximum likelihood techniques and later Bayesian modeling. In particular, models that take into account rate variation across lineages have been proposed in order to obtain better estimates of divergence times. These models are called relaxed molecular clocks[33] because they represent an intermediate position between the 'strict' molecular clock hypothesis and Joseph Felsenstein's many-rates model[34] and are made possible through MCMC techniques that explore a weighted range of tree topologies and simultaneously estimate parameters of the chosen substitution model."
>
>
>> The more fossil calibration points, the better. But of course,
>> strictly speaking, a fossil can only provide a lower limit on the age of
>> a node. Generally, some distribution of allowed ages is also used in the
>> model. A common program used for such things is called BEAST, which you
>> might look up if you're interested.
>
> Unlike you and "Mickey Mortimer," I don't know the first thing about computer programming.
> When I needed something to be programmed when I was a "first Louie" in the Army, I gave
> it to a bright high school student who worked for us in his spare time.

You should be able to understand the models used by the algorithms, and
that should be enough. Programming skill not needed.

>>>> Read the rest at
>>>>
>>>> https://en.wikipedia.org/wiki/Paleontology
>>>
>>> Molecular phylogenetics, by itself, is reasonably reliable. But there are problems with it,
>>> of which I'll be reminding Harshman this week. So far, though, the disagreements are not of major importance.
>
> I'm reminding you now of the problem of analyzing relationships within Euarchontoglires.
> In the July 1 reply I did to Pandora on this thread, I detailed the way the 2022 popularization, linked by Daud in the OP,
> clashes with the one in the 2017 research paper that Pandora linked:
>
> https://academic.oup.com/gbe/article/9/9/2308/4095375?login=false
>
> In particular, it supported the clade Primatomorpha, which the paper Daud linked did not.
> It also detailed a lot of other difficulties with the molecular phylogenetics of various orders,
> which you might need to look at carefully.

Of course some questions are more difficult to answer than others. I
don't see this as an example of a problem with molecular systematics,
per se. The questions are not, almost always, less difficult for
morphological systematics.

> PS You haven't replied to either Pandora's post or my reply to it yet.

Maybe there was nothing worth a reply? I'll look again.

[Peter Nyikos]

On Thursday, July 7, 2022 at 10:11:54 PM UTC-4, John Harshman wrote:
> On 7/7/22 6:51 PM, Peter Nyikos wrote:
> > On Tuesday, July 5, 2022 at 11:54:14 AM UTC-4, John Harshman wrote:
> >> On 7/5/22 6:26 AM, Peter Nyikos wrote:
> >>> On Friday, July 1, 2022 at 5:34:05 PM UTC-4, Glenn wrote:
> >>>> On Wednesday, June 29, 2022 at 8:20:49 PM UTC-7, John Harshman wrote:
> >>>>> On 6/29/22 5:38 PM, Peter Nyikos wrote:
> >>>
> >>>>>> By the way, Cetacea is conspicuous by its absence. Paleontology would put it close to artiodactyla, though perhaps not as close as the molecular evidence does. The mesonychids are indicated as the sister group paleontologically, and their paraxonian digits argue for closeness with artiodactyls.
> >
> > I didn't want to stir things up too much at the beginning, so I talked in terms of "sister group" rather
> > than "ancestor, as in "mesonychids may be ancestral to whales."

> I didn't notice that bit. Paleontology in fact puts Cetacea within
> Artiodactyla, not adjacent, since Thewissen's and Gingrich's independent
> papers in 2001. Mesonychids, on the other hand, have been outside
> Artiodactyla in all analyses I'm aware of.

Why didn't you reference a single one of them?

> >>>>> Cetacea as an order doesn't exist. Whales are artiodactyls.
> >>>>> Andrewsarchus, interestingly, is probably not a mesonychid but an
> >>>>> artiodactyl close to whales.
> >>>
> >>> Harshman is here ignoring the evidence that mesonychids were also artiodactyls, some of which I gave him (see above)
> >>> though not yet giving the same conclusion.
> >
> >> "Close to artiodactyls" and "were also artiodactyls" are not the same thing.
> >
> > I take it you wrote this for Glenn's benefit. Or didn't you understand what
> > "though not yet giving the same conclusion" was all about?

> It was in fact unclear what that meant.

I was very clear, you just didn't pay close attention when reading it.

>Still wondering what the
> evidence is that mesonychids are artiodactyls.

evidence = overwhelming evidence?

> >> If you want to present evidence that mesonychids, like whales,
> >> are nested within Artiodactyla, I would be interested.
> >
> > In my 1945 second edition of Romer's _Vertebrate_Paleontology_, p. 367, Fig. 277, on the left, there are
> > detailed drawings of the manus and pes of *Synoplotherium,* a mesonychid. Its astragalus looks
> > quite a lot like that of the Entelodont *Dinohyus,* shown on p. 452, Fig. 342 and that of the Camelid
> > *Poebrotherium,* p. 461, Fig. 348. All three astraguli look double-pulleyed to me.

> I'm suspicious of this conclusion, which has never been mentioned in any
> phylogenetic analysis I know about.

What kept you from opening your 1966 edition and looking up the drawings in there?
The index should have made it child's play.

Don't you think *you* can recognize a double-pulleyed astragalus?

> Either you are wrong about this or
> mammal paleontologists who have done the work are all incompetent.

If these are the only alternatives you could think of, then you must be suffering from
a long lack of meaningful discussions with research biologists *qua* research biologists.
I thought of two others right off the bat:

1. *Synoplotherium* was found to have been mis-classified after Romer's 1945 edition.

2. The reconstruction to which I referred was inaccurate. [The reconstruction of the skull
of Archaeopteryx in the same edition was criticized in a peer-reviewed paper comparing
it with modern reconstructions of the Eichstatt specimen and the London specimen.]

Later, while investigating 1. and finding it was incorrect, I discovered what is
almost surely the real reason:

3. The most complete skeleton of *Synoplotherium* was never subjected to a modern character analysis,
and so, other mesonychids were used in all the phylogenetic analyses that placed them outside Artiodactyla.

I conclude this from a 2016 account by Riley Black in _Scientific American_, long after the all
the analyses of mesonychid affinities of which I know.

https://blogs.scientificamerican.com/laelaps/yale-s-mysterious-fossil-beast/
Subtitle: Despite being discovered over a century ago, *Synoplotherium* is still an enigma
The crucial excerpt:
"The first and last major source of information on Synoplotherium is a section of a massive review of Eocene mammals laid out by paleontologist Jacob Wortman in 1902. He covered the whole animal from its oversized skull to what remained of its tail, detecting a few surprises along the way."

It's a fascinating and very readable article by a prolific contributor to SA, but for the sake of brevity I
mention only one more detail: the almost complete skeleton in the Peabody Museum of Yale, which was the
focus of the article, is displayed to the public, despite the accepted practice of only displaying casts.

> > Isn't the lack of a double-pulleyed astragalus the item that convinced most vertebrate paleontologists
> > to eliminate mesonychids from cetacean ancestry? AFAIK we have never found any mesonychid DNA,
> > so disqualification would have to use fossil morphology.

> It's not the lack of a double-pulleyed astragalus alone, but the
> combined verdict of the many characters used in the phylogenetic
> analyses of the past 30 or so years. You could look up some of those

> see how characters of mesonychid astragali (and any other relevant
> characters) are scored if you want to know how that happened.

> > I know you dislike the language of the preceding paragraph, so I put it to you this way:
> > what else, if anything, disqualifies mesonychids from being artiodactyls?

> I don't know. I would have to look up the previously mentioned
> literature, which I have not done for quite a few years.

Could you do that now? I am preparing for two back to back conferences in topology next week
and the one after, and I am giving talks about my recent research in both of them.

> > Even if I am wrong in the interpretation of the astragalus of *Synoplotherium,* I have to ask:
> > how difficult would it have been for a double pulleyed astragalus to emerge independently in Mesonychidae?
> There is no way of answering that question, as far as I can tell. But of
> course phylogenetics should not be based on single characters. It's the
> agreement among many independent characters that gives us confidence in
> any phylogeny.
> > A look at other illustrations in Romer's _Vertebrate_Paleontology_ gives me the impression that
> > a double-pulleyed astragalus is not all that hard to come by. Perissodactyls (p. 433, Fig. 329),
> > litopterns (p.402, Fig. 305) and some notoungulates (p. 397, Fig. 299) had what look to me like
> > double-pulleyed astragali.

Some more clearly than others. Among the three litopternans whose pes is shown, that
of *Thoatherium* looks most obvious. Among the notoungulates, the best example
is *Protypotherium*. Both animals were built for fast running; *Thoatherium* was even
more completely monodactyl than *Equus*.

> Does Romer mention the term? It's conceivable that you are unclear on
> the definition.

A quick look at your copy of Romer's 1966 edition should settle this baseless
speculation of yours one way or the other.

Romer talks about astraguli in three places, noting that it is the term used
by mammalologists for what in other classes is called the intermedium.
He writes that it "develops a rolling joint over which the tibia moves freely" (p. 307).

He doesn't mention "double-pulleyed," though.

> If so, that might explain why paleontologists make so
> much of the character and do generally find it diagnostic of Artiodactyla.

"diagnostic" within what clade? See above about perissodactyls and litopterns.

> > > I don't recall
> >> ever seeing an analysis that did that. Except for Andrewsarchus, that is.
> >
> > Did it have a double-pulleyed astragalus?

> I don't believe its astragalus is known. It's other characters that put
> it within artiodactyls (and away from mesonychids).

That's not very harmonious with what I read here:

"However, the close grouping of whales with hippopotami in cladistic analyses only surfaces following the deletion of Andrewsarchus, which has often been included within the mesonychids.[13][14] One possible conclusion is that Andrewsarchus has been incorrectly classified. The current uncertainty may, in part, reflect the fragmentary nature of the remains of some crucial fossil taxa, such as Andrewsarchus.[13]"

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

[erik simpson]

Andrewsarchus isn't a good taxon to hang anybody's hat on. It's a single skull (no astralagus), and it's placement within Mesonichids or Entelodonts is
controversial. At least genetic data is available for whales, hippos and other Artiodactyls. No such luck for Mesonichids or Entelodonts.

[Peter Nyikos]

On Friday, July 8, 2022 at 1:52:49 PM UTC-4, Glenn wrote:
> On Friday, July 8, 2022 at 10:12:32 AM UTC-7, peter2...@gmail.com wrote:
> > This is my second and final reply to this post by John:
> > On Tuesday, July 5, 2022 at 11:54:14 AM UTC-4, John Harshman wrote:

> > > These days, time calibrations don't generally use a molecular clock.
> > > They use maximum likelihood models that allow the rate of evolution to
> > > vary among taxa, with some parameter limiting the rate of change in
> > > rate.

> > How does ML help with that? The only place where the webpage that I linked mentions ML
> > is very vague about how it is used and how useful it is:
> >
> > "Molecular clock users have developed workaround solutions using a number of statistical approaches including maximum likelihood techniques and later Bayesian modeling. In particular, models that take into account rate variation across lineages have been proposed in order to obtain better estimates of divergence times. These models are called relaxed molecular clocks[33] because they represent an intermediate position between the 'strict' molecular clock hypothesis and Joseph Felsenstein's many-rates model[34] and are made possible through MCMC techniques that explore a weighted range of tree topologies and simultaneously estimate parameters of the chosen substitution model."
> > > The more fossil calibration points, the better. But of course,
> > > strictly speaking, a fossil can only provide a lower limit on the age of
> > > a node. Generally, some distribution of allowed ages is also used in the
> > > model. A common program used for such things is called BEAST, which you
> > > might look up if you're interested.

"BEAST" has so many meanings that I haven't been able to find
a reference to the program in Wikipedia. The best I've been able
to do so far is the following:

https://bioinformatics-tutorials.readthedocs.io/en/latest/beastintro/

Most of it is Greek to me, partly because of what I wrote next:

> > Unlike you and "Mickey Mortimer," I don't know the first thing about computer programming.
> > When I needed something to be programmed when I was a "first Louie" in the Army, I gave
> > it to a bright high school student who worked for us in his spare time.

> You only need to know that computers are programmed. Say it out loud.
>
> ML is "guessing", but believers won't accept that.

As used by Harshman and myself, it is guessing of a specific sort about phylogeny, but the long article you link
below doesn't seem to get into that application at all.

> "We now under-
> stand the limitations of maximum likelihood better
> than Fisher did, but far from well enough to guar-
> antee safety in its application in complex situations
> where it is most needed. Maximum likelihood re-
> mains a truly beautiful theory, even though tragedy
> may lurk around a corner."
>
> https://arxiv.org/pdf/0804.2996.pdf

Or, as one graphic variation has it, "A beautiful theory was murdered by a brutal gang of cold facts."

Harshman, as I remarked in my reply to a post of his along another sub-thread today,
seems to have become badly handicapped at thinking like a biologist through
lack of meaningful interaction with research biologists *qua* research biologists,
as opposed to *qua* computer programmers.

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

Univ. of South Carolina at Columbia
http://people.math.sc.edu/nyikos

PS Harshman's response to what I wrote below was very unsatisfactory.

[John Harshman]

On 7/11/22 11:52 AM, Peter Nyikos wrote:
> On Thursday, July 7, 2022 at 10:11:54 PM UTC-4, John Harshman wrote:
>> On 7/7/22 6:51 PM, Peter Nyikos wrote:
>>> On Tuesday, July 5, 2022 at 11:54:14 AM UTC-4, John Harshman wrote:
>>>> On 7/5/22 6:26 AM, Peter Nyikos wrote:
>>>>> On Friday, July 1, 2022 at 5:34:05 PM UTC-4, Glenn wrote:
>>>>>> On Wednesday, June 29, 2022 at 8:20:49 PM UTC-7, John Harshman wrote:
>>>>>>> On 6/29/22 5:38 PM, Peter Nyikos wrote:
>>>>>
>>>>>>>> By the way, Cetacea is conspicuous by its absence. Paleontology would put it close to artiodactyla, though perhaps not as close as the molecular evidence does. The mesonychids are indicated as the sister group paleontologically, and their paraxonian digits argue for closeness with artiodactyls.
>>>
>>> I didn't want to stir things up too much at the beginning, so I talked in terms of "sister group" rather
>>> than "ancestor, as in "mesonychids may be ancestral to whales."
>
>> I didn't notice that bit. Paleontology in fact puts Cetacea within
>> Artiodactyla, not adjacent, since Thewissen's and Gingrich's independent
>> papers in 2001. Mesonychids, on the other hand, have been outside
>> Artiodactyla in all analyses I'm aware of.
>
> Why didn't you reference a single one of them?

Surely you can google as easily as I can. Skip the middle man.

>>>>>>> Cetacea as an order doesn't exist. Whales are artiodactyls.
>>>>>>> Andrewsarchus, interestingly, is probably not a mesonychid but an
>>>>>>> artiodactyl close to whales.
>>>>>
>>>>> Harshman is here ignoring the evidence that mesonychids were also artiodactyls, some of which I gave him (see above)
>>>>> though not yet giving the same conclusion.
>>>
>>>> "Close to artiodactyls" and "were also artiodactyls" are not the same thing.
>>>
>>> I take it you wrote this for Glenn's benefit. Or didn't you understand what
>>> "though not yet giving the same conclusion" was all about?
>
>> It was in fact unclear what that meant.
>
> I was very clear, you just didn't pay close attention when reading it.

Agree to disagree.

>> Still wondering what the
>> evidence is that mesonychids are artiodactyls.
>
> evidence = overwhelming evidence?

Still wondering what the evidence is, whether it's strong or weak.

>>>> If you want to present evidence that mesonychids, like whales,
>>>> are nested within Artiodactyla, I would be interested.
>>>
>>> In my 1945 second edition of Romer's _Vertebrate_Paleontology_, p. 367, Fig. 277, on the left, there are
>>> detailed drawings of the manus and pes of *Synoplotherium,* a mesonychid. Its astragalus looks
>>> quite a lot like that of the Entelodont *Dinohyus,* shown on p. 452, Fig. 342 and that of the Camelid
>>> *Poebrotherium,* p. 461, Fig. 348. All three astraguli look double-pulleyed to me.
>
>> I'm suspicious of this conclusion, which has never been mentioned in any
>> phylogenetic analysis I know about.
>
> What kept you from opening your 1966 edition and looking up the drawings in there?
> The index should have made it child's play.
>
> Don't you think *you* can recognize a double-pulleyed astragalus?

I think so, though I may be wrong. The question you should consider is
why no paleontologist can apparently recognize one, if your
understanding is correct.

> > Either you are wrong about this or
>> mammal paleontologists who have done the work are all incompetent.
>
> If these are the only alternatives you could think of, then you must be suffering from
> a long lack of meaningful discussions with research biologists *qua* research biologists.
> I thought of two others right off the bat:
>
> 1. *Synoplotherium* was found to have been mis-classified after Romer's 1945 edition.
>
> 2. The reconstruction to which I referred was inaccurate. [The reconstruction of the skull
> of Archaeopteryx in the same edition was criticized in a peer-reviewed paper comparing
> it with modern reconstructions of the Eichstatt specimen and the London specimen.]

Both possible. Can we at least agree that, based on the failure of
paleontologists to note it (and the excitement of paleontologists upon
discovering whale astragali), that no mesonychid is known to have a
double-pulley astragalus, despite anything you may have found in Romer 1045?

> Later, while investigating 1. and finding it was incorrect, I discovered what is
> almost surely the real reason:
>
> 3. The most complete skeleton of *Synoplotherium* was never subjected to a modern character analysis,
> and so, other mesonychids were used in all the phylogenetic analyses that placed them outside Artiodactyla.
>
> I conclude this from a 2016 account by Riley Black in _Scientific American_, long after the all
> the analyses of mesonychid affinities of which I know.
>
> https://blogs.scientificamerican.com/laelaps/yale-s-mysterious-fossil-beast/
> Subtitle: Despite being discovered over a century ago, *Synoplotherium* is still an enigma
> The crucial excerpt:
> "The first and last major source of information on Synoplotherium is a section of a massive review of Eocene mammals laid out by paleontologist Jacob Wortman in 1902. He covered the whole animal from its oversized skull to what remained of its tail, detecting a few surprises along the way."
>
> It's a fascinating and very readable article by a prolific contributor to SA, but for the sake of brevity I
> mention only one more detail: the almost complete skeleton in the Peabody Museum of Yale, which was the
> focus of the article, is displayed to the public, despite the accepted practice of only displaying casts.

Interesting. I do wonder why nobody has put it into any analyses,
especially since it's so complete. But what accepted practice of only
displaying casts? I see original vertebrae fossils in museums all the
time. The Field Museum doesn't display only casts. Sue, in all her
glory, is right in the main hall.

>>> Isn't the lack of a double-pulleyed astragalus the item that convinced most vertebrate paleontologists
>>> to eliminate mesonychids from cetacean ancestry? AFAIK we have never found any mesonychid DNA,
>>> so disqualification would have to use fossil morphology.
>
>> It's not the lack of a double-pulleyed astragalus alone, but the
>> combined verdict of the many characters used in the phylogenetic
>> analyses of the past 30 or so years. You could look up some of those
>> see how characters of mesonychid astragali (and any other relevant
>> characters) are scored if you want to know how that happened.
>
>>> I know you dislike the language of the preceding paragraph, so I put it to you this way:
>>> what else, if anything, disqualifies mesonychids from being artiodactyls?
>
>> I don't know. I would have to look up the previously mentioned
>> literature, which I have not done for quite a few years.
>
> Could you do that now? I am preparing for two back to back conferences in topology next week
> and the one after, and I am giving talks about my recent research in both of them.

I've been looking a bit. Everything so far is paywalled. O'Leary's
matrix has several astragalar characters, but I don't see any of them as
referring specifically to the double-pulley situation.

Here's a link. It's not to O'Leary's analysis, but that's where they got
the data matrix.

https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0007062#s4

>>> Even if I am wrong in the interpretation of the astragalus of *Synoplotherium,* I have to ask:
>>> how difficult would it have been for a double pulleyed astragalus to emerge independently in Mesonychidae?
>> There is no way of answering that question, as far as I can tell. But of
>> course phylogenetics should not be based on single characters. It's the
>> agreement among many independent characters that gives us confidence in
>> any phylogeny.
>>> A look at other illustrations in Romer's _Vertebrate_Paleontology_ gives me the impression that
>>> a double-pulleyed astragalus is not all that hard to come by. Perissodactyls (p. 433, Fig. 329),
>>> litopterns (p.402, Fig. 305) and some notoungulates (p. 397, Fig. 299) had what look to me like
>>> double-pulleyed astragali.
>
> Some more clearly than others. Among the three litopternans whose pes is shown, that
> of *Thoatherium* looks most obvious. Among the notoungulates, the best example
> is *Protypotherium*. Both animals were built for fast running; *Thoatherium* was even
> more completely monodactyl than *Equus*.
>
>> Does Romer mention the term? It's conceivable that you are unclear on
>> the definition.
>
> A quick look at your copy of Romer's 1966 edition should settle this baseless
> speculation of yours one way or the other.
>
> Romer talks about astraguli in three places, noting that it is the term used
> by mammalologists for what in other classes is called the intermedium.
> He writes that it "develops a rolling joint over which the tibia moves freely" (p. 307).
>
> He doesn't mention "double-pulleyed," though.

Pity.

>> If so, that might explain why paleontologists make so
>> much of the character and do generally find it diagnostic of Artiodactyla.
>
> "diagnostic" within what clade? See above about perissodactyls and litopterns.

It's diagnostic of Artiodactyla. Not Perissodactyla, not Litopterna, not
Ungulata (which does not appear to exist). Try the papers by Thewissen
and Gingerich.

Gingerich P.D., ul Haq M., Zalmout I.S., Khan I.H., Malkani M.S. Origin
of whales from early artiodactyls: Hands and feet of Eocene Protocetidae
from Pakistan. Science 2001; 293:2239-2242.

Thewissen J.G.M., Williams E.M., Roe L.J. Hussain S.T. Skeletons of
terrestrial cetaceans and the relationship of whales to artiodactyls.
Nature 2001; 413: 277–281.

>>>> I don't recall
>>>> ever seeing an analysis that did that. Except for Andrewsarchus, that is.
>>>
>>> Did it have a double-pulleyed astragalus?
>
>> I don't believe its astragalus is known. It's other characters that put
>> it within artiodactyls (and away from mesonychids).
>
> That's not very harmonious with what I read here:
>
> "However, the close grouping of whales with hippopotami in cladistic analyses only surfaces following the deletion of Andrewsarchus, which has often been included within the mesonychids.[13][14] One possible conclusion is that Andrewsarchus has been incorrectly classified. The current uncertainty may, in part, reflect the fragmentary nature of the remains of some crucial fossil taxa, such as Andrewsarchus.[13]"

Not actually incompatible with what I said. In those analyses, I suspect
that Andrewsarchus pulls whales away from hippos but not out of
Artiodactyla.

> https://en.wikipedia.org/wiki/Mesonychid

[John Harshman]

On 7/11/22 12:57 PM, Peter Nyikos wrote:
> On Friday, July 8, 2022 at 1:52:49 PM UTC-4, Glenn wrote:
>> On Friday, July 8, 2022 at 10:12:32 AM UTC-7, peter2...@gmail.com wrote:
>>> This is my second and final reply to this post by John:
>>> On Tuesday, July 5, 2022 at 11:54:14 AM UTC-4, John Harshman wrote:
>
>>>> These days, time calibrations don't generally use a molecular clock.
>>>> They use maximum likelihood models that allow the rate of evolution to
>>>> vary among taxa, with some parameter limiting the rate of change in
>>>> rate.
>
>>> How does ML help with that? The only place where the webpage that I linked mentions ML
>>> is very vague about how it is used and how useful it is:
>>>
>>> "Molecular clock users have developed workaround solutions using a number of statistical approaches including maximum likelihood techniques and later Bayesian modeling. In particular, models that take into account rate variation across lineages have been proposed in order to obtain better estimates of divergence times. These models are called relaxed molecular clocks[33] because they represent an intermediate position between the 'strict' molecular clock hypothesis and Joseph Felsenstein's many-rates model[34] and are made possible through MCMC techniques that explore a weighted range of tree topologies and simultaneously estimate parameters of the chosen substitution model."
>>>> The more fossil calibration points, the better. But of course,
>>>> strictly speaking, a fossil can only provide a lower limit on the age of
>>>> a node. Generally, some distribution of allowed ages is also used in the
>>>> model. A common program used for such things is called BEAST, which you
>>>> might look up if you're interested.
>
> "BEAST" has so many meanings that I haven't been able to find
> a reference to the program in Wikipedia. The best I've been able
> to do so far is the following:
>
> https://bioinformatics-tutorials.readthedocs.io/en/latest/beastintro/

Here: https://beast.community

> Most of it is Greek to me, partly because of what I wrote next:
>
>>> Unlike you and "Mickey Mortimer," I don't know the first thing about computer programming.
>>> When I needed something to be programmed when I was a "first Louie" in the Army, I gave
>>> it to a bright high school student who worked for us in his spare time.
>
>> You only need to know that computers are programmed. Say it out loud.
>>
>> ML is "guessing", but believers won't accept that.
>
> As used by Harshman and myself, it is guessing of a specific sort about phylogeny, but the long article you link
> below doesn't seem to get into that application at all.

I wouldn't call it guessing at all. Would you? BEAST isn't exactly ML;
it's Bayesian. But it uses a model of molecular evolution and computes
likelihoods of the data under various combinations of parameters.

>> "We now under-
>> stand the limitations of maximum likelihood better
>> than Fisher did, but far from well enough to guar-
>> antee safety in its application in complex situations
>> where it is most needed. Maximum likelihood re-
>> mains a truly beautiful theory, even though tragedy
>> may lurk around a corner."
>>
>> https://arxiv.org/pdf/0804.2996.pdf
>
> Or, as one graphic variation has it, "A beautiful theory was murdered by a brutal gang of cold facts."

Are you in fact buying this?

> Harshman, as I remarked in my reply to a post of his along another sub-thread today,
> seems to have become badly handicapped at thinking like a biologist through
> lack of meaningful interaction with research biologists *qua* research biologists,
> as opposed to *qua* computer programmers.

This is a ridiculous accusation. You have no idea what you're talking about.

Of course, Glenn cares about none of this. You're talking to yourself here.

[Glenn]

I care deeply, John. Much more than you.

"Discordance between gene trees and species tree in their topologies and times can lead to incorrect species tree estimates from concatenated gene sequences—this has been shown to occur with both maximum likelihood and Bayesian methods like those implemented in BEAST. "

https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1006650


Why should I believe a word you say?

[Peter Nyikos]

On Monday, July 11, 2022 at 5:52:06 PM UTC-4, John Harshman wrote:
> On 7/11/22 12:57 PM, Peter Nyikos wrote:
> > On Friday, July 8, 2022 at 1:52:49 PM UTC-4, Glenn wrote:
> >> On Friday, July 8, 2022 at 10:12:32 AM UTC-7, peter2...@gmail.com wrote:
> >>> This is my second and final reply to this post by John:
> >>> On Tuesday, July 5, 2022 at 11:54:14 AM UTC-4, John Harshman wrote:
> >
> >>>> These days, time calibrations don't generally use a molecular clock.
> >>>> They use maximum likelihood models that allow the rate of evolution to
> >>>> vary among taxa, with some parameter limiting the rate of change in
> >>>> rate.
> >
> >>> How does ML help with that? The only place where the webpage that I linked mentions ML
> >>> is very vague about how it is used and how useful it is:

Turns out far below that you never really answered this question, John.
Instead you recommended a program that is different from ML.

> >>> "Molecular clock users have developed workaround solutions using a number of statistical approaches including maximum likelihood techniques and later Bayesian modeling. In particular, models that take into account rate variation across lineages have been proposed in order to obtain better estimates of divergence times. These models are called relaxed molecular clocks[33] because they represent an intermediate position between the 'strict' molecular clock hypothesis and Joseph Felsenstein's many-rates model[34] and are made possible through MCMC techniques that explore a weighted range of tree topologies and simultaneously estimate parameters of the chosen substitution model."
> >>>> The more fossil calibration points, the better. But of course,
> >>>> strictly speaking, a fossil can only provide a lower limit on the age of
> >>>> a node. Generally, some distribution of allowed ages is also used in the
> >>>> model. A common program used for such things is called BEAST, which you
> >>>> might look up if you're interested.
> >
> > "BEAST" has so many meanings that I haven't been able to find
> > a reference to the program in Wikipedia. The best I've been able
> > to do so far is the following:
> >
> > https://bioinformatics-tutorials.readthedocs.io/en/latest/beastintro/
> Here: https://beast.community
> > Most of it is Greek to me, partly because of what I wrote next:
> >
> >>> Unlike you and "Mickey Mortimer," I don't know the first thing about computer programming.
> >>> When I needed something to be programmed when I was a "first Louie" in the Army, I gave
> >>> it to a bright high school student who worked for us in his spare time.
> >
> >> You only need to know that computers are programmed. Say it out loud.
> >>
> >> ML is "guessing", but believers won't accept that.
> >
> > As used by Harshman and myself, it is guessing of a specific sort about phylogeny, but the long article you link
> > below doesn't seem to get into that application at all.


> I wouldn't call it guessing at all. Would you?

The way I understand it, ML tries to find a phylogeny that comes as close to possible to the ideal,
of all evolution proceeding at the same rate, as a first approximation. If some branch grows much
longer or much shorter than average, the branch is calibrated to bring the relative times as close
to each other as possible, and then the phylogeny is re-calculated. This goes on as long as it takes
a final tree to stabilize.

It's anybody's guess, though, whether this algorithm has really arrived at the true tree,
or even how far off it might be. Hence, "guessing." Am I correct?

> BEAST isn't exactly ML;
> it's Bayesian. But it uses a model of molecular evolution and computes
> likelihoods of the data under various combinations of parameters.

Notice how vague and general this description of yours is compared to mine.
Mine might be incorrect, but at least it gives a starting point to take off from.

> >> "We now under-
> >> stand the limitations of maximum likelihood better
> >> than Fisher did, but far from well enough to guar-
> >> antee safety in its application in complex situations
> >> where it is most needed. Maximum likelihood re-
> >> mains a truly beautiful theory, even though tragedy
> >> may lurk around a corner."
> >>
> >> https://arxiv.org/pdf/0804.2996.pdf
> >
> > Or, as one graphic variation has it, "A beautiful theory was murdered by a brutal gang of cold facts."

> Are you in fact buying this?

I don't buy anything until I know the scientific facts behind it.
Hence my probing above.

> > Harshman, as I remarked in my reply to a post of his along another sub-thread today,
> > seems to have become badly handicapped at thinking like a biologist through
> > lack of meaningful interaction with research biologists *qua* research biologists,
> > as opposed to *qua* computer programmers.

> This is a ridiculous accusation.

On the other thread, the context of my remark was unmistakably specific, and essentially the same
statement went like water off a duck's back. You acted as if I had never said it, probably
because you were at a loss for an answer.

And your reply to it came *before* you posted this reply. In fact, it is the post in this thread
that immediately precedes the one to which I am replying. Moreover, it was posted a bit less
than an hour before.

>You have no idea what you're talking about.

Nonsense. If anything, YOU had no idea what *I* was talking about.

But now, you have no excuse for not having any idea. Deal with it.

> Of course, Glenn cares about none of this. You're talking to yourself here.

Now that you've put your foot in it, both of these sentences are obsolete.

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

University of So. Carolina at Columbia
http://people.math.sc.edu/nyikos

[John Harshman]

Believe it or don't. What point were you trying to make with that quote
and link? Did you have one?

[John Harshman]

On 7/13/22 4:36 PM, Peter Nyikos wrote:
> On Monday, July 11, 2022 at 5:52:06 PM UTC-4, John Harshman wrote:
>> On 7/11/22 12:57 PM, Peter Nyikos wrote:
>>> On Friday, July 8, 2022 at 1:52:49 PM UTC-4, Glenn wrote:
>>>> On Friday, July 8, 2022 at 10:12:32 AM UTC-7, peter2...@gmail.com wrote:
>>>>> This is my second and final reply to this post by John:
>>>>> On Tuesday, July 5, 2022 at 11:54:14 AM UTC-4, John Harshman wrote:
>>>
>>>>>> These days, time calibrations don't generally use a molecular clock.
>>>>>> They use maximum likelihood models that allow the rate of evolution to
>>>>>> vary among taxa, with some parameter limiting the rate of change in
>>>>>> rate.
>>>
>>>>> How does ML help with that? The only place where the webpage that I linked mentions ML
>>>>> is very vague about how it is used and how useful it is:
>
> Turns out far below that you never really answered this question, John.
> Instead you recommended a program that is different from ML.

Different in some ways, the same in others. The big difference is that
ML methods calculate likelihoods, the conditional probabilities of
observing the data given a model, while Bayesian methods calculate the
posterior probability of a tree given a likelihood, a prior probability,
and Bayes' theorem.

No, that's nothing like what ML does. It's possible to have a ML model
that incorporates a molecular clock, such that the path from root to tip
must have the same length for all taxa, but that's not the usual. What
the model usually does is assign a length to each branch of a tree in
terms of the expected number of changes on that branch. There's an
algorithm that determines the branch lengths that give the highest
likelihood to the data, and there are other parameters in the model,
also estimated from the data. The point of ML is that it determines the
likelihood of observing the actual data given the model (the model
includes various evolutionary parameters and a particular tree). It
tries to find the model parameter values that maximize the likelihood.

Bayesian methods perform a likelihood calculation as input to the
Bayesian posterior probability of a tree.

> It's anybody's guess, though, whether this algorithm has really arrived at the true tree,
> or even how far off it might be. Hence, "guessing." Am I correct?
>
>> BEAST isn't exactly ML;
>> it's Bayesian. But it uses a model of molecular evolution and computes
>> likelihoods of the data under various combinations of parameters.
>
> Notice how vague and general this description of yours is compared to mine.
> Mine might be incorrect, but at least it gives a starting point to take off from.

It's quite complicated, and you would be better off looking at an actual
publication. Joe Felsenstein's book Inferring Phylogenies should be
available at your library and will tell you everything you need to know
about likelihood models.

>>>> "We now under-
>>>> stand the limitations of maximum likelihood better
>>>> than Fisher did, but far from well enough to guar-
>>>> antee safety in its application in complex situations
>>>> where it is most needed. Maximum likelihood re-
>>>> mains a truly beautiful theory, even though tragedy
>>>> may lurk around a corner."
>>>>
>>>> https://arxiv.org/pdf/0804.2996.pdf
>>>
>>> Or, as one graphic variation has it, "A beautiful theory was murdered by a brutal gang of cold facts."
>
>> Are you in fact buying this?
>
> I don't buy anything until I know the scientific facts behind it.
> Hence my probing above.

Your response seemed to be agreeing. Were you just rephrasing without
implying agreement?

>>> Harshman, as I remarked in my reply to a post of his along another sub-thread today,
>>> seems to have become badly handicapped at thinking like a biologist through
>>> lack of meaningful interaction with research biologists *qua* research biologists,
>>> as opposed to *qua* computer programmers.
>
>> This is a ridiculous accusation.
>
> On the other thread, the context of my remark was unmistakably specific, and essentially the same
> statement went like water off a duck's back. You acted as if I had never said it, probably
> because you were at a loss for an answer.
>
> And your reply to it came *before* you posted this reply. In fact, it is the post in this thread
> that immediately precedes the one to which I am replying. Moreover, it was posted a bit less
> than an hour before.

It was a ridiculous accusation there too. I didn't answer it because it
was ridiculous.

>> You have no idea what you're talking about.
>
> Nonsense. If anything, YOU had no idea what *I* was talking about.
>
> But now, you have no excuse for not having any idea. Deal with it.
>
>
>> Of course, Glenn cares about none of this. You're talking to yourself here.
>
> Now that you've put your foot in it, both of these sentences are obsolete.

?

[Glenn]

Should I include a maximum likelihood with that?

[Glenn]

It would likely tell you everything you need to know about Felsenstein, and you as well for that matter. He's an atheist activist, and a true believer in naturalism. What a wasted life.

[John Harshman]

I ask again: What point were you trying to make, if any?

[John Harshman]

What are you a true believer in? How do you waste your life?

[Glenn]

I need not be a true believer in anything, John. And I don't regard being a skeptic and allowing for the possibility of almost anything to be a waste of my life. On the other hand, obsessing over whether dinos had brown eyes or red eyes and whether they were cousin to the frog, I consider to be a complete waste of time, and nothing useful can ever be gained from such speculations, except to reinforce a preexisting assumption, or true belief, in the power of "nature" to produce the profound mysteries we see all around us. A janitor would contribute more to society, and a bookkeeper might someday shake up your whole world.

You asked. Now you'll say "See, I told you so". Nah, can't have any criticism while discussing paleontology. Only believers allowed.

[Glenn]

Are you aware that the title of this thread is "Man closer kin to naked mole rats than bats"?

[John Harshman]

I am indeed. But what does that have to do with my question?

[John Harshman]

Beyond the fact that you think paleontology is a waste of time, and that
you don't think "nature" can produce what we see (But what do you think
could?), I don't get much out of that rant. I suppose I also get that
you have no interest in communication.

[Glenn]

I got nothing out of that rant.

[John Harshman]

OK, I also get that you're an angry troll.

[Glenn]

You're the angry troll, John. Do you think the unfounded comments you've made about me have escaped my notice? Clearly you don't like your true beliefs questioned or criticized at all, and trolls ask what a person's point is after they have provided an experts opinion in the scientific literature criticizing those true beliefs. Constantly asking my what my "point" was or is, is trolling. And you have a clear history of trolling. Peter may not have seen our little exchanges in the thread about bird migration, but you know you're a troll. You might think you're good at it, but it doesn't take much of a subject to see the obviousness of that.

What happened to your own advice, bud? "Glenn is best ignored"...I suppose now you'll say "See, I told you why".

[John Harshman]

Dunno. Which unfounded comments?

> Clearly you don't like your true beliefs questioned or criticized at all

Have you questioned or criticized my true beliefs? When?

>, and trolls ask what a person's point is after they have provided an experts opinion in the scientific literature criticizing those true beliefs.

In general, if you think the expert is criticizing my true beliefs, you
have been mistaken about what the expert was actually saying. But you
seem to be saying here that your point has been to attack my supposed
beliefs. Is that true?

> Constantly asking my what my "point" was or is, is trolling.

Not if I actually want to know what your point is. You drop occasional
vague hints, but you never actually say.

> And you have a clear history of trolling. Peter may not have seen our little exchanges in the thread about bird migration, but you know you're a troll. You might think you're good at it, but it doesn't take much of a subject to see the obviousness of that.

You understand nothing about any of that, as far as I can tell.

> What happened to your own advice, bud? "Glenn is best ignored"...I suppose now you'll say "See, I told you why".

Still good advice, but I don't always follow it. Would you prefer to be
ignored or not ignored?

[Glenn]

You have my permission to ignore my posts, and any post that replies to mine.

Since you don't follow your own advice though, I doubt that you would take that seriously. Since you're a troll.

[John Harshman]

Well of course I don't need your permission. I was asking for your
preference, and you didn't say.

[Glenn]

You don't control what I say. What I might prefer is not relevant to your own behavior and advice.

[Peter Nyikos]

On Monday, July 11, 2022 at 4:54:33 PM UTC-4, John Harshman wrote:
> On 7/11/22 11:52 AM, Peter Nyikos wrote:
> > On Thursday, July 7, 2022 at 10:11:54 PM UTC-4, John Harshman wrote:
> >> On 7/7/22 6:51 PM, Peter Nyikos wrote:
> >>> On Tuesday, July 5, 2022 at 11:54:14 AM UTC-4, John Harshman wrote:

> >>>> If you want to present evidence that mesonychids, like whales,
> >>>> are nested within Artiodactyla, I would be interested.
> >>>
> >>> In my 1945 second edition of Romer's _Vertebrate_Paleontology_, p. 367, Fig. 277, on the left, there are

> >>> detailed drawings of the manus and pes of a mesonychid. Its astragalus looks

> >>> quite a lot like that of the Entelodont *Dinohyus,* shown on p. 452, Fig. 342 and that of the Camelid
> >>> *Poebrotherium,* p. 461, Fig. 348. All three astraguli look double-pulleyed to me.
> >
> >> I'm suspicious of this conclusion, which has never been mentioned in any
> >> phylogenetic analysis I know about.

There are good reasons, some mentioned below, and also the fact that Romer's 1945 book
had been superseded by the 1966 edition, and might not have kept the drawing of
*Synoplotherium,* but you've stubbornly resisted cracking that edition open:

> > What kept you from opening your 1966 edition and looking up the drawings in there?

<crickets>

> > The index should have made it child's play.
> >
> > Don't you think *you* can recognize a double-pulleyed astragalus?
> I think so, though I may be wrong.

Aren't you in touch with professional anatomists who could verify any guesses you make?

>The question you should consider is
> why no paleontologist can apparently recognize one, if your
> understanding is correct.

That is a highly insulting question, almost as intensely derogatory as your
second alternative below:

> > > Either you are wrong about this or
> >> mammal paleontologists who have done the work are all incompetent.
> >
> > If these are the only alternatives you could think of, then you must be suffering from
> > a long lack of meaningful discussions with research biologists *qua* research biologists.

When I remided you of this comment yesterday, you thumbed your nose at it.
But you ignored what I had subsequently written:

> > I thought of two others right off the bat:
> >
> > 1. *Synoplotherium* was found to have been mis-classified after Romer's 1945 edition.
> >
> > 2. The reconstruction to which I referred was inaccurate. [The reconstruction of the skull
> > of Archaeopteryx in the same edition was criticized in a peer-reviewed paper comparing
> > it with modern reconstructions of the Eichstatt specimen and the London specimen.]

> Both possible.

And neither possibility occurred to you? Such incompetence!

> Can we at least agree that, based on the failure of
> paleontologists to note it (and the excitement of paleontologists upon
> discovering whale astragali), that no mesonychid is known to have a
> double-pulley astragalus, despite anything you may have found in Romer 1045?

We have no business agreeing on such a thing until you take a look
at your 1966 edition and tell us your opinion of what you see.

What keeps you from looking at it? Afraid it might cramp your style?

> > Later, while investigating 1. and finding it was incorrect, I discovered what is
> > almost surely the real reason:
> >
> > 3. The most complete skeleton of *Synoplotherium* was never subjected to a modern character analysis,
> > and so, other mesonychids were used in all the phylogenetic analyses that placed them outside Artiodactyla.
> >
> > I conclude this from a 2016 account by Riley Black in _Scientific American_, long after the all
> > the analyses of mesonychid affinities of which I know.
> >
> > https://blogs.scientificamerican.com/laelaps/yale-s-mysterious-fossil-beast/
> > Subtitle: Despite being discovered over a century ago, *Synoplotherium* is still an enigma
> > The crucial excerpt:
> > "The first and last major source of information on Synoplotherium is a section of a massive review of Eocene mammals laid out by paleontologist Jacob Wortman in 1902. He covered the whole animal from its oversized skull to what remained of its tail, detecting a few surprises along the way."

I wonder how easily his "coverage" can be translated into a modern character analysis.
It might require a whole new study of the specimen.

> > It's a fascinating and very readable article by a prolific contributor to SA, but for the sake of brevity I
> > mention only one more detail: the almost complete skeleton in the Peabody Museum of Yale, which was the
> > focus of the article, is displayed to the public, despite the accepted practice of only displaying casts.

> Interesting. I do wonder why nobody has put it into any analyses,
> especially since it's so complete. But what accepted practice of only
> displaying casts? I see original vertebrae fossils in museums all the
> time. The Field Museum doesn't display only casts. Sue, in all her
> glory, is right in the main hall.

OK, I may have been unduly influenced by another part of Riley Black's account:

"Marsh fervently believed that bones were for experts only and any attempt at public display hindered research. Reconstructions should only be done on paper, never with real bones. It’s clear that not everyone agreed with this view, though, as the Peabody’s paleontologists went about mounting the bones of Synoplotherium almost immediately after Marsh’s death in 1899. Synoplotherium was the first creature to ever be mounted at the Peabody."

Such a distinguished fossil, and so little understood!

> >>> Isn't the lack of a double-pulleyed astragalus the item that convinced most vertebrate paleontologists
> >>> to eliminate mesonychids from cetacean ancestry? AFAIK we have never found any mesonychid DNA,
> >>> so disqualification would have to use fossil morphology.
> >
> >> It's not the lack of a double-pulleyed astragalus alone, but the
> >> combined verdict of the many characters used in the phylogenetic
> >> analyses of the past 30 or so years.

Even if this is true, other characters of Synoplotherium may be different
from those of the mesonychids that had actually been used in the analyses that
put them outside Artirodactyla. Seems like a new analysis is called for.


Concluded in next reply, possibly today but perhaps only several days from now.


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

PS The reason for the uncertainty is that I am busy packing for my conference trip,
and this might be the only post I do to either s.b.p. or to t.o. today.

[Peter Nyikos]

OK, this is enough of a no-brainer that I can spare the time for it, despite what I wrote
a little while ago.

On Thursday, July 14, 2022 at 5:02:43 PM UTC-4, John Harshman wrote:
> On 7/14/22 1:30 PM, Glenn wrote:
> > On Thursday, July 14, 2022 at 1:09:41 PM UTC-7, John Harshman wrote:
> >> On 7/14/22 12:54 PM, Glenn wrote:
> >>> On Thursday, July 14, 2022 at 12:34:24 PM UTC-7, John Harshman wrote:
> >>>> On 7/14/22 11:18 AM, Glenn wrote:
> >>>>> On Thursday, July 14, 2022 at 6:13:14 AM UTC-7, John Harshman wrote:
> >>>>>> On 7/13/22 10:12 PM, Glenn wrote:
> >>>>>>> On Wednesday, July 13, 2022 at 6:47:20 PM UTC-7, John Harshman wrote:

> >>>>>>>> It's quite complicated, and you would be better off looking at an actual
> >>>>>>>> publication. Joe Felsenstein's book Inferring Phylogenies should be
> >>>>>>>> available at your library and will tell you everything you need to know
> >>>>>>>> about likelihood models.

[Glenn:]

> >>>>>>> It would likely tell you everything you need to know about Felsenstein, and you as well for that matter. He's an atheist activist, and a true believer in naturalism. What a wasted life.

I gather that Felsenstein is what I call an 'evilutionist', someone who goes beyond a belief
in common descent, to a conviction that no supernatural entity guided evolution (however
sporadically), and beyond to a conviction that all talk of a supernatural creator of our
universe is a "fairy tale" that normal people need to grow out of.

You once claimed that the belief of each of us having an immortal soul is "a fairy tale",
and I wonder whether you subscribe to the "and beyond" part as well.

> >>>>>> What are you a true believer in? How do you waste your life?
> >>>>>
> >>>>> I need not be a true believer in anything, John. And I don't regard being a skeptic and allowing for the possibility of almost anything to be a waste of my life. On the other hand, obsessing over whether dinos had brown eyes or red eyes and whether they were cousin to the frog, I consider to be a complete waste of time, and nothing useful can ever be gained from such speculations, except to reinforce a preexisting assumption, or true belief, in the power of "nature" to produce the profound mysteries we see all around us. A janitor would contribute more to society, and a bookkeeper might someday shake up your whole world.

Glenn is mistaken here about "dinosaur cousin to the frog". Michael Behe believes in common descent,
but he does not believe it can happen without a good lot of intelligent direction.
The agnostic Loren Eiseley generally seems to believe in unguided evolution,
but allowing for direction at crucial moments, such as when he says that the appearance
of the cerebral hemispheres may have been due to "the eternal mystery, the finger of God."

> >>>>>
> >>>>> You asked. Now you'll say "See, I told you so". Nah, can't have any criticism while discussing paleontology. Only believers allowed.
> >>>> Beyond the fact that you think paleontology is a waste of time, and that
> >>>> you don't think "nature" can produce what we see (But what do you think
> >>>> could?), I don't get much out of that rant. I suppose I also get that
> >>>> you have no interest in communication.
> >>>
> >>> I got nothing out of that rant.
> >> OK, I also get that you're an angry troll.
> >
> > You're the angry troll, John. Do you think the unfounded comments you've made about me have escaped my notice?
> Dunno. Which unfounded comments?

Don't play dumb, John. You painted yourself into a corner on the thread another thread, where you could not
produce any evidence for militant creationist-like accusations that you made against Glenn.
Erik jumped to your "defense" by attacking Glenn himself, and the thread came to a complete standstill
for six days until I let Erik know he had made a fool of himself. Last I looked, neither of you had done
any replies to either my final clarification of how badly you had painted yourself into the corner,
or my taking Erik to task.

> > Clearly you don't like your true beliefs questioned or criticized at all
> Have you questioned or criticized my true beliefs? When?

I have, but I am too pressed for time to do a rehash here. But I do
wonder how you will reply to my question above, about "evilutionism."

> >, and trolls ask what a person's point is after they have provided an experts opinion in the scientific literature criticizing those true beliefs.
> In general, if you think the expert is criticizing my true beliefs, you
> have been mistaken about what the expert was actually saying. But you
> seem to be saying here that your point has been to attack my supposed
> beliefs. Is that true?
> > Constantly asking my what my "point" was or is, is trolling.

> Not if I actually want to know what your poll int is.

I told Glenn some juicy details about you and especially Erik making
a scam out of it the way y'all did it to me, but I expect that will be
another one of the posts that y'all "can't see because you don't want to see it."

>You drop occasional
> vague hints, but you never actually say.

> > And you have a clear history of trolling. Peter may not have seen our little exchanges in the thread about bird migration, but you know you're a troll. You might think you're good at it, but it doesn't take much of a subject to see the obviousness of that.

> You understand nothing about any of that, as far as I can tell.

Which, I suspect, is not far at all. Erik in particular ran the scam for a year before
I realized what he was doing, and then I nailed him on it.

One difference is that he was more inclined to claim I was 'unclear,' but then
he would ask me questions having no bearing on what I was supposedly unclear about.


Peter Nyikos

[John Harshman]

No, I just thought you might satisfy my curiosity. There's always the
possibility I might do as you prefer, which I would suppose you would like.

[John Harshman]

On 7/14/22 5:52 PM, Peter Nyikos wrote:
> OK, this is enough of a no-brainer that I can spare the time for it, despite what I wrote
> a little while ago.
>
> On Thursday, July 14, 2022 at 5:02:43 PM UTC-4, John Harshman wrote:
>> On 7/14/22 1:30 PM, Glenn wrote:
>>> On Thursday, July 14, 2022 at 1:09:41 PM UTC-7, John Harshman wrote:
>>>> On 7/14/22 12:54 PM, Glenn wrote:
>>>>> On Thursday, July 14, 2022 at 12:34:24 PM UTC-7, John Harshman wrote:
>>>>>> On 7/14/22 11:18 AM, Glenn wrote:
>>>>>>> On Thursday, July 14, 2022 at 6:13:14 AM UTC-7, John Harshman wrote:
>>>>>>>> On 7/13/22 10:12 PM, Glenn wrote:
>>>>>>>>> On Wednesday, July 13, 2022 at 6:47:20 PM UTC-7, John Harshman wrote:
>
>>>>>>>>>> It's quite complicated, and you would be better off looking at an actual
>>>>>>>>>> publication. Joe Felsenstein's book Inferring Phylogenies should be
>>>>>>>>>> available at your library and will tell you everything you need to know
>>>>>>>>>> about likelihood models.
>
> [Glenn:]
>>>>>>>>> It would likely tell you everything you need to know about Felsenstein, and you as well for that matter. He's an atheist activist, and a true believer in naturalism. What a wasted life.
>
> I gather that Felsenstein is what I call an 'evilutionist', someone who goes beyond a belief
> in common descent, to a conviction that no supernatural entity guided evolution (however
> sporadically), and beyond to a conviction that all talk of a supernatural creator of our
> universe is a "fairy tale" that normal people need to grow out of.

Dunno. I've never asked him. What he is, more importantly, is an eminent
evolutionary biologist who did seminal work on phylogenetic analysis.
What he believes about God, etc., would seem wholly irrelevant to his
ability to explain ML to you.

> You once claimed that the belief of each of us having an immortal soul is "a fairy tale",
> and I wonder whether you subscribe to the "and beyond" part as well.

Yes I do, but that also is irrelevant to the subject here.

>>>>>>>> What are you a true believer in? How do you waste your life?
>>>>>>>
>>>>>>> I need not be a true believer in anything, John. And I don't regard being a skeptic and allowing for the possibility of almost anything to be a waste of my life. On the other hand, obsessing over whether dinos had brown eyes or red eyes and whether they were cousin to the frog, I consider to be a complete waste of time, and nothing useful can ever be gained from such speculations, except to reinforce a preexisting assumption, or true belief, in the power of "nature" to produce the profound mysteries we see all around us. A janitor would contribute more to society, and a bookkeeper might someday shake up your whole world.
>
> Glenn is mistaken here about "dinosaur cousin to the frog".

I think he was joking about that one, perhaps in a veiled reference to
Jurassic Park. Depending on how close one has to be in order to be a
cousin, that is. But he was also suggesting that the idea is ridiculous.
Glenn, you may figure from that, is a creationist. He also appears to
think that paleontology and evolutionary biology are useless endeavors.
Would you agree?

> Michael Behe believes in common descent,
> but he does not believe it can happen without a good lot of intelligent direction.
> The agnostic Loren Eiseley generally seems to believe in unguided evolution,
> but allowing for direction at crucial moments, such as when he says that the appearance
> of the cerebral hemispheres may have been due to "the eternal mystery, the finger of God."

So what? Glenn has never, that I know of, claimed that Behe or Eiseley
are worth paying attention to.

>>>>>>>
>>>>>>> You asked. Now you'll say "See, I told you so". Nah, can't have any criticism while discussing paleontology. Only believers allowed.
>>>>>> Beyond the fact that you think paleontology is a waste of time, and that
>>>>>> you don't think "nature" can produce what we see (But what do you think
>>>>>> could?), I don't get much out of that rant. I suppose I also get that
>>>>>> you have no interest in communication.
>>>>>
>>>>> I got nothing out of that rant.
>>>> OK, I also get that you're an angry troll.
>>>
>>> You're the angry troll, John. Do you think the unfounded comments you've made about me have escaped my notice?
>> Dunno. Which unfounded comments?
>
> Don't play dumb, John. You painted yourself into a corner on the thread another thread, where you could not
> produce any evidence for militant creationist-like accusations that you made against Glenn.
> Erik jumped to your "defense" by attacking Glenn himself, and the thread came to a complete standstill
> for six days until I let Erik know he had made a fool of himself. Last I looked, neither of you had done
> any replies to either my final clarification of how badly you had painted yourself into the corner,
> or my taking Erik to task.

Don't mistake "didn't choose to" for "couldn't".

>>> Clearly you don't like your true beliefs questioned or criticized at all
>> Have you questioned or criticized my true beliefs? When?
>
> I have, but I am too pressed for time to do a rehash here. But I do
> wonder how you will reply to my question above, about "evilutionism."

It seems completely irrelevant. And of course the term is insulting,
though you may not intend it that way.

>>> , and trolls ask what a person's point is after they have provided an experts opinion in the scientific literature criticizing those true beliefs.
>> In general, if you think the expert is criticizing my true beliefs, you
>> have been mistaken about what the expert was actually saying. But you
>> seem to be saying here that your point has been to attack my supposed
>> beliefs. Is that true?
>>> Constantly asking my what my "point" was or is, is trolling.
>
>> Not if I actually want to know what your poll int is.
>
> I told Glenn some juicy details about you and especially Erik making
> a scam out of it the way y'all did it to me, but I expect that will be
> another one of the posts that y'all "can't see because you don't want to see it."

You overestimate the interest people have in your posts. It's not
necessary not to want to see it, merely to have no interest in responding.

>> You drop occasional
>> vague hints, but you never actually say.
>
>>> And you have a clear history of trolling. Peter may not have seen our little exchanges in the thread about bird migration, but you know you're a troll. You might think you're good at it, but it doesn't take much of a subject to see the obviousness of that.
>
>> You understand nothing about any of that, as far as I can tell.
>
> Which, I suspect, is not far at all. Erik in particular ran the scam for a year before
> I realized what he was doing, and then I nailed him on it.

I don't know what scam you're talking about, but do you really think I'm
a troll?

> One difference is that he was more inclined to claim I was 'unclear,' but then
> he would ask me questions having no bearing on what I was supposedly unclear about.

You are often unclear, you know, though not as often as Glenn is.

[John Harshman]

On 7/14/22 4:45 PM, Peter Nyikos wrote:
> On Monday, July 11, 2022 at 4:54:33 PM UTC-4, John Harshman wrote:
>> On 7/11/22 11:52 AM, Peter Nyikos wrote:
>>> On Thursday, July 7, 2022 at 10:11:54 PM UTC-4, John Harshman wrote:
>>>> On 7/7/22 6:51 PM, Peter Nyikos wrote:
>>>>> On Tuesday, July 5, 2022 at 11:54:14 AM UTC-4, John Harshman wrote:
>
>>>>>> If you want to present evidence that mesonychids, like whales,
>>>>>> are nested within Artiodactyla, I would be interested.
>>>>>
>>>>> In my 1945 second edition of Romer's _Vertebrate_Paleontology_, p. 367, Fig. 277, on the left, there are
>>>>> detailed drawings of the manus and pes of a mesonychid. Its astragalus looks
>>>>> quite a lot like that of the Entelodont *Dinohyus,* shown on p. 452, Fig. 342 and that of the Camelid
>>>>> *Poebrotherium,* p. 461, Fig. 348. All three astraguli look double-pulleyed to me.
>>>
>>>> I'm suspicious of this conclusion, which has never been mentioned in any
>>>> phylogenetic analysis I know about.
>
> There are good reasons, some mentioned below, and also the fact that Romer's 1945 book
> had been superseded by the 1966 edition, and might not have kept the drawing of
> *Synoplotherium,* but you've stubbornly resisted cracking that edition open:
>
>>> What kept you from opening your 1966 edition and looking up the drawings in there?
>
> <crickets>

Because there wouldn't be much point to it.

>>> The index should have made it child's play.
>>>
>>> Don't you think *you* can recognize a double-pulleyed astragalus?
>> I think so, though I may be wrong.
>
> Aren't you in touch with professional anatomists who could verify any guesses you make?

Well of course Pandora could do it, if she's paying attention to any of
this stuff.

> >The question you should consider is
>> why no paleontologist can apparently recognize one, if your
>> understanding is correct.
>
> That is a highly insulting question, almost as intensely derogatory as your
> second alternative below:

But you're the one who's being insulting; I'm only the person who
notices and points out the implication. Is that not a good argument
against mesonychids having double-pulley astragali?

>>>> Either you are wrong about this or
>>>> mammal paleontologists who have done the work are all incompetent.
>>>
>>> If these are the only alternatives you could think of, then you must be suffering from
>>> a long lack of meaningful discussions with research biologists *qua* research biologists.
>
> When I remided you of this comment yesterday, you thumbed your nose at it.

And rightly so. Now that *is* insulting.

> But you ignored what I had subsequently written:
>
>>> I thought of two others right off the bat:
>>>
>>> 1. *Synoplotherium* was found to have been mis-classified after Romer's 1945 edition.
>>>
>>> 2. The reconstruction to which I referred was inaccurate. [The reconstruction of the skull
>>> of Archaeopteryx in the same edition was criticized in a peer-reviewed paper comparing
>>> it with modern reconstructions of the Eichstatt specimen and the London specimen.]
>
>> Both possible.
>
> And neither possibility occurred to you? Such incompetence!

You win some, you lose some. But it turns out that Synoplotherium is
still considered a mesonychid, so hypothesis 1 is wrong. And nobody has
studied it since 1902, so nobody is in a position to say that the
reconstruction is wrong unless it disagrees with that old study.

>> Can we at least agree that, based on the failure of
>> paleontologists to note it (and the excitement of paleontologists upon
>> discovering whale astragali), that no mesonychid is known to have a
>> double-pulley astragalus, despite anything you may have found in Romer 1045?
>
> We have no business agreeing on such a thing until you take a look
> at your 1966 edition and tell us your opinion of what you see.

Not relevant. What Romer said in 1966 is not relevant to what
paleontologists notice thirty or more years later.

> What keeps you from looking at it? Afraid it might cramp your style?

Notice how you keep accusing me of being afraid of this or that? Bad
habit, which you should strive to restrain.

>>> Later, while investigating 1. and finding it was incorrect, I discovered what is
>>> almost surely the real reason:
>>>
>>> 3. The most complete skeleton of *Synoplotherium* was never subjected to a modern character analysis,
>>> and so, other mesonychids were used in all the phylogenetic analyses that placed them outside Artiodactyla.
>>>
>>> I conclude this from a 2016 account by Riley Black in _Scientific American_, long after the all
>>> the analyses of mesonychid affinities of which I know.
>>>
>>> https://blogs.scientificamerican.com/laelaps/yale-s-mysterious-fossil-beast/
>>> Subtitle: Despite being discovered over a century ago, *Synoplotherium* is still an enigma
>>> The crucial excerpt:
>>> "The first and last major source of information on Synoplotherium is a section of a massive review of Eocene mammals laid out by paleontologist Jacob Wortman in 1902. He covered the whole animal from its oversized skull to what remained of its tail, detecting a few surprises along the way."
>
> I wonder how easily his "coverage" can be translated into a modern character analysis.
> It might require a whole new study of the specimen.

Not if it's a proper description accompanied by clear photos and/or
drawings.

>>> It's a fascinating and very readable article by a prolific contributor to SA, but for the sake of brevity I
>>> mention only one more detail: the almost complete skeleton in the Peabody Museum of Yale, which was the
>>> focus of the article, is displayed to the public, despite the accepted practice of only displaying casts.
>
>> Interesting. I do wonder why nobody has put it into any analyses,
>> especially since it's so complete. But what accepted practice of only
>> displaying casts? I see original vertebrae fossils in museums all the
>> time. The Field Museum doesn't display only casts. Sue, in all her
>> glory, is right in the main hall.
>
> OK, I may have been unduly influenced by another part of Riley Black's account:
>
> "Marsh fervently believed that bones were for experts only and any attempt at public display hindered research. Reconstructions should only be done on paper, never with real bones. It’s clear that not everyone agreed with this view, though, as the Peabody’s paleontologists went about mounting the bones of Synoplotherium almost immediately after Marsh’s death in 1899. Synoplotherium was the first creature to ever be mounted at the Peabody."
>
> Such a distinguished fossil, and so little understood!

Sounds like everyone knew Marsh was a dick, and his aversion to
displaying fossils was unique to him.

>>>>> Isn't the lack of a double-pulleyed astragalus the item that convinced most vertebrate paleontologists
>>>>> to eliminate mesonychids from cetacean ancestry? AFAIK we have never found any mesonychid DNA,
>>>>> so disqualification would have to use fossil morphology.
>>>
>>>> It's not the lack of a double-pulleyed astragalus alone, but the
>>>> combined verdict of the many characters used in the phylogenetic
>>>> analyses of the past 30 or so years.
>
> Even if this is true, other characters of Synoplotherium may be different
> from those of the mesonychids that had actually been used in the analyses that
> put them outside Artirodactyla. Seems like a new analysis is called for.

Could be. You could ask a paleontologist why it wasn't included.

[Glenn]

I said nothing about Behe, or ID, but simply commented on the obsession of fitting everything together in a tree of life.yes, yes, I know, it's to some now a bush, to others, who knows. The point is not that I question the concept of common descent itself. I don't, nor have any reason to. The *theory* of common descent and all that goes with it is a different matter. But again, that wasn't the issue.

- Houston, we have a problem.

[Trolidan7]

I am thinking that my posting handle used to be more similar to 'troll'
than it is now. I guess a 'troll' is generally considered to be bad
but who knows.

>> One difference is that he was more inclined to claim I was 'unclear,'
>> but then
>> he would ask me questions having no bearing on what I was supposedly
>> unclear about.
>
> You are often unclear, you know, though not as often as Glenn is.

You know, there is a vast array of different words here with varying
meanings to different people.

Among them are god, mind, soul, belief, religion, science, and guess.

It may be very close to impossible to know what weights, values, or
meanings anyone attaches to any of these words and more, if any.

Reasably in a foreign language, many of these symbol patterns and
the sounds that they might represent might have other symbol patterns
that might stand for slightly similar or different ideas, and it is
not obvious what would be a viable translation for different symbol
sets some times with subtle differences.

Various words however might have different meanings even within
those that might use specific languages.

Getting back to biology however, what is a 'maximum likelihood' model?

To me, it would seem feasible to me that there could be differences
in which some genes might change throughout time. If a critical gene
causes fatality to an organism if it is mutated in an array of different
places throughout the gene it seems to me that it might be more conserved
than 'junk DNA', but it might be that genes called 'junk' might have some
need to have some of its sequences conserved because sometimes they might
produce some viability even if they have been called 'junk' by someone.

There are also generation times and varying exposure to mutagens for germ
cells during the life time of an organism that might also effect the
rate of mutagenesis. Is there some standardized way of trying to estimate
these factors also?

When I type something like that into Wikipedia, I tend to get something
like formulas for Maxwell's distribution of speeds and probability
functions concerning the like.

Is this something vastly different when concerning equations estimating
comparisons of DNA sequences?

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

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

[John Harshman]

Was that a question? A maximum likelihood model is one that's used in a
maximum likelihood estimator. A maximum likelihood estimator attempts to
find the mode, or peak, of a conditional probability distribution. In
the phylogenetic case, if determines the probability of observing the
data conditional on a particular tree and model of evolution.

> To me, it would seem feasible to me that there could be differences
> in which some genes might change throughout time.  If a critical gene
> causes fatality to an organism if it is mutated in an array of different
> places throughout the gene it seems to me that it might be more conserved
> than 'junk DNA', but it might be that genes called 'junk' might have some
> need to have some of its sequences conserved because sometimes they might
> produce some viability even if they have been called 'junk' by someone.
>
> There are also generation times and varying exposure to mutagens for germ
> cells during the life time of an organism that might also effect the
> rate of mutagenesis.  Is there some standardized way of trying to estimate
> these factors also?

Yes, but they aren't generally used in phylogenetic analyses. Instead,
the estimated parameters are branch lengths and the relative frequencies
of particular sorts of changes. Different rates of evolution would of
course produce different branch lengths, and so would different amounts
of time, and there is no attempt to untangle those two. Trees are
estimated as unrooted and time-reversible, so there's no need to do so.

> When I type something like that into Wikipedia, I tend to get something
> like formulas for Maxwell's distribution of speeds and probability
> functions concerning the like.
>
> Is this something vastly different when concerning equations estimating
> comparisons of DNA sequences?

Not vastly, no. Try "maximum likelihood phylogeny" and you might get
better results.

> https://en.wikipedia.org/wiki/Maximum_likelihood_estimation
>
> https://en.wikipedia.org/wiki/Likelihood_function

[Trolidan7]

Ok. Thank you for your response.

>> https://en.wikipedia.org/wiki/Maximum_likelihood_estimation
>>
>> https://en.wikipedia.org/wiki/Likelihood_function
>

[Peter Nyikos]

On Tuesday, July 19, 2022 at 11:31:08 AM UTC-4, Trolidan7 wrote:
> On 7/14/22 7:47 PM, John Harshman wrote:
> > On 7/14/22 5:52 PM, Peter Nyikos wrote:
> >> OK, this is enough of a no-brainer that I can spare the time for it,
> >> despite what I wrote
> >> a little while ago.
> >>
> >> On Thursday, July 14, 2022 at 5:02:43 PM UTC-4, John Harshman wrote:
> >>> On 7/14/22 1:30 PM, Glenn wrote:

> >>>> And you have a clear history of trolling. Peter may not have seen
> >>>> our little exchanges in the thread about bird migration, but you
> >>>> know you're a troll. You might think you're good at it, but it
> >>>> doesn't take much of a subject to see the obviousness of that.
> >>
> >>> You understand nothing about any of that, as far as I can tell.
> >>
> >> Which, I suspect, is not far at all. Erik in particular ran the scam
> >> for a year before
> >> I realized what he was doing, and then I nailed him on it.
> >
> > I don't know what scam you're talking about, but do you really think I'm
> > a troll?

"troll" is not a very useful word because people seldom bother to
say what it means to them.

I've found "polemical opportunist" to better describe what Harshman does,
because it has a narrower range of meanings. It denotes someone who
frequently says things because they sound clever or convincing but
don't care if they are true, false, or of unknown veracity.

> I am thinking that my posting handle used to be more similar to 'troll'
> than it is now. I guess a 'troll' is generally considered to be bad
> but who knows.

In s.b.p. and talk.origins, it is a negative term. However, different people
attach different meanings to it, some much more negative than others.

<snip for focus>

> You know, there is a vast array of different words here with varying
> meanings to different people.
>
> Among them are god, mind, soul, belief, religion, science, and guess.

Very perceptive.

> It may be very close to impossible to know what weights, values, or
> meanings anyone attaches to any of these words and more, if any.

Yes. For instance, Harshman attaches one meaning to the word "insult"
when accusing others of it, and another when he is accused of it.

> https://en.wikipedia.org/wiki/Likelihood_functionI

I found these too general to be of value in understanding how ML
produces phylogenetic trees that are supposed to be "most likely."

The first webpage drew a blank after "phyl" in a word search.

The second is somewhat better, but [47] looks too general to be promising:

"A. W. F. Edwards (1972) established the axiomatic basis for use of the log-likelihood ratio as a measure of relative support for one hypothesis against another. The support function is then the natural logarithm of the likelihood function. Both terms are used in phylogenetics, but were not adopted in a general treatment of the topic of statistical evidence.[47]"


Peter Nyikos

[John Harshman]

You have no evidence that I've ever done any such thing, and in fact I
deny ever having done so.

>> I am thinking that my posting handle used to be more similar to 'troll'
>> than it is now. I guess a 'troll' is generally considered to be bad
>> but who knows.
>
>
> In s.b.p. and talk.origins, it is a negative term. However, different people
> attach different meanings to it, some much more negative than others.
>
> <snip for focus>
>
>
>> You know, there is a vast array of different words here with varying
>> meanings to different people.
>>
>> Among them are god, mind, soul, belief, religion, science, and guess.
>
> Very perceptive.
>
>
>> It may be very close to impossible to know what weights, values, or
>> meanings anyone attaches to any of these words and more, if any.
>
> Yes. For instance, Harshman attaches one meaning to the word "insult"
> when accusing others of it, and another when he is accused of it.

This is all from your fertile imagination.

Once again I recommend Joe Felsenstein's book *Inferring Phylogenies*.
Should be available in any decent university library.

Also a search for "maximum likelihood phylogeny" turned up several
explanatory articles (as well as random tangential stuff).

> The first webpage drew a blank after "phyl" in a word search.
>
> The second is somewhat better, but [47] looks too general to be promising:
>
> "A. W. F. Edwards (1972) established the axiomatic basis for use of the log-likelihood ratio as a measure of relative support for one hypothesis against another. The support function is then the natural logarithm of the likelihood function. Both terms are used in phylogenetics, but were not adopted in a general treatment of the topic of statistical evidence.[47]"

Once again, briefly, a likelihood is a conditional probability. A
maximum likelihood algorithm in phylogeny computes the probability of
observing the observed data given a particular tree and a model of
evolution. One then examines a great many trees, preferring the one with
the highest likelihood. A likelihood ratio test attempts to determine if
one tree is *significantly* better than another, using the computed
likelihoods of both trees. (Likelihoods of trees are extremely small
numbers and are usually presented as the negative logs of the computed
likelihoods.) Does that help?

[Trolidan7]

I think it might be a good idea to bow down and out of this
one and insult myself if needed. I will try to look it up
however.

[erik simpson]

On Tuesday, August 2, 2022 at 4:46:22 PM UTC-7, Trolidan7 wrote:

...

> I think it might be a good idea to bow down and out of this
> one and insult myself if needed. I will try to look it up
> however.

....
Wise choice. It's always hard to follow exchanges like this, and trust me, if you
participate it's inevitable that you will be insulted.

[Glenn]

I sure do. Here is just part of one exchange, among many others over the years, between you and I:

Begin section of a talk.origins post:

Quail are observed flying from Crete to Northern Libya, specifically to Derma. That's about 150 miles.
>
Where is that observation? Could you cite the publication?
>
Well, I could load a webpage with a line drawn from Crete to Derma if you wish. But yes, I did read it in a publication.
>
Wait, I'll just use your earlier referenced map:
https://www.researchgate.net/figure/Main-migration-paths-of-the-Common-Quail-Coturnix-coturnix-to-from-the-West-Palearctic_fig2_318885611
>
Looks like the ones that head directly for Alexandria go thru Turkey and Cyprus!

Yes, on that map. But that's not the one I'm referring to now.
>
I unfortunately neglected to cite mine, but it shows a map on which the
migration is a line from eastern Crete straight to Alexandria.
>>
https://flightforsurvival.org/common-quail/
>
Oh Lord, another "scientific" site of yours. Is that a "publication"?
>
Best I can do on the web. Is there a problem?
>
And you provide no reason why quail fly greater distances when crossing open water.
>
Because they have to?
>
That appears to be a question, [redacted name]. The only thing they have to do is die.

That seems counterproductive.

End repost.

Lets see if any interested parties at this point can guess who is doing the talking.

https://groups.google.com/g/talk.origins/c/UwwAoH2n7sM/m/82Xbe-V8CAAJ

One of the initial claim that spurred the conversation above originated with

"You would have seen, if you had looked, that C. coturnix flies across
the Mediterranean without stopping from approximately Rome to Tunis,
about 360 miles. But I'll agree it's not the easiest thing to find out.
That's the greatest distance I have found for them so far."

John didn't care if he didn't have evidence. It did sound clever, to provide examples of long range bird migration in the larger subject of over ocean migration. But John stuck to his guns, and rested his claim on a non-scientific website, explaining that that was the "Best I can do on the web". John basically ignored all the evidence I provided, as well as the reasoning, for why his claim was wrong and unsupported.
Did John care about whether his claim was true or false, when he answered the question implied by "And you provide no reason why quail fly greater distances when crossing open water " with the question "Because they have to?"...yet he claimed my sarcasm "seemed counterproductive".

So much for John "I don't care what 50 Nobel Prize winners say" Harshman.

[Glenn]

On Tuesday, August 2, 2022 at 4:46:22 PM UTC-7, Trolidan7 wrote:

I recommend taking an objective look, at all claims and the reasoning and language used to support such methodologies.

"As a 2012 paper published in Biological Reviews of the Cambridge philosophical Society reported, “Incongruence between phylogenies derived from morphological versus molecular analysis, and between trees based on different subsets of molecular sequences has become pervasive as datasets have expanded rapidly in both characters and species”.
Another paper published the following year in the journal Nature, highlighted the extent of the problem. The authors compared 1,070 genes in twenty different yeasts and got 1,070 different trees."

https://uncommondescent.com/intelligent-design/at-sci-news-new-interactive-phylogenetic-map-shows-full-diversity-of-life-on-earth/#comment-762264

2020:

"Incongruence among phylogenetic results has become a common occurrence in analyses of genome-scale data sets."

https://academic.oup.com/mbe/article/38/3/1090/5976982

There are many more to consider. Accepting there is "nothing to see here, move along" type of apologetics John practices shouldn't impress an honest unbiased person.

[Glenn]

This looks interesting, and worth researching and comparing other practices:

"Networks: expanding evolutionary thinking"

https://par.nsf.gov/servlets/purl/10054654

Not sure though why evolutionary "thinking" would be required.

[John Harshman]

You and me.

But I did have evidence.

> It did sound clever, to provide examples of long range bird migration in the larger subject of over ocean migration. But John stuck to his guns, and rested his claim on a non-scientific website, explaining that that was the "Best I can do on the web". John basically ignored all the evidence I provided, as well as the reasoning, for why his claim was wrong and unsupported.
> Did John care about whether his claim was true or false, when he answered the question implied by "And you provide no reason why quail fly greater distances when crossing open water " with the question "Because they have to?"...yet he claimed my sarcasm "seemed counterproductive".

None of your understanding of that is true. Sorry. Still, if all you're
looking for is long-distance, over-water migration, there's plenty
that's more impressive than Coternix quails. Arctic tern, for example.

> So much for John "I don't care what 50 Nobel Prize winners say" Harshman.

That seems to be another quote, but where is it from?

[John Harshman]

You really have to stop relying on creationist web sites for your
science information.

> 2020:
>
> "Incongruence among phylogenetic results has become a common occurrence in analyses of genome-scale data sets."
>
> https://academic.oup.com/mbe/article/38/3/1090/5976982

And I have to question whether you actually read that article.

> There are many more to consider. Accepting there is "nothing to see here, move along" type of apologetics John practices shouldn't impress an honest unbiased person.

For "honest unbiased person" read "creationist".

[Glenn]

Goody for you.

> > There are many more to consider. Accepting there is "nothing to see here, move along" type of apologetics John practices shouldn't impress an honest unbiased person.
> For "honest unbiased person" read "creationist".

No thanks.

[Glenn]

That's your claim, not mine.

"The National Science Foundation (NSF)-funded scientist says it is not unusual for high quality research to report genealogies that conflict with each other over the origins of certain organisms."

""We found 1,070 genes, and made 1,070 trees, and each one was different," Rokas says."

https://phys.org/news/2014-05-reconstructing-tree-life.html

Is that a creationist site, John? Will you claim that is "junk" or "crap", like Ron does?
Bet you would like to. And you'll think to weasel you way out of it, and essentially say "nothing to see here, move along" or "we got better at it".

[Glenn]

Heil!

No you didn't. And you still don't. You had no reasoning either.

> > It did sound clever, to provide examples of long range bird migration in the larger subject of over ocean migration. But John stuck to his guns, and rested his claim on a non-scientific website, explaining that that was the "Best I can do on the web". John basically ignored all the evidence I provided, as well as the reasoning, for why his claim was wrong and unsupported.
> > Did John care about whether his claim was true or false, when he answered the question implied by "And you provide no reason why quail fly greater distances when crossing open water " with the question "Because they have to?"...yet he claimed my sarcasm "seemed counterproductive".
> None of your understanding of that is true. Sorry. Still, if all you're
> looking for is long-distance, over-water migration, there's plenty
> that's more impressive than Coternix quails. Arctic tern, for example.

You know that this was a result of a specific claim you made of quails. But you can't help yourself.

> > So much for John "I don't care what 50 Nobel Prize winners say" Harshman.
> That seems to be another quote, but where is it from?

From many years ago, and it does not seem to be a direct quote. You did say something very much like that, and it could very well be the exact words you used. You have also claimed that everything is inference. I suppose you would deny both. I don't care.

[John Harshman]

You're denying that UD is creationist, or what?

> "The National Science Foundation (NSF)-funded scientist says it is not unusual for high quality research to report genealogies that conflict with each other over the origins of certain organisms."
>
> ""We found 1,070 genes, and made 1,070 trees, and each one was different," Rokas says."
>
> https://phys.org/news/2014-05-reconstructing-tree-life.html
>
> Is that a creationist site, John? Will you claim that is "junk" or "crap", like Ron does?

No, the creationist site was Uncommon Descent. What you have there is
still not a real scientific source, though, just "science journalism",
and they tend toward the sensational. Have you looked at the real papers?

[John Harshman]

Well, as long as you don't care, there seems no reason to reply.

[John Harshman]

On 8/2/22 11:14 AM, Peter Nyikos wrote:

This is a pretty clear, brief description:

http://www.deduveinstitute.be/~opperd/private/max_likeli.html

[Glenn]

I deny that I rely on any website for anything. I sure don't rely on you for anything, other than disinformation and subterfuge. But to your strawman, I don't consider UD to be "creationist". Do you consider that since you believe UD is creationist, that the link above found in UD is creationist, since UD posted it on their website? If not, why did you ask?

> > "The National Science Foundation (NSF)-funded scientist says it is not unusual for high quality research to report genealogies that conflict with each other over the origins of certain organisms."
> >
> > ""We found 1,070 genes, and made 1,070 trees, and each one was different," Rokas says."
> >
> > https://phys.org/news/2014-05-reconstructing-tree-life.html
> >
> > Is that a creationist site, John? Will you claim that is "junk" or "crap", like Ron does?

> No, the creationist site was Uncommon Descent. What you have there is
> still not a real scientific source, though, just "science journalism",
> and they tend toward the sensational. Have you looked at the real papers?

What you call the "creationist site" referenced another 'creationist site" which referenced the actual paper in Nature. Phys.org, another site, also references the Nature article, and quotes from one of two of the authors. What is "sensational" about that? Have you looked at the "real papers"? Before you made a fool of yourself? You might as well have called phys.org liars that misquoted a scientist, or made it up.

https://www.nature.com/articles/nature12130

What is sensational about the paper, John? They admit and identify the problem with incongruence in such methodologies, and provide a startling example. Yet, as so many papers do and have for years, essentially claimed they had a method of fixing incongurences, all expressing the belief that either it "is getting better" or it "will get better".
There are facts and there are opinions based on a variety of reasons, such as belief in common descent. From the paper's abstract:

"These results question the exclusive reliance on concatenation and associated practices, and argue that selecting genes with strong phylogenetic signals and demonstrating the absence of significant incongruence are essential for accurately reconstructing ancient divergences."

> > Bet you would like to. And you'll think to weasel you way out of it, and essentially say "nothing to see here, move along" or "we got better at it".

Well I was wrong, and should have included more ways that you try to weasel out of challenges. You did essentially do what Ron often does. But I suppose I should be satisfied that you don't practice in the lab, and that the subject amounts to a big fat zero.

[John Harshman]

This is not the sort of thing that can be discussed rationally with you.
Not sure there's any sort of thing, or if there is it hasn't come up. It
isn't clear what you point is, since you never actually state one, but
I'm supposing you want to suppose that maximum likelihood is an
unreliable method that should not be used. If so, you grossly
misinterpret the paper in question.

[Glenn]

I had a sub for breakfast. Anything else you'd like to contribute?

[Peter Nyikos]

I have plenty of evidence, John, going back over more than a decade.
Let's start with one case that occurred less than a month ago, on this very thread:

"Either you are wrong about this or mammal paleontologists who have done the work are all incompetent."

--https://groups.google.com/g/sci.bio.paleontology/c/WP7JtxgJIio/m/58Cewad9BQAJ
Jul 7, 2022, 10:11:54 PM

You obviously did not care whether there were any other possibilities,
yet anyone who is competent at thinking scientifically and has one-tenth
of our knowledge could have come up with at least two of them, as I did immediately.
And you had to ackowledge that they were valid alternatives to your false dichotomy.

And it certainly looks like a case of polemical opportunism: the evidence powerfully suggests that
you purposely chose a highly insulting false dichotomy for its "cleverness"
and didn't give a hoot about coming across as a competent scientist.

> and in fact I deny ever having done so.

You are in denial, in the psychologically negative sense of the term.
It continued in subsequent back-and-forth, and your last post in
that sequence was a masterpiece of polemical opportunism.

I haven't replied to that post yet, but if you keep on being in denial,
I can go through it with a fine toothed comb.

> >> I am thinking that my posting handle used to be more similar to 'troll'
> >> than it is now. I guess a 'troll' is generally considered to be bad
> >> but who knows.
> >
> >
> > In s.b.p. and talk.origins, it is a negative term. However, different people
> > attach different meanings to it, some much more negative than others.
> >
> > <snip for focus>
> >
> >
> >> You know, there is a vast array of different words here with varying
> >> meanings to different people.
> >>
> >> Among them are god, mind, soul, belief, religion, science, and guess.
> >
> > Very perceptive.
> >
> >
> >> It may be very close to impossible to know what weights, values, or
> >> meanings anyone attaches to any of these words and more, if any.
> >
> > Yes. For instance, Harshman attaches one meaning to the word "insult"
> > when accusing others of it, and another when he is accused of it.

> This is all from your fertile imagination.

This formulaic taunt is devoid of reality in this context.

When I give evidence of you behaving in a irresponsible way,
you very frequently label it an "insult" without trying to refute what I wrote.

In contrast, when you make insulting comments about me, without any evidence, like the one
you have uttered just now, you sometimes retort, "that wasn't an insult, it was an observation."

Your behavior was even more blatantly hypocritical and irresponsible on the thread
. . Questions about BBC’s “Prehistoric Planet” Episode 1
This was where you wrote such highly insulting things about Glenn as:

"most of his posts seeming intended to show that "evolutionists" are bad and/or idiots,"
. . --https://groups.google.com/g/sci.bio.paleontology/c/o7a-W7iuTZE/m/XtrSILUmAQAJ
. . Jul 5, 2022, 9:05:48 AM

You never came up with an example of a single post where that "seeming" would
be clear to an unbiased person reading the post. What's more, Erik badly
misunderstood what the two links that you tried to use as evidence against Glenn
were all about, and you blindly followed Erik without even reading the links.


Double standards: one for your insults, and one for things you label insults by others.


At times like these, you and Erik typically claim to want to get back to paleontology,
yet when he made the claim near the end of the thread where the above took place, he didn't
make any moves in that direction. I challenged him to demonstrate his fondness
for getting back to paleontology, and he folded.

I've snipped the rest, which I could use to get us back to paleontology, but since you
were perfectly OK with Erik's behavior, I'll pass for now to see whether you can be sincere
in dealing with the above false claims of yours.


Peter Nyikos

[John Harshman]

This is all purely in your imagination.

>> and in fact I deny ever having done so.
>
> You are in denial, in the psychologically negative sense of the term.
> It continued in subsequent back-and-forth, and your last post in
> that sequence was a masterpiece of polemical opportunism.
>
> I haven't replied to that post yet, but if you keep on being in denial,
> I can go through it with a fine toothed comb.

Please don't. It will all be your imagined characterization of purely
innocent statements.

How about paleontology?

>>>> I am thinking that my posting handle used to be more similar to 'troll'
>>>> than it is now. I guess a 'troll' is generally considered to be bad
>>>> but who knows.
>>>
>>>
>>> In s.b.p. and talk.origins, it is a negative term. However, different people
>>> attach different meanings to it, some much more negative than others.
>>>
>>> <snip for focus>
>>>
>>>
>>>> You know, there is a vast array of different words here with varying
>>>> meanings to different people.
>>>>
>>>> Among them are god, mind, soul, belief, religion, science, and guess.
>>>
>>> Very perceptive.
>>>
>>>
>>>> It may be very close to impossible to know what weights, values, or
>>>> meanings anyone attaches to any of these words and more, if any.
>>>
>>> Yes. For instance, Harshman attaches one meaning to the word "insult"
>>> when accusing others of it, and another when he is accused of it.
>
>> This is all from your fertile imagination.
>
> This formulaic taunt is devoid of reality in this context.

I'm afraid you aren't in a position to judge, living within your fertile
imagination as you do.

> When I give evidence of you behaving in a irresponsible way,
> you very frequently label it an "insult" without trying to refute what I wrote.
>
> In contrast, when you make insulting comments about me, without any evidence, like the one
> you have uttered just now, you sometimes retort, "that wasn't an insult, it was an observation."
>
> Your behavior was even more blatantly hypocritical and irresponsible on the thread
> . . Questions about BBC’s “Prehistoric Planet” Episode 1
> This was where you wrote such highly insulting things about Glenn as:
>
> "most of his posts seeming intended to show that "evolutionists" are bad and/or idiots,"
> . . --https://groups.google.com/g/sci.bio.paleontology/c/o7a-W7iuTZE/m/XtrSILUmAQAJ
> . . Jul 5, 2022, 9:05:48 AM

Are those insulting? Do you disagree? What do you think is the intention
behind most of his posts?

> You never came up with an example of a single post where that "seeming" would
> be clear to an unbiased person reading the post. What's more, Erik badly
> misunderstood what the two links that you tried to use as evidence against Glenn
> were all about, and you blindly followed Erik without even reading the links.
>
>
> Double standards: one for your insults, and one for things you label insults by others.
>
>
> At times like these, you and Erik typically claim to want to get back to paleontology,
> yet when he made the claim near the end of the thread where the above took place, he didn't
> make any moves in that direction. I challenged him to demonstrate his fondness
> for getting back to paleontology, and he folded.
>
> I've snipped the rest, which I could use to get us back to paleontology, but since you
> were perfectly OK with Erik's behavior, I'll pass for now to see whether you can be sincere
> in dealing with the above false claims of yours.

So essentially, you snip all the paleontology in order to concentrate on
something more interesting to you, the supposed crimes of other people.

[Glenn]

I learned a long time ago that the only person that I would fool is myself.

You take such things as I say as an insult, yet I am serious. You don't appear to have learned such lessons, and appear not to have grown up. Perhaps, probably, I am too eager at times to consider the reasons, but even then I do confront myself before the first word comes out of my mouth. I don't get defensive, although many seem to be convinced that I do. Invoking "most of my posts" isn't really a wise thing to do, unproductive. The specific instance Peter cites should be the subject of your question. Yet instead of addressing that, you ask leading questions, add some innuendo, and deflect from responding directly to Peter. He is right on the money about insults. I recall times when you have argued that your insults were not insults, or not ad hominem, because they were true. It appears you do not or will not consider others may sincerely feel the same way. If so, why not?

[Peter Nyikos]

On Wednesday, August 3, 2022 at 9:35:10 AM UTC-4, John Harshman wrote:
> On 8/2/22 11:33 PM, Glenn wrote:
> > On Tuesday, August 2, 2022 at 7:46:15 PM UTC-7, John Harshman wrote:
> >> On 8/2/22 7:13 PM, Glenn wrote:

> >>> I recommend taking an objective look, at all claims and the reasoning and language used to support such methodologies.
> >>>
> >>> "As a 2012 paper published in Biological Reviews of the Cambridge philosophical Society reported, “Incongruence between phylogenies derived from morphological versus molecular analysis, and between trees based on different subsets of molecular sequences has become pervasive as datasets have expanded rapidly in both characters and species”.
> >>> Another paper published the following year in the journal Nature, highlighted the extent of the problem. The authors compared 1,070 genes in twenty different yeasts and got 1,070 different trees."
> >>>
> >>> https://uncommondescent.com/intelligent-design/at-sci-news-new-interactive-phylogenetic-map-shows-full-diversity-of-life-on-earth/#comment-762264

> >> You really have to stop relying on creationist web sites for your
> >> science information.

As long as Glenn comes up with scientific statements that you cannot undermine, originally from
respected scientific sources, you haven't a leg to stand on. What's more, your wording here makes
you sound like a control freak.

> > That's your claim, not mine.
> You're denying that UD is creationist, or what?

Non sequitur. See what I wrote in response just now. I dealt with the bottom line, while talk
about secondary or tertiary sources is just a diversion.

> > "The National Science Foundation (NSF)-funded scientist says it is not unusual for high quality research to report genealogies that conflict with each other over the origins of certain organisms."
> >
> > ""We found 1,070 genes, and made 1,070 trees, and each one was different," Rokas says."
> >
> > https://phys.org/news/2014-05-reconstructing-tree-life.html
> >
> > Is that a creationist site, John? Will you claim that is "junk" or "crap", like Ron does?

> No, the creationist site was Uncommon Descent.

You and Ron O falsely equate ID with creationism. While the majority of the rank and file who
read UD are probably creationists, I have found lots of the writing of authors to be
noncommittal about creationism and common descent.

Heck, even Glenn told me, much to my pleasant surprise, that he has no
problems with common descent.

> What you have there is
> still not a real scientific source, though, just "science journalism",
> and they tend toward the sensational. Have you looked at the real papers?

Why should Glenn do that, when you have made a blatantly false claim
about a link you ADMITTEDLY hadn't even read:

"it seems clear that he's attacking paleontology, especially in
the second link, which plays up the prevalence of faked (composite) fossils."

This was decisively refuted by me in:

https://groups.google.com/g/sci.bio.paleontology/c/o7a-W7iuTZE/m/MG_AiPnsAAAJ
Re: Questions about BBC’s “Prehistoric Planet” Episode 1
Jul 4, 2022, 3:27:50 PM

> > Bet you would like to. And you'll think to weasel you way out of it, and essentially say "nothing to see here, move along" or "we got better at it".


> >>> 2020:
> >>>
> >>> "Incongruence among phylogenetic results has become a common occurrence in analyses of genome-scale data sets."
> >>>
> >>> https://academic.oup.com/mbe/article/38/3/1090/5976982
> >> And I have to question whether you actually read that article.
> >>> There are many more to consider. Accepting there is "nothing to see here, move along" type of apologetics John practices shouldn't impress an honest unbiased person.

> >> For "honest unbiased person" read "creationist".

More polemical opportunism from you. You've been spoiled rotten by Erik and Oxyaena here
and by something like a dozen irresponsible people in talk.origins for many years.
One consequence is that you have, or pretend to have, a completely false picture of what an "honest unbiased person" is like.


Peter Nyikos

[*Hemidactylus*]

Harshman and especially Lawyer Daggett often add value to these venues with
very informative posts. I never see Glenn adding anything of value to the
discussion whatsoever. He subtracts value. This newsgroup is yet another he
haunts. So much for actual paleontology.

What value are you actually adding here or in talk.origins?

[Peter Nyikos]

Harshman is reduced to outright lies below, and you, Glenn, are right on the money at the end.
Harshman can be expected to shove the evidence I talk about below down his memory hole,
and Erik [1] to follow suit. And so, I want you, Glenn, to bear witness to it and refer to it if the occasion calls for it.

[1] and Hemidactylus, who has jumped into a thread he knows next to nothing about.

On Friday, August 5, 2022 at 5:09:14 PM UTC-4, Glenn wrote:
> On Friday, August 5, 2022 at 12:34:32 PM UTC-7, John Harshman wrote:
> > On 8/5/22 11:46 AM, Peter Nyikos wrote:
> > > On Tuesday, August 2, 2022 at 6:27:36 PM UTC-4, John Harshman wrote:
> > >> On 8/2/22 11:14 AM, Peter Nyikos wrote:

> > >>> "troll" is not a very useful word because people seldom bother to
> > >>> say what it means to them.
> > >>>
> > >>> I've found "polemical opportunist" to better describe what Harshman does,
> > >>> because it has a narrower range of meanings. It denotes someone who
> > >>> frequently says things because they sound clever or convincing but
> > >>> don't care if they are true, false, or of unknown veracity.

Harshman immediately gave a strong demonstration of polemical opportunism:

> > >> You have no evidence that I've ever done any such thing,
> > >
> > > I have plenty of evidence, John, going back over more than a decade.
> > > Let's start with one case that occurred less than a month ago, on this very thread:
> > >
> > > "Either you are wrong about this or mammal paleontologists who have done the work are all incompetent."
> > > --https://groups.google.com/g/sci.bio.paleontology/c/WP7JtxgJIio/m/58Cewad9BQAJ
> > > Jul 7, 2022, 10:11:54 PM
> > >
> > > You obviously did not care whether there were any other possibilities,
> > > yet anyone who is competent at thinking scientifically and has one-tenth
> > > of our knowledge could have come up with at least two of them, as I did immediately.
> > > And you had to ackowledge that they were valid alternatives to your false dichotomy.

The closing line is absolutely true, while the preceding three lines are true beyond a reasonable doubt,
and Harshman is powerless to argue otherwise.

> > > And it certainly looks like a case of polemical opportunism: the evidence powerfully suggests that
> > > you purposely chose a highly insulting false dichotomy for its "cleverness"
> > > and didn't give a hoot about coming across as a competent scientist.

Harshman, having painted himself into a corner, just plain lied with:

> > This is all purely in your imagination.

Wait, it gets worse below:

> > >> and in fact I deny ever having done so.
> > >
> > > You are in denial, in the psychologically negative sense of the term.
> > > It continued in subsequent back-and-forth, and your last post in
> > > that sequence was a masterpiece of polemical opportunism.
> > >
> > > I haven't replied to that post yet, but if you keep on being in denial,
> > > I can go through it with a fine toothed comb.

> > Please don't. It will all be your imagined characterization of purely
> > innocent statements.

Even Iago, in Shakespeare's "Othello," wasn't so brazenly self-righteous and insincere.

More about this below, when I address what you wrote at the end.

> >
> > How about paleontology?

The ENTIRE issue about which Harshman told his false dichotomy was
pure paleontology. I gave scientific evidence which he has adamantly avoided
looking at, and has indulged in pure polemic about it.

All of which reinforces my immediate reaction to that false dichotomy:

"If these are the only alternatives you could think of, then you must be suffering from
a long lack of meaningful discussions with research biologists *qua* research biologists."

Harshman has aggressively responded to that conclusion, but hasn't dared to address the conditional clause,
"If these are the only alternatives you could think of," let alone confirm or deny it.


<snip for focus>

> > >>> For instance, Harshman attaches one meaning to the word "insult"
> > >>> when accusing others of it, and another when he is accused of it.
> > >
> > >> This is all from your fertile imagination.
> > >
> > > This formulaic taunt is devoid of reality in this context.


> > I'm afraid you aren't in a position to judge, living within your fertile
> > imagination as you do.

Harshman is bordering on libel with his closing clause. And "I'm afraid you aren't" is his standard
code for "I'd love it if people thought you weren't."

> > > When I give evidence of you behaving in a irresponsible way,
> > > you very frequently label it an "insult" without trying to refute what I wrote.
> > >
> > > In contrast, when you make insulting comments about me, without any evidence, like the one
> > > you have uttered just now, you sometimes retort, "that wasn't an insult, it was an observation."
> > >
> > > Your behavior was even more blatantly hypocritical and irresponsible on the thread
> > > . . Questions about BBC’s “Prehistoric Planet” Episode 1
> > > This was where you wrote such highly insulting things about Glenn as:
> > >
> > > "most of his posts seeming intended to show that "evolutionists" are bad and/or idiots,"
> > > . . --https://groups.google.com/g/sci.bio.paleontology/c/o7a-W7iuTZE/m/XtrSILUmAQAJ
> > > . . Jul 5, 2022, 9:05:48 AM

> > Are those insulting?

Harshman knows HE was insulting, and without credible evidence for his allegation
about you "seeming" to be this way, Glenn.

He is here grasping at straws in trying to divert attention from that fact.

> > Do you disagree? What do you think is the intention
> > behind most of his posts?


> I learned a long time ago that the only person that I would fool is myself.

> You take such things as I say as an insult, yet I am serious. You don't appear to have learned such lessons, and appear not to have grown up. Perhaps, probably, I am too eager at times to consider the reasons, but even then I do confront myself before the first word comes out of my mouth.

Like you, Glenn, I've learned to wait a LONG time before coming to the kind of conclusions
about people that Harshman is all too eager to make about you and me, often without
having a leg to stand on. I waited until late in 2018 to come to an opinion about him that is
close to the one I have now, after almost 8 years of hoping that he will get better.
I gave him every possible benefit of the doubt up to that point, but I finally realized that I was being played for a sucker.

>I don't get defensive, although many seem to be convinced that I do. Invoking "most of my posts" isn't really a wise thing to do, unproductive.

And he has not come up with ONE post as described, never mind "most". Not even after I suggested
he turn to the many people in talk.origins who are as down in the mouth about you as he is.

> The specific instance Peter cites should be the subject of your question. Yet instead of addressing that, you ask leading questions, add some innuendo, and deflect from responding directly to Peter. He is right on the money about insults. I recall times when you have argued that your insults were not insults, or not ad hominem, because they were true.

Were they? Harshman has a dismal track record where that kind of comeback to me is concerned.

> It appears you do not or will not consider others may sincerely feel the same way. If so, why not?

I doubt that Harshman will respond to this, but I will show you on Monday just how right you
are about him.

On Monday, I will reply in detail to the post about which Harshman brazenly lied above with:

"It will all be your imagined characterization of purely innocent statements."

None of it will be imagined; everything will be irrefutable, in a post that you can
remind Harshman about each time he gets too domineering and insulting.
It will also be useful against allies such as Hemidactylus and Simpson when
they play "see no evil, hear no evil, speak no evil" about Harshman.

Peter Nyikos

[*Hemidactylus*]

Peter Nyikos

[snip]

> It will also be useful against allies such as Hemidactylus and Simpson when
> they play "see no evil, hear no evil, speak no evil" about Harshman.
>

I’m playing the sympathy for this newsgroup card now that Glenn is becoming
a permanent fixture, though it was circling the drain already. Quantity is
not quality as can be seen from perpetual death spiral threads on
talk.origins. Lurkers looking here aghast.

I’m also wondering what positive value you happen to bring here as in
justifying your existence. Harshman, Simpson, Harshman. Whatever. Enjoy
Glenn’s presence. I have better things to do than watch the plumbing burst
on a sci. hierarchy group.

[John Harshman]

Get a room.

[John Harshman]

On 8/5/22 2:32 PM, Peter Nyikos wrote:
> On Wednesday, August 3, 2022 at 9:35:10 AM UTC-4, John Harshman wrote:
>> On 8/2/22 11:33 PM, Glenn wrote:
>>> On Tuesday, August 2, 2022 at 7:46:15 PM UTC-7, John Harshman wrote:
>>>> On 8/2/22 7:13 PM, Glenn wrote:
>
>>>>> I recommend taking an objective look, at all claims and the reasoning and language used to support such methodologies.
>>>>>
>>>>> "As a 2012 paper published in Biological Reviews of the Cambridge philosophical Society reported, “Incongruence between phylogenies derived from morphological versus molecular analysis, and between trees based on different subsets of molecular sequences has become pervasive as datasets have expanded rapidly in both characters and species”.
>>>>> Another paper published the following year in the journal Nature, highlighted the extent of the problem. The authors compared 1,070 genes in twenty different yeasts and got 1,070 different trees."
>>>>>
>>>>> https://uncommondescent.com/intelligent-design/at-sci-news-new-interactive-phylogenetic-map-shows-full-diversity-of-life-on-earth/#comment-762264
>
>>>> You really have to stop relying on creationist web sites for your
>>>> science information.
>
> As long as Glenn comes up with scientific statements that you cannot undermine, originally from
> respected scientific sources, you haven't a leg to stand on. What's more, your wording here makes
> you sound like a control freak.

I submit that this is a fine example of Glenn trying to undermine
science by quote-mining at second hand, and from a creationist web site.
Do you disagree?

Do you think that supporting your arguments (well, he doesn't actually
make an argument, does he?) by quoting a creationist web site is a good
thing?

>>> That's your claim, not mine.
>> You're denying that UD is creationist, or what?
>
> Non sequitur. See what I wrote in response just now. I dealt with the bottom line, while talk
> about secondary or tertiary sources is just a diversion.

We could talk about what the actual paper really says if you like. Have
you read it?

>>> "The National Science Foundation (NSF)-funded scientist says it is not unusual for high quality research to report genealogies that conflict with each other over the origins of certain organisms."
>>>
>>> ""We found 1,070 genes, and made 1,070 trees, and each one was different," Rokas says."
>>>
>>> https://phys.org/news/2014-05-reconstructing-tree-life.html
>>>
>>> Is that a creationist site, John? Will you claim that is "junk" or "crap", like Ron does?
>
>> No, the creationist site was Uncommon Descent.
>
> You and Ron O falsely equate ID with creationism. While the majority of the rank and file who
> read UD are probably creationists, I have found lots of the writing of authors to be
> noncommittal about creationism and common descent.

Isn't "nomcommittal" an unsupportable position when it comes to
creationism vs. common descent? Now of course they sometimes fail to
connect the dots directly to creationism. But that article was an
attempt to deny that phylogenetic analyses provide evidence of common
descent. What's your position on that?

> Heck, even Glenn told me, much to my pleasant surprise, that he has no
> problems with common descent.

When did he do that? And what does "no problems" mean, exactly? Why is
he constantly attacking the evidence for common descent? What do you
think was his purpose in bringing up the UD article?

>> What you have there is
>> still not a real scientific source, though, just "science journalism",
>> and they tend toward the sensational. Have you looked at the real papers?
>
> Why should Glenn do that, when you have made a blatantly false claim
> about a link you ADMITTEDLY hadn't even read:
>
> "it seems clear that he's attacking paleontology, especially in
> the second link, which plays up the prevalence of faked (composite) fossils."
>
> This was decisively refuted by me in:
>
> https://groups.google.com/g/sci.bio.paleontology/c/o7a-W7iuTZE/m/MG_AiPnsAAAJ
> Re: Questions about BBC’s “Prehistoric Planet” Episode 1
> Jul 4, 2022, 3:27:50 PM

"Decisively refuted" is your characterization. What do you think he was
trying to do?

>>> Bet you would like to. And you'll think to weasel you way out of it, and essentially say "nothing to see here, move along" or "we got better at it".
>
>
>>>>> 2020:
>>>>>
>>>>> "Incongruence among phylogenetic results has become a common occurrence in analyses of genome-scale data sets."
>>>>>
>>>>> https://academic.oup.com/mbe/article/38/3/1090/5976982
>>>> And I have to question whether you actually read that article.
>>>>> There are many more to consider. Accepting there is "nothing to see here, move along" type of apologetics John practices shouldn't impress an honest unbiased person.
>
>>>> For "honest unbiased person" read "creationist".
>
> More polemical opportunism from you. You've been spoiled rotten by Erik and Oxyaena here
> and by something like a dozen irresponsible people in talk.origins for many years.
> One consequence is that you have, or pretend to have, a completely false picture of what an "honest unbiased person" is like.

I certainly don't find you honest and unbiased. I think you're looking
at Glenn from the perspective of a potential ally against the atheist
Darwinist anti-ID cabal, and that blinds you to what he actually does.

[Peter Nyikos]

You picked just about the worst possible thread make this assertion. You have
even violated one of your alleged standards by naming someone who has never
participated on this thread.

> I never see Glenn adding anything of value to the
> discussion whatsoever.

You chose exactly the right thread in which to see him adding a lot of value,
as I saw when I took the trouble to actually READ two webpages that he linked.

However, you may well choose to blind yourself lest you see it, just to maintain
your perfect track record.

> He subtracts value. This newsgroup is yet another he haunts.
> So much for actual paleontology.

You've ignored everything that went before your bottom-posting.
Perhaps if you had bothered to read things like the following,
you might not have written something not so abysmally clueless:

[my words, repeated from above:]

> > you have made a blatantly false claim
> > about a link you ADMITTEDLY hadn't even read:
> >
> > "it seems clear that he's attacking paleontology, especially in
> > the second link, which plays up the prevalence of faked (composite) fossils."
> >
> > This was decisively refuted by me in:
> >
> > https://groups.google.com/g/sci.bio.paleontology/c/o7a-W7iuTZE/m/MG_AiPnsAAAJ
> > Re: Questions about BBC’s “Prehistoric Planet” Episode 1
> > Jul 4, 2022, 3:27:50 PM

In reality, Harshman was blindly relying on a misleading statement by Erik Simpson
about the following article:

https://fossilcollector.wordpress.com/2015/07/13/fakingit2/
.
The article was about fossils and how they are handled: legitimately by
almost all museums, and illegitimately by dishonest merchants peddling fakes.
It gave some valuable tips in the second half about how to distinguish
between a genuine fossil and a fake, to avoid being bilked.

Perhaps you don't see any value added in knowing about these tips?
Have you never bought a fossil?

More importantly, will you stick by your guns and say that Harshman's
allegation of an "attack on paleontology" added value to the discussion?


I found Glenn's first link highly informative in a direct paleontological sense:

https://www.nature.com/articles/d41586-019-00739-0

The article is primarily about 3-d scans of rare fossils that make it
easy for researchers to study their details without having to make
costly trips. There is, unfortunately, a huge reluctance of most paleontologists
to share these scans that they have made, even for a reasonable fee,
thereby impeding progress in understanding the relationships and lifestyle of the animals involved.

The article gave a fascinating example of "lifestyle" using two views of a 3-d scan of the
skull of the extinct carnivore *Kolponomos*. It concluded that it bit its prey in the same
way as the iconic sabretooth *Smilodon*.

But I didn't just read the article, I went on to learn more about *Kolponomos*
and in the process, greatly updated my understanding about carnivore relationships.
As I told Sight Reader, I learned a lot about the suborder *Arctoidea* from the placement
of *Kolponomos* in it:

"From the name, I had long assumed that Arctoidea only included the bear family Ursidae and the raccoon family
Procyonidae, but it includes Canidae, Mustelidae, the pinnipeds, and various extinct groups like the dog-bears
(Hemicyonidae), but not the bear-dogs (Amphycionidae). Kolponomos belongs to the sister group of the pinnipeds,
it seems. All this was new to me."

https://groups.google.com/g/sci.bio.paleontology/c/o7a-W7iuTZE/m/A38DCXzvAAAJ

Re: Questions about BBC’s “Prehistoric Planet” Episode 1

Jul 4, 2022, 4:13:50 PM

> What value are you actually adding here or in talk.origins?

See above. Try not to be so clueless about threads you have jumped into.

More importantly, when will YOU try to add value to sci.bio.paleontology?


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

PS. As for talk.origins, I'm the resident expert on spheres and related
objects, and even though the subject is off-topic, that hasn't deterred
Robert Carnegie from talking seriously about it.

I have far more respect for Robert than I have for you, so I have
tried to remove all confusion about the various geometrical objects.

[*Hemidactylus*]

So you took a dive and gleaned stuff you found value in. What was Glenn’s
actual part in that process? Outcome in one person and original intent in
another can diverge.

Actually I have long been curious about dog-bear connections. There are
caniforms vs feliforms no? Superficially bears look like oversize dogs, but
looks fan deceive.

>
> https://groups.google.com/g/sci.bio.paleontology/c/o7a-W7iuTZE/m/A38DCXzvAAAJ
> Re: Questions about BBC’s “Prehistoric Planet” Episode 1
> Jul 4, 2022, 4:13:50 PM
>
>
>> What value are you actually adding here or in talk.origins?
>
> See above. Try not to be so clueless about threads you have jumped into.
>
> More importantly, when will YOU try to add value to sci.bio.paleontology?
>
>
> Peter Nyikos
> Professor, Dept. of Mathematics -- standard disclaimer--
> Univ. of South Carolina in Columbia
> http://people.math.sc.edu/nyikos
>
> PS. As for talk.origins, I'm the resident expert on spheres and related
> objects, and even though the subject is off-topic, that hasn't deterred
> Robert Carnegie from talking seriously about it.
>
> I have far more respect for Robert than I have for you, so I have
> tried to remove all confusion about the various geometrical objects.
>

But can you distinguish a bagel from a coffee cup?

[Glenn]

So that it is perfectly clear, that meant John claimed they were true. It is how he has argued in the past. I haven't seen the argument for some time, but he is not the only one that has used that angle.

[Glenn]

You think too highly of John, Peter.

[jillery]

On Fri, 5 Aug 2022 19:38:59 -0700 (PDT), Glenn <GlennS...@msn.com>
wrote:

<snip mindless back-and-forth>

>> > > It appears you do not or will not consider others may sincerely feel the same way. If so, why not?
>> >
>> > I doubt that Harshman will respond to this, but I will show you on Monday just how right you
>> > are about him.
>> >
>> > On Monday, I will reply in detail to the post about which Harshman brazenly lied above with:

The above is notice of the peter preparing for extended spam.

>> > "It will all be your imagined characterization of purely innocent statements."
>> >
>> > None of it will be imagined; everything will be irrefutable, in a post that you can
>> > remind Harshman about each time he gets too domineering and insulting.
>> > It will also be useful against allies such as Hemidactylus and Simpson when
>> > they play "see no evil, hear no evil, speak no evil" about Harshman.
>> Get a room.
>
>You think too highly of John, Peter.

Even if so, that would put you on the bottom of a three-way.

[Glenn]

On Saturday, August 6, 2022 at 5:20:18 AM UTC-7, 69jp...@gmail.com wrote:
> On Fri, 5 Aug 2022 19:38:59 -0700 (PDT), Glenn <GlennS...@msn.com>
> wrote:
>
> <snip mindless back-and-forth>

Uh oh, here comes Booger Queen...

[erik simpson]

On Saturday, August 6, 2022 at 5:20:18 AM UTC-7, 69jp...@gmail.com wrote:

Don't be so negative. Peter is going to document how Glenn's insight and helpfulness enhances all of us in our understanding
of the interrelatedness of not just paleontology but of all the facets of our struggles to understand the universe and our
place in it. Glenn may be slightly embarrassed by the fulsomeness of praise to be bestowed upon him, but he has no need
to be. I suspect Peter my be grooming him for something, God wot what.

[Glenn]

You're an interesting subject in the study of certain origins.

[jillery]

Some people prefrer to be on the bottom, so no negativity implied. And
if what you suggest above is factually correct, Glenn has my deepest
sympathy.

[jillery]

On Sat, 6 Aug 2022 09:53:46 -0700 (PDT), Glenn <GlennS...@msn.com>

wrote:

>On Saturday, August 6, 2022 at 5:20:18 AM UTC-7, 69jp...@gmail.com wrote:
>> On Fri, 5 Aug 2022 19:38:59 -0700 (PDT), Glenn <GlennS...@msn.com>
>> wrote:
>>
>> <snip mindless back-and-forth>
>
>Uh oh, here comes Booger Queen...

Didn't your mommy tell you not to play with yourself while posting?

[erik simpson]

God forbid it being factually correct. I'm at the cusp between being amused and being bored
by all this. I'll check in Monday and see what comes out.

[Peter Nyikos]

Get a life, John. Preferably as a responsible adult.

As I told Erik, it is never too late to learn how to be a responsible adult.
I hope that, before both of you die, one of you at least will learn how to become one.


Peter Nyikos

*********** QUOTE OF THE WEEK *********

If you don't own up to your own elemental truth, falsehood will ultimately end up owning you.

-- https://www.inc.com/lolly-daskal/12-things-people-regret-the-most-before-they-die.html

[Glenn]

“I am the way, the truth, and the life. No one comes to the Father except through Me.”

[Peter Nyikos]

On Friday, August 5, 2022 at 8:35:29 PM UTC-4, John Harshman wrote:
> On 8/5/22 2:32 PM, Peter Nyikos wrote:
> > On Wednesday, August 3, 2022 at 9:35:10 AM UTC-4, John Harshman wrote:
> >> On 8/2/22 11:33 PM, Glenn wrote:
> >>> On Tuesday, August 2, 2022 at 7:46:15 PM UTC-7, John Harshman wrote:
> >>>> On 8/2/22 7:13 PM, Glenn wrote:
> >
> >>>>> I recommend taking an objective look, at all claims and the reasoning and language used to support such methodologies.
> >>>>>
> >>>>> "As a 2012 paper published in Biological Reviews of the Cambridge philosophical Society reported, “Incongruence between phylogenies derived from morphological versus molecular analysis, and between trees based on different subsets of molecular sequences has become pervasive as datasets have expanded rapidly in both characters and species”.
> >>>>> Another paper published the following year in the journal Nature, highlighted the extent of the problem. The authors compared 1,070 genes in twenty different yeasts and got 1,070 different trees."
> >>>>>
> >>>>> https://uncommondescent.com/intelligent-design/at-sci-news-new-interactive-phylogenetic-map-shows-full-diversity-of-life-on-earth/#comment-762264
> >
> >>>> You really have to stop relying on creationist web sites for your
> >>>> science information.
> >
> > As long as Glenn comes up with scientific statements that you cannot undermine, originally from
> > respected scientific sources, you haven't a leg to stand on. What's more, your wording here makes
> > you sound like a control freak.

> I submit that this is a fine example of Glenn trying to undermine
> science by quote-mining at second hand, and from a creationist web site.

Your obsession with painting Glenn as anti-science, without a smidgin of evidence, is becoming pathological.
There is NOTHING in the quotes to suggest any such thing. They are so accurate, you could have
quoted them yourself.

By the way, did you even LOOK at the post he linked? It cites a wonderful resource, rendered all the more valuable
by it being a partial replacement for the pitifully inadequate Tree of Life on Tolweb.

It's a pity it only treats extant organisms, so that it is more a resource for talk.origins than for sci.bio.paleontology.
But it is amazingly comprehensive.


> Do you disagree?

Pathetic, John, really pathetic.

> Do you think that supporting your arguments (well, he doesn't actually
> make an argument, does he?) by quoting a creationist web site is a good
> thing?

Your question is rendered illogical by the fact that Glenn didn't even try to make an argument.
He preceded his quotes with a noncommittal [see meaning below] recommendation, not an argument.

> >>> That's your claim, not mine.
> >> You're denying that UD is creationist, or what?
> >
> > Non sequitur. See what I wrote in response just now. I dealt with the bottom line, while talk
> > about secondary or tertiary sources is just a diversion.
> We could talk about what the actual paper really says if you like. Have
> you read it?
> >>> "The National Science Foundation (NSF)-funded scientist says it is not unusual for high quality research to report genealogies that conflict with each other over the origins of certain organisms."
> >>>
> >>> ""We found 1,070 genes, and made 1,070 trees, and each one was different," Rokas says."
> >>>
> >>> https://phys.org/news/2014-05-reconstructing-tree-life.html

Did you look at this site, John? It explains the experiment behind Glenn's second quote.

> >>> Is that a creationist site, John? Will you claim that is "junk" or "crap", like Ron does?
> >
> >> No, the creationist site was Uncommon Descent.
> >
> > You and Ron O falsely equate ID with creationism. While the majority of the rank and file who
> > read UD are probably creationists, I have found lots of the writing of authors to be
> > noncommittal about creationism and common descent.

> Isn't "nomcommittal" an unsupportable position when it comes to
> creationism vs. common descent?

Nonsense. By "noncommittal" I mean, in this case, that there is no mention of the whole issue
in a science-based article.

That is why the canard that ID is a "creationist pseudo-science" is so prevalent: it is a dishonest
way of evading scientific arguments. You see it whenever people jeer at Directed Panspermia
without trying to undermine it: it is a purely scientific way of bringing the issue of
ID into the beginning of life on earth.

> Now of course they sometimes fail to
> connect the dots directly to creationism. But that article was an
> attempt to deny that phylogenetic analyses provide evidence of common
> descent.

That's your highly biased reading of the article. Try quoting it and DEFENDING your allegation.

This would be a novelty for you: you have yet to defend any of the "mind-reading" allegations
you've made of anything Glenn quoted on this thread.


To be concluded, possibly today, and certainly this week.

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

University of So. Carolina in Columbia
http://people.math.sc.edu/nyikos

******************** QUOTE OF THE WEEK ****************

[Glenn]

It wouldn't have mattered. He'd claim quote mining till the cows came home.

[Peter Nyikos]

On Tuesday, August 2, 2022 at 7:52:28 PM UTC-4, erik simpson wrote:
> On Tuesday, August 2, 2022 at 4:46:22 PM UTC-7, Trolidan7 wrote:
>

> ...

> > I think it might be a good idea to bow down and out of this
> > one and insult myself if needed. I will try to look it up
> > however.

> ....
> Wise choice. It's always hard to follow exchanges like this, and trust me, if you
> participate it's inevitable that you will be insulted.

There is no reason to trust you on this, unless you know more about Trolidan7 than
I do, and are itching for a chance to insult him, or you know that John is.


With your staunch ally Oxyaena long gone, only the two of you come into consideration here [1].
I remember well how Oxyaena insulted both Daud and Mario because they made clueless comments
about paleontology, but they just laughed it off, while I've tried to steer them on a proper course in a civil fashion.

[1] It is possible that Glenn may some day decide Trolidan7 is an outspoken atheist,
like John, but right now I am going on the evidence I have. As for your buddies
jillery and Hemidactylus, I don't think you know whether they are inclined to insult Trolidan 7 or not.


I do not insult people, not even in the self-centered sense in which John uses the word
"insult," unless they behave very irresponsibly. And I have seen no sign of Trolidan7
behaving irresponsibly.

I have said many times, both here and in talk.origins: I suffer fools gladly, but knaves with difficulty or not at all.
You and Harshman are just about the opposite.

Peter Nyikos

*********** QUOTE OF THE WEEK *********

If you don't own up to your own elemental truth, falsehood will ultimately end up owning you.

-- https://www.inc.com/lolly-daskal/12-things-people-regret-the-most-before-they-die.html

The above, plus the sentence "Honesty is the clearest path." was under item number 6 of the 12.

[Peter Nyikos]

I don't know which "actual paper" Harshman is referring to, do you, Glenn?

If it is the one you link below, I've read enough of it to be able to discuss it,
but I don't think Harshman will be happy if I do.

> > > >>> "The National Science Foundation (NSF)-funded scientist says it is not unusual for high quality research to report genealogies that conflict with each other over the origins of certain organisms."
> > > >>>
> > > >>> ""We found 1,070 genes, and made 1,070 trees, and each one was different," Rokas says."
> > > >>>
> > > >>> https://phys.org/news/2014-05-reconstructing-tree-life.html

> > Did you look at this site, John? It explains the experiment behind Glenn's second quote.

> It wouldn't have mattered. He'd claim quote mining till the cows came home.

I'm postponing my fine-toothed-comb analysis of his masterpiece of polemical opportunism until tomorrow.
Duty calls.

Meanwhile, you might want to see the post to which I am referring. It was done on this thread on Jul 14, 2022, 11:00:55 PM
In between, I was gone to participate in two intense research conferences.

Peter Nyikos

[John Harshman]

On 8/8/22 10:28 AM, Peter Nyikos wrote:
> On Friday, August 5, 2022 at 8:35:29 PM UTC-4, John Harshman wrote:
>> On 8/5/22 2:32 PM, Peter Nyikos wrote:
>>> On Wednesday, August 3, 2022 at 9:35:10 AM UTC-4, John Harshman wrote:
>>>> On 8/2/22 11:33 PM, Glenn wrote:
>>>>> On Tuesday, August 2, 2022 at 7:46:15 PM UTC-7, John Harshman wrote:
>>>>>> On 8/2/22 7:13 PM, Glenn wrote:
>>>
>>>>>>> I recommend taking an objective look, at all claims and the reasoning and language used to support such methodologies.
>>>>>>>
>>>>>>> "As a 2012 paper published in Biological Reviews of the Cambridge philosophical Society reported, “Incongruence between phylogenies derived from morphological versus molecular analysis, and between trees based on different subsets of molecular sequences has become pervasive as datasets have expanded rapidly in both characters and species”.
>>>>>>> Another paper published the following year in the journal Nature, highlighted the extent of the problem. The authors compared 1,070 genes in twenty different yeasts and got 1,070 different trees."
>>>>>>>
>>>>>>> https://uncommondescent.com/intelligent-design/at-sci-news-new-interactive-phylogenetic-map-shows-full-diversity-of-life-on-earth/#comment-762264
>>>
>>>>>> You really have to stop relying on creationist web sites for your
>>>>>> science information.
>>>
>>> As long as Glenn comes up with scientific statements that you cannot undermine, originally from
>>> respected scientific sources, you haven't a leg to stand on. What's more, your wording here makes
>>> you sound like a control freak.
>
>> I submit that this is a fine example of Glenn trying to undermine
>> science by quote-mining at second hand, and from a creationist web site.
>
> Your obsession with painting Glenn as anti-science, without a smidgin of evidence, is becoming pathological.
> There is NOTHING in the quotes to suggest any such thing. They are so accurate, you could have
> quoted them yourself.

The question to ask is why Glenn is quoting them. He never says, so you
have to come up with your own explanations. What do you come up with?

> By the way, did you even LOOK at the post he linked? It cites a wonderful resource, rendered all the more valuable
> by it being a partial replacement for the pitifully inadequate Tree of Life on Tolweb.
>
> It's a pity it only treats extant organisms, so that it is more a resource for talk.origins than for sci.bio.paleontology.
> But it is amazingly comprehensive.

I'm not thrilled with it. It has nice pictures but the display is
confusing and doesn't make a good substitute for a tree. Also, I find
that the bird part leaves out (or makes invisible to me??) many small
orders and families.

>> Do you disagree?
>
> Pathetic, John, really pathetic.

I presume that by that you mean that you disagree. But it would be good
to clarify.

>> Do you think that supporting your arguments (well, he doesn't actually
>> make an argument, does he?) by quoting a creationist web site is a good
>> thing?
>
> Your question is rendered illogical by the fact that Glenn didn't even try to make an argument.
> He preceded his quotes with a noncommittal [see meaning below] recommendation, not an argument.

Was he making that recommendation for no purpose whatsoever? Do you
think creationist web sites are, in general, good to recommend here?

>>>>> That's your claim, not mine.
>>>> You're denying that UD is creationist, or what?
>>>
>>> Non sequitur. See what I wrote in response just now. I dealt with the bottom line, while talk
>>> about secondary or tertiary sources is just a diversion.
>> We could talk about what the actual paper really says if you like. Have
>> you read it?
>>>>> "The National Science Foundation (NSF)-funded scientist says it is not unusual for high quality research to report genealogies that conflict with each other over the origins of certain organisms."
>>>>>
>>>>> ""We found 1,070 genes, and made 1,070 trees, and each one was different," Rokas says."
>>>>>
>>>>> https://phys.org/news/2014-05-reconstructing-tree-life.html
>
> Did you look at this site, John? It explains the experiment behind Glenn's second quote.

Yes. Do you think it means what Glenn thinks it means? For that matter,
what does Glenn think it means?

>>>>> Is that a creationist site, John? Will you claim that is "junk" or "crap", like Ron does?
>>>
>>>> No, the creationist site was Uncommon Descent.
>>>
>>> You and Ron O falsely equate ID with creationism. While the majority of the rank and file who
>>> read UD are probably creationists, I have found lots of the writing of authors to be
>>> noncommittal about creationism and common descent.
>
>> Isn't "nomcommittal" an unsupportable position when it comes to
>> creationism vs. common descent?
>
> Nonsense. By "noncommittal" I mean, in this case, that there is no mention of the whole issue
> in a science-based article.

Oh. That isn't what "noncommittal" means. But isn't there a pattern to
what UD picks for "science-based" articles? Can you spot that pattern?

> That is why the canard that ID is a "creationist pseudo-science" is so prevalent: it is a dishonest
> way of evading scientific arguments. You see it whenever people jeer at Directed Panspermia
> without trying to undermine it: it is a purely scientific way of bringing the issue of
> ID into the beginning of life on earth.

Let's not go further into that particular one of your obsessions. But
one common theme at UD is casting doubt on common descent by seizing on
anything that they think discredits the tree of life. Had you noticed that?

> > Now of course they sometimes fail to
>> connect the dots directly to creationism. But that article was an
>> attempt to deny that phylogenetic analyses provide evidence of common
>> descent.
>
> That's your highly biased reading of the article. Try quoting it and DEFENDING your allegation.

What's your reading? Why was it considered suitable for UD?

> This would be a novelty for you: you have yet to defend any of the "mind-reading" allegations
> you've made of anything Glenn quoted on this thread.

Since Glenn never says, mind-reading is the only way to figure out why
he posted. Or one could try to spot the pattern in his postings. He
posts things that appear (often based on quote-mining) to show that
"evolutionists" are wrong about stuff. His use of Rokas et al. was an
attempt to show that phylogenetic analysis doesn't work, because ever
gene produced a different tree. Of course if you read, you will find
that the trees are only slightly different, and there are reasons given
for the slight differences that do not invalidate phylogenetic analysis.
Would you agree?

[Peter Nyikos]

On Monday, August 8, 2022 at 9:25:14 PM UTC-4, John Harshman wrote:
> On 8/8/22 10:28 AM, Peter Nyikos wrote:
> > On Friday, August 5, 2022 at 8:35:29 PM UTC-4, John Harshman wrote:
> >> On 8/5/22 2:32 PM, Peter Nyikos wrote:
> >>> On Wednesday, August 3, 2022 at 9:35:10 AM UTC-4, John Harshman wrote:
> >>>> On 8/2/22 11:33 PM, Glenn wrote:
> >>>>> On Tuesday, August 2, 2022 at 7:46:15 PM UTC-7, John Harshman wrote:
> >>>>>> On 8/2/22 7:13 PM, Glenn wrote:
> >>>
> >>>>>>> I recommend taking an objective look, at all claims and the reasoning and language used to support such methodologies.
> >>>>>>>
> >>>>>>> "As a 2012 paper published in Biological Reviews of the Cambridge philosophical Society reported, “Incongruence between phylogenies derived from morphological versus molecular analysis, and between trees based on different subsets of molecular sequences has become pervasive as datasets have expanded rapidly in both characters and species”.
> >>>>>>> Another paper published the following year in the journal Nature, highlighted the extent of the problem. The authors compared 1,070 genes in twenty different yeasts and got 1,070 different trees."
> >>>>>>>
> >>>>>>> https://uncommondescent.com/intelligent-design/at-sci-news-new-interactive-phylogenetic-map-shows-full-diversity-of-life-on-earth/#comment-762264
> >>>
> >>>>>> You really have to stop relying on creationist web sites for your
> >>>>>> science information.
> >>>
> >>> As long as Glenn comes up with scientific statements that you cannot undermine, originally from
> >>> respected scientific sources, you haven't a leg to stand on. What's more, your wording here makes
> >>> you sound like a control freak.
> >
> >> I submit that this is a fine example of Glenn trying to undermine
> >> science by quote-mining at second hand, and from a creationist web site.
> >
> > Your obsession with painting Glenn as anti-science, without a smidgin of evidence, is becoming pathological.
> > There is NOTHING in the quotes to suggest any such thing. They are so accurate, you could have
> > quoted them yourself.

> The question to ask is why Glenn is quoting them. He never says, so you
> have to come up with your own explanations. What do you come up with?

I don't come up with wild paranoid conclusions like you do.


As long as you keep feeding your obsession, I might as well ask:

Were you ever sincere when you said that you have suspicions of me
being a creationist?

Do you still claim to have such suspicions?

If you do, you are either irresponsibly insincere about your true suspicions,
or you are paranoid, and far more deeply into a fantasy world than any sane
person would think I am, given the context of your accusations that I live in one.


Remainder deleted, pending a reply to the above.


Peter Nyikos

[John Harshman]

Can you come up with any conclusions? Any hypotheses?

> As long as you keep feeding your obsession, I might as well ask:
>
> Were you ever sincere when you said that you have suspicions of me
> being a creationist?

I don't recall the circumstances of whatever happened, but if I said I
had suspicions I was sincere.

> Do you still claim to have such suspicions?

No, assuming I ever did.

[Glenn]

He's asked me that before. Don't ask why, the world ended the day he lost my love.
Skeeter:
https://www.youtube.com/watch?v=NZ5WeXtOacU

>
> As long as you keep feeding your obsession, I might as well ask:
>
> Were you ever sincere when you said that you have suspicions of me
> being a creationist?
>
> Do you still claim to have such suspicions?
>
> If you do, you are either irresponsibly insincere about your true suspicions,
> or you are paranoid, and far more deeply into a fantasy world than any sane
> person would think I am, given the context of your accusations that I live in one.
>
>
> Remainder deleted, pending a reply to the above.
>

You really expect an honest answer?

From the Nature paper
https://www.nature.com/articles/nature12130

"These results question the exclusive reliance on concatenation and associated practices, and argue that selecting genes with strong phylogenetic signals and demonstrating the absence of significant incongruence are essential for accurately reconstructing ancient divergences."

Your mileage may vary, John's will never run out of mileage, but this appears to me to be a little too obvious. Just choose genes that fit your belief and all will be well. And the comment about belief is being generous and assumes honesty.

[John Harshman]

Needless to say, "just chooses genes to fit your belief" is not an
accurate characterization of what that quote says or what the paper
shows. But could you make your point explicitly?

[Glenn]

Could you be more specific?

[Glenn]

I'm not sure what you mean, and you don't say why you ask, or why you post at all.

But perhaps this will help -

"The argument is that even in a case where we all agree common ancestry is true (i.e., within humans), we don’t necessarily find a tree-like dataset. This is supposed to get common descent off the hook, so it isn’t falsified or challenged in the numerous other cases where the data isn’t tree-like."

https://evolutionnews.org/2022/08/yes-winston-ewerts-dependency-graph-is-a-real-model/

Is Emily just Joshing us? And is "dataset" a proper English word?