Response to Colin Patterson "Quote:" Need Help
K-Man
"Tell me one thing that you _know_ about evolution." I've searched the
TO archive and googled the topic, and I've come up with some good
info--including the Patterson Misquoted A Tale of Two 'Cites'
article--but I don't think I have enough, or understand enough, to do a
rebuttal that will be convincing to a confirmed creationist. (Impossible
task, I know).
My questions:
1) Did Patterson really say, "One morning I woke up and something had
happened in the night, and it struck me that I had been working on this
stuff for 20 years and there was not one thing I knew about it."? If
so, what's the full context for this statement? What "stuff" was he
talking about.
2) Did Patterson really ask "Can you tell me anything you know about
evolution, any one thing, any one thing that is true?" Again, if so,
what was the context?
3) Did someone in the audience really say "I do know one thing - it
ought not to be taught in high school." Again, if so, what was the
context? What was he or she saying shouldn't be taught in high school.
4) What is going to be the best way to present this information to a
staunch creationist?
TIA
Ken
Thomas H. Faller
<snip of questions>
>4) What is going to be the best way to present this information to a
>staunch creationist?
>TIA
Inscribed in the head of a baseball bat.
Tom Faller
syvanen
Patterson was a fanatical cladist and according to the strict set of
rules that this sect obeys, he realized one morning he had nothing to
say about evolution. He was right. But don't get bent out of shape
about it, these guys are real extremists of the scientific world and
even if they are unable to reconstruct evolutionary histories, it
doesn't mean it can't be done.\
Mike Syvanen
catshark
Yes (more or less). I found the text while doing a little research into
the quote mines in _Darwin on Trial_. You can read it in full here:
<http://www.linnean.org/html/publications/publications_lin_04_02.htm>
Keep in mind Patterson seems to be one of those people who enjoy stirring
the hornet's nest and then acts surprised when he gets stung. Oh, well, to
each . . .
I was going to get around to asking someone who understands cladistics in
depth to see what he was really talking about, but keep in mind what he
said at the beginning of his talk, as follows:
Well, I'm not interested in the controversy over high school
teaching, and if any militant creationists have come here
looking for political ammunition, I hope they will be dis-
appointed. As an aside, I think the high school evolution-
creation controversy is easily solved - all you need is an
established religion, which is automatically taught in schools
as the Church of England is, and creationists have no ground
for complaint. But it's 200 years too late for that solution
here. Anyway, I'm not talking about that controversy - this
is a systematics discussion group, and I shall talk about
evolutionism and creationism as they apply to systematics.
And since it's a discussion group, I only want to be outrageous
enough to get a discussion going.
>
>4) What is going to be the best way to present this information to a
>staunch creationist?
How about:
"Patterson ain't Moses, much less god . . . maybe he was right or
maybe he was wrong, I don't much care which. He is just *one*
scientist, just one *man*.
"But, if *you* want to discuss whether he was right or wrong,
you're gonna have to read more than a few books on the *science*
of evolution. If you want me to help, I'll be glad to.
---------------
J. Pieret
---------------
How extremely stupid not to have thought of that!
- Thomas Henry Huxley -
catshark
wrote:
>On Thu, 28 Aug 2003 18:55:54 +0000 (UTC), K-Man <k-...@nospam.com> wrote:
[First attempt to post this is overdue and presumed lost. My apologies
should the missing return from the dead to make for a duplicate]
Oh, and just as an interesting bit of serendipity, Patterson says:
Instead, I shall take my texts from this book - Gillespie's
"Charles Darwin and the problem of Creation " [1979 - a first
rate book, and I want to consider the ways in which two
alternative world views - creationism and evolutionism -
have affected or might influence systematists and systematics].
I was in a used book store earlier today and picked up a fairly expensive
but intriguing-looking book. It was not until after I posted the previous
response to K-Man that I realized that it was the same book that Patterson
cited.
As the IDers would say . . . spooky!
---------------
J. Pieret
---------------
Science, as a practice, process, or institution,
has no metaphysics other than the assumption
that if you can measure it you can study it . . .
- John Wilkins -
John Wilkins
> On Fri, 29 Aug 2003 02:06:42 +0000 (UTC), catshark <cats...@yahoo.com>
> wrote:
>
> >On Thu, 28 Aug 2003 18:55:54 +0000 (UTC), K-Man <k-...@nospam.com> wrote:
>
>
> [First attempt to post this is overdue and presumed lost. My apologies
> should the missing return from the dead to make for a duplicate]
>
>
> Oh, and just as an interesting bit of serendipity, Patterson says:
>
> Instead, I shall take my texts from this book - Gillespie's
> "Charles Darwin and the problem of Creation " [1979 - a first
> rate book, and I want to consider the ways in which two
> alternative world views - creationism and evolutionism -
> have affected or might influence systematists and systematics].
>
> I was in a used book store earlier today and picked up a fairly expensive
> but intriguing-looking book. It was not until after I posted the previous
> response to K-Man that I realized that it was the same book that Patterson
> cited.
Gillispie? Same guy who did
Gillispie, Charles Coulston. 1959. Genesis and geology: a study in the
relations of scientific thought, natural theology, and social opinion in
Great Britain, 1790-1850. New York: Harper Torchbooks / The Cloister
Library. Original edition, 1951?
>
> As the IDers would say . . . spooky!
>
> ---------------
> J. Pieret
> ---------------
>
> Science, as a practice, process, or institution,
> has no metaphysics other than the assumption
> that if you can measure it you can study it . . .
>
> - John Wilkins -
>
>
--
John Wilkins
DARK IN HERE, ISN'T IT?
wilkins.id.au
John Harshman
syvanen wrote:
That's a severe misunderstanding of Patterson and of cladistics.
I don't know what Patterson said at that meeting, nor do I know what the
immediate context of whatever he did say might have been. But I do
know that Patterson's views were no comfort to creationists.
catshark
Wilkins) wrote:
>catshark <cats...@yahoo.com> wrote:
>
>> On Fri, 29 Aug 2003 02:06:42 +0000 (UTC), catshark <cats...@yahoo.com>
>> wrote:
>>
>> >On Thu, 28 Aug 2003 18:55:54 +0000 (UTC), K-Man <k-...@nospam.com> wrote:
>>
>>
>> [First attempt to post this is overdue and presumed lost. My apologies
>> should the missing return from the dead to make for a duplicate]
>>
>>
>> Oh, and just as an interesting bit of serendipity, Patterson says:
>>
>> Instead, I shall take my texts from this book - Gillespie's
>> "Charles Darwin and the problem of Creation " [1979 - a first
>> rate book, and I want to consider the ways in which two
>> alternative world views - creationism and evolutionism -
>> have affected or might influence systematists and systematics].
>>
>> I was in a used book store earlier today and picked up a fairly expensive
>> but intriguing-looking book. It was not until after I posted the previous
>> response to K-Man that I realized that it was the same book that Patterson
>> cited.
>
>Gillispie? Same guy who did
>
>Gillispie, Charles Coulston. 1959. Genesis and geology: a study in the
>relations of scientific thought, natural theology, and social opinion in
>Great Britain, 1790-1850. New York: Harper Torchbooks / The Cloister
>Library. Original edition, 1951?
Nope, Neal C. Gillespie, a professor of history at Georgia State University
(as of 1979). Other than _Charles Darwin and the problem of Creation_, the
only other book by him listed on the cover or in Amazon is: _The collapse
of orthodoxy; the intellectual ordeal of George Frederick Holmes_. Both
books are out-of-print.
It isn't all that much of a book, though, more like a monograph . . . only
201 pages, *including* 54 pages of notes, bibliography and index, in not
particularly dense typeface. I hesitated for a long time because it was
marked $12.50 (US), almost as much as most trade paperbacks (though it is
hardback and in good condition). A tad more luck: either the clerk misread
the price or gave me a "I've seen your face a *lot*" discount and I got it
for $10.
---------------
J. Pieret
---------------
Cogito sum, ergo sum, cogito.
- Robert Carroll -
John Wilkins
And note that he is explicitly restricting the discussion to
systematics. This is the dispute:
1. "Evolutionary systematics" of around 1940- assumes that we can work
out from our knowledge of how organisms evolve what the ancestors must
have looked like, and so identify them in the fossil record.
2. In the 1960s, the fashion is all "theory-free classification" so a
school that attempts to classify species on the basis of any old
characters grouping together in a graph, called "phenetics" by its
detractors, is promoted. Because it uses computers a lot to process the
data, it gets called "numerical taxonomy".
3. In 1966 an English edition of a work by Willi Hennig, an East German
entomologist, is published. Hennig argues that neither evolutionary
systematics nor phenetics tells us much about ancestors, and proposes
instead a formal method to reconstruct phylogenetic trees from modern
characters. This gets called "phylogenetic systematics", or "cladism" by
its detractors. By about 1974, phenetics is dead and phylogenetics is
the go.
4. Ernst Mayr, a leading evolutionary systematist, begins a campaign
against cladistics because it conflicts with his intuitions about what
happens in evolution. He is joined by Peter Ashlock, and battle is
joined. Mayr uses his political and personal influence to attack
cladistics wherever he can.
5. A split in the cladistic camp arises between the ordinary variety
(known more recently as "process cladists") and a variety that calls
themselves "transformed cladists", who later become known as "pattern
cladists". One of the originators of this view is Colin Patterson, who,
along with Donn Rosen, Gareth Nelson and Norman Platnick attempt to show
that the trees formed by cladistic techniques are not histories but
relationships between sets of characters, and hence are abstractions.
6. Process cladists, led mainly by the acid pen of Jamie Farris, attack
pattern cladists as being non-Darwinian and anti-evolutionary. Mayr
attacks them as being essentialists, typologists, and idealists, and
pre-Darwinian. Pattern cladists respond by referring to both schools as
the "Darwinians", since both assume the mantle and authority of Darwin
to justify their own views.
7. The debate gets sidelined on matters of epistemology, with all sides
claiming they are being most Popperian. Process cladism gets taught
anyway, and pattern cladism falls by the wayside. Evolutionary, or as it
is sometimes called, "eclectic", systematics is a strong but
increasingly less successful school of thought.
In this context, what Patterson had to say becomes obvious - he is not
denying evolution *happened*, only that we can know what happened in
systematics. Unfortunately both he and Nelson are very quick with a
phrase capable of being taken out of this complex context and misused.
And so it has been, just as similar problems with punctuated equilibrium
were taken out of context by creationists. Film at 11.
For the record I am highly sympathetic to pattern cladism and not merely
because Gary Nelson is my supervisor (and in recent years something of a
mentor), but because I was convinced by the logic of the arguments. gary
has actually never discussed it with me, although I am sure he would if
I asked (when the thesis is done...).
I think that assuming evolution *has* occurred but that we aren't given
a chronicle before we start doing paleontology and all the other
disciplines that investigate the past of life, we can only say that some
sequence is more likely than some other, in a Bayesian sort of way, and
there's the rub: what priors do we assume to make the likelihoods
useful? the obvious, and unhelpful, answer is, we use the best knowledge
we now have to assess new claims. Of course, if our priors are wrong, we
will assess new claims wrongly. Moreover, *how* do we test new claims of
ancestry and history? The pattern cladist account is - by using the
*known* data sets and the relations they evidence. In short, we test
phylogenies with well-attested cladograms. Hence, cladograms are not, in
themselves, phylogenies.
So, in Patterson's words, what do we *know* about the evolutionary
history of any group of organisms? In the most rigid sense of "to know"
we know very little indeed. He and the other pattern cladists overstated
this, I believe, but the point is firm. And to date, *nobody* of any
school has managed to resolve it.
>
> >
> >4) What is going to be the best way to present this information to a
> >staunch creationist?
>
> How about:
>
> "Patterson ain't Moses, much less god . . . maybe he was right or
> maybe he was wrong, I don't much care which. He is just *one*
> scientist, just one *man*.
>
> "But, if *you* want to discuss whether he was right or wrong,
> you're gonna have to read more than a few books on the *science*
> of evolution. If you want me to help, I'll be glad to.
or, "you have to read up on systematics. Email Wilkins and he'll give
you a reading list, or if you want to get a less-partisan view, John
Harshman".
catshark
wrote:
>On Thu, 28 Aug 2003 18:55:54 +0000 (UTC), K-Man <k-...@nospam.com> wrote:
Oh, and just as an interesting bit of serendipity, Patterson says:
Instead, I shall take my texts from this book - Gillespie's
"Charles Darwin and the problem of Creation " [1979 - a first
rate book, and I want to consider the ways in which two
alternative world views - creationism and evolutionism -
have affected or might influence systematists and systematics].
I was in a used book store earlier today and picked up a fairly expensive
but intriguing-looking book. It was not until after I posted the previous
response to K-Man that I realized that it was the same book that Patterson
cited.
As the IDers would say . . . spooky!
John Wilkins
With respect Mike, that's just stupid and inflammatory. Also false.
Patterson was a pattern cladist. These are cladists who recognise that
the nodes in a cladogram are not automatically ancestors. If anyone
thinks that this is fanatical, they need to investigate some
mathematical work on minimum message length encoding.
I'm not going to get into a debate over pattern cladism except to note
that even "*non*-fanatical" cladists such as Elliot Sober note the
difficulty in reconstructing the past on the basis on modern
relationships. Information is lost through insufficient data, horizontal
transfer, homoplasy and noise. *Process* cladists like Farris are
perhaps fanatical in that they denigrate anyone who does not think
cladism is the magic method for finding out the past. Pattern cladists
have been intemperate in their claims about evolution and
"neo-Darwinism" and the like, but they have a very real, and so far
unanswered, point - cladograms are arrangements of datasets, not
histories, and anyone who thinks a cladogram is a history in and of
itself is reinventing Haeckel's recapitulation theory.
catshark
Thanks, John, for the clear exposition of what is going on in Patterson's
talk (with it as a guide, I was able to reread it with maybe 85%
understanding this time ;-). This is going in my Quote Mine file for when
we get the original up and start expanding it. (Ok, I have a long weekend
and I promise I'll make a major push to get that finally ready!)
>
>And note that he is explicitly restricting the discussion to
>systematics. This is the dispute:
>
>1. "Evolutionary systematics" of around 1940- assumes that we can work
>out from our knowledge of how organisms evolve what the ancestors must
>have looked like, and so identify them in the fossil record.
For those of us <cough> who are a little less versed in the finer points of
evolutionary theory (and academic jargon), definitions may help. Are these
adequate?:
After Darwin, it was realized that organisms could indeed change.
Scientists began to construct _phylogenies_, lists or diagrams that
showed the evolutionary paths taken by populations of organisms
through many generations and over long periods of time. These
phylogenetic diagrams quickly started to look a bit like a family
tree, showing who the nearest relatives were and who shared a common
ancestor, and when.
Two main types of taxonomy have arisen from this work. They look
alike, but there are serious differences:
_Evolutionary systematics_ makes an attempt to construct trees that
accurately show phyletic lineages (proper branching on the family
tree), along with a consideration of when and how new species arose
and moved into new habitats and niches (established a 'new' way of
life as opposed to some trivial character change).
_Cladistics_, however, ignores when and where a branch occurs, tries
to use purely objective criteria, and defines each branch point by a
fundamental character of evolutionary significance. Both methods have
their strengths and weaknesses.
[Taken from: <http://www.brooklyn.cuny.edu/bc/ahp/CLAS/CLAS.Clad.html>]
And a clarification, please: Patterson also seems, in several places, to
use "systematics" (without "evolutionary" tacked on) in a broader sense, to
include *both* taxonomic methodologies, or am I misreading him?
That last is your answer to the problem, not necessarily Patterson's,
right?
>Of course, if our priors are wrong, we
>will assess new claims wrongly. Moreover, *how* do we test new claims of
>ancestry and history? The pattern cladist account is - by using the
>*known* data sets and the relations they evidence. In short, we test
>phylogenies with well-attested cladograms. Hence, cladograms are not, in
>themselves, phylogenies.
Is this what Patterson is getting at when he says that the 'theory of
evolution has got between scientists and the data'?
>
>So, in Patterson's words, what do we *know* about the evolutionary
>history of any group of organisms? In the most rigid sense of "to know"
>we know very little indeed. He and the other pattern cladists overstated
>this, I believe, but the point is firm. And to date, *nobody* of any
>school has managed to resolve it.
But the IDiots have: 'Throw it out, bassinet, bathwater, baby and all.'
>>
>> >
>> >4) What is going to be the best way to present this information to a
>> >staunch creationist?
>>
>> How about:
>>
>> "Patterson ain't Moses, much less god . . . maybe he was right or
>> maybe he was wrong, I don't much care which. He is just *one*
>> scientist, just one *man*.
>>
>> "But, if *you* want to discuss whether he was right or wrong,
>> you're gonna have to read more than a few books on the *science*
>> of evolution. If you want me to help, I'll be glad to.
>
>or, "you have to read up on systematics. Email Wilkins and he'll give
>you a reading list,
But be careful what you wish for. ;-)
>or if you want to get a less-partisan view, John
>Harshman".
---------------
Andrew Criddle
One can certainly perform cladistic systematics without overt
reference to evolution, and there may well be important advantages
to limiting the reliance of cladistic analysis on particular
evolutionary models.
I have difficulties in seeing how one can justify doing cladistics
at all, instead of say phenetics/"numerical taxonomy", without
a prior belief in some form of common descent.
Andrew Criddle
John Harshman
John Wilkins wrote:
And Patterson's expressed reason for not considering cladograms to be
evolutionary trees *a priori* was that you could then turn around and
use the congruence of data represented by the cladogram as evidence for
evolution/common descent without the danger of circular argument.
(My position, for what it's worth: cladograms are *estimates* of
history. If they aren't, we have no reason to care about them.)
John Harshman
John Wilkins wrote:
> catshark <cats...@yahoo.com> wrote:
>
>
>>On Thu, 28 Aug 2003 18:55:54 +0000 (UTC), K-Man <k-...@nospam.com> wrote:
>>
>>
>>>This past weekend my pastor repeated the old Colin Patterson quote re
>>>"Tell me one thing that you _know_ about evolution." I've searched the
>>>TO archive and googled the topic, and I've come up with some good
>>>info--including the Patterson Misquoted A Tale of Two 'Cites'
>>>article--but I don't think I have enough, or understand enough, to do a
>>>rebuttal that will be convincing to a confirmed creationist. (Impossible
>>>task, I know).
>>>
>>>My questions:
>>>
>>>1) Did Patterson really say, "One morning I woke up and something had
>>>happened in the night, and it struck me that I had been working on this
>>>stuff for 20 years and there was not one thing I knew about it."? If
>>>so, what's the full context for this statement? What "stuff" was he
>>>talking about.
[snip]
> So, in Patterson's words, what do we *know* about the evolutionary
> history of any group of organisms? In the most rigid sense of "to know"
> we know very little indeed. He and the other pattern cladists overstated
> this, I believe, but the point is firm. And to date, *nobody* of any
> school has managed to resolve it.
I don't see this as being any different from anything else in science.
We "know" some things about evolution as much as science can claim to
"know" anything. There being no absolute way to establish truth, we must
be content to have at least implied quotes around the word. But with
that proviso, we know plenty, at least about the branching pattern of
life, and are learning more all the time. Guinea pigs *are* rodents.
Despite Patterson's diatribe, that's equivalent to saying they are
descended from rodents, and it is a meaningful, and true, statement
about evolution.
(I found it particularly unlikely that Jim Hopson, mentioned
specifically in Patterson's speech, would have been unable to come up
with a bunch of other things we know. One wonders if the question was at
all clear.)
>>>4) What is going to be the best way to present this information to a
>>>staunch creationist?
>>>
>>How about:
>>
>> "Patterson ain't Moses, much less god . . . maybe he was right or
>> maybe he was wrong, I don't much care which. He is just *one*
>> scientist, just one *man*.
>>
>> "But, if *you* want to discuss whether he was right or wrong,
>> you're gonna have to read more than a few books on the *science*
>> of evolution. If you want me to help, I'll be glad to.
>>
>
> or, "you have to read up on systematics. Email Wilkins and he'll give
> you a reading list, or if you want to get a less-partisan view, John
> Harshman".
And the quoted diatribe
(<http://www.linnean.org/html/publications/publications_lin_04_02.htm>)
is Patterson being controversial and combative, and is not a fair
representation of his views. How Patterson's own statements can be a
mis-statement of his views (or at least easily interpreted that way)
would be an interesting study in contrarian psychology, and is a trait
he shares with Gareth Nelson. But I heard him speak when he wasn't being
weird and combative, and he was much more reasonable.
John Harshman
Andrew Criddle wrote:
> wil...@wehi.edu.au (John Wilkins) wrote in message news:<1g0gget.zsby1frxv5oN%wil...@wehi.edu.au>...
[snip]
>>
>>So, in Patterson's words, what do we *know* about the evolutionary
>>history of any group of organisms? In the most rigid sense of "to know"
>>we know very little indeed. He and the other pattern cladists overstated
>>this, I believe, but the point is firm. And to date, *nobody* of any
>>school has managed to resolve it.
>>
>
> One can certainly perform cladistic systematics without overt
> reference to evolution, and there may well be important advantages
> to limiting the reliance of cladistic analysis on particular
> evolutionary models.
> I have difficulties in seeing how one can justify doing cladistics
> at all, instead of say phenetics/"numerical taxonomy", without
> a prior belief in some form of common descent.
I think that's the weakness in Patterson's position. His response would
have been that these patterns exist regardless of our prior beliefs
about their cause, and are interesting simply because they are patterns
in the data demanding explanation. The explanation, for Patterson as
much as for anyone, was common descent. But he took pains to make
evolution the explanation rather than the axiom. I myself see no reason
why it can't be both, without being circular; reciprocal illumination.
John Harshman
catshark wrote:
> On Thu, 28 Aug 2003 18:55:54 +0000 (UTC), K-Man <k-...@nospam.com> wrote:
>
>
>>This past weekend my pastor repeated the old Colin Patterson quote re
>>"Tell me one thing that you _know_ about evolution." I've searched the
>>TO archive and googled the topic, and I've come up with some good
>>info--including the Patterson Misquoted A Tale of Two 'Cites'
>>article--but I don't think I have enough, or understand enough, to do a
>>rebuttal that will be convincing to a confirmed creationist. (Impossible
>>task, I know).
>>
>>My questions:
>>
>>1) Did Patterson really say, "One morning I woke up and something had
>>happened in the night, and it struck me that I had been working on this
>>stuff for 20 years and there was not one thing I knew about it."? If
>>so, what's the full context for this statement? What "stuff" was he
>>talking about.
>>
>>2) Did Patterson really ask "Can you tell me anything you know about
>>evolution, any one thing, any one thing that is true?" Again, if so,
>>what was the context?
>>
>>3) Did someone in the audience really say "I do know one thing - it
>>ought not to be taught in high school." Again, if so, what was the
>>context? What was he or she saying shouldn't be taught in high school.
>>
>
> Yes (more or less). I found the text while doing a little research into
> the quote mines in _Darwin on Trial_. You can read it in full here:
>
> <http://www.linnean.org/html/publications/publications_lin_04_02.htm>
This was weird, and I would be very interested in knowing the context.
This was published in 2002 in The Linnean, though it's the (presumed)
text of a speech given in 1981. It's unusual that in web form at least
it contains no explanatory or prefatory material. I'd be interested in
knowing whether anyone with access to a hard copy of The Linnean (April,
2002) can find an introduction or other explanation that was left off
the web site, particularly since the only previous source for the speech
was a transcript of a recording made by creationist Luther Sunderland.
Colin Patterson was an odd guy, but I didn't know he was that odd. Be
aware that when Patterson says "evolutionist" in this speech, he means
"evolutionary systematist", which is to me quite a different thing.
[snip]
Bigdakine
>From: wil...@wehi.edu.au (John Wilkins)
>Date: 8/28/03 8:36 PM Hawaiian Standard Time
>Message-id: <1g0gget.zsby1frxv5oN%wil...@wehi.edu.au
>
<snip good stuff>
>--
>John Wilkins
>DARK IN HERE, ISN'T IT?
>wilkins.id.au
>
ID? Say it isn't so, John.
Stuart
Dr. Stuart A. Weinstein
Ewa Beach Institute of Tectonics
"To err is human, but to really foul things up
requires a creationist"
catshark
There is an editorial about the trancript and some letters to the editor in
the _following_ issue, July 2002 (PDF format) here:
<http://www.linnean.org/html/publications/linnean18_3.pdf>
The only part of the April 2002 issue I've found yet is the Patterson
transcript.
>
>Colin Patterson was an odd guy, but I didn't know he was that odd. Be
>aware that when Patterson says "evolutionist" in this speech, he means
>"evolutionary systematist", which is to me quite a different thing.
>
>[snip]
It may be a European usage. Peter J. Bowler (from Queen's University in
Belfast) in his _Evolution: The History of an Idea_ uses "evolutionist" all
the time and not limited to evolutionary systemists or evolutionary
biologists.
John Wilkins
....
>
> And Patterson's expressed reason for not considering cladograms to be
> evolutionary trees *a priori* was that you could then turn around and
> use the congruence of data represented by the cladogram as evidence for
> evolution/common descent without the danger of circular argument.
>
> (My position, for what it's worth: cladograms are *estimates* of
> history. If they aren't, we have no reason to care about them.)
Patterson and his colleagues tended at that time to rely on a rather
curious version of Popperian epistemology. In this account, all claims
made in science are conjectures not arrived at inductively (of course)
and hence needing to be theory-freely tested (subjected to
falsification). This is what cladograms were for.
I prefer (as later on did Patterson, Nelson and Platnick all) Hennig's
notion of "reciprocal illumination", in large part because it is a
better approximation of actual scientific practice.
In the Bayesian debate now ongoing in various disciplines, the claim
appears to be that we are all bayesians now, and that each hypothesis is
assessed on the basis of our prior information and knowledge, which
raises no small set of problems in itself. I think that we should see
data as a test of any hypothesis, even if this is not a process of
Popperian falsification, and to some extent we have to rely on prior
commitments. But these need not be coherent priors.
IMNSHO, cladograms are first approximations of history, but they are
rather more than that - they are inductive generalisations (and the
Popperian cladists would *hate* that, although I get the idea from
Nelson himself, in Otte and Endler). They are ways to represent complex
data in the most generalisable fashion. From *this* we test scenarios of
evolutionary history, and much of the time the translation is relatively
straightforward, or it would be if cladograms were immediately
resolvable from anything but the most simple of datasets. But lateral
transfer, hybridisation (which amounts to the same thing) and
undiscovered homoplasy makes it a non-trivial move to assign historicity
to a cladogram (that is, to convert it to a phylogenetic dendrogram).
John Wilkins
> On Fri, 29 Aug 2003 06:36:59 +0000 (UTC), wil...@wehi.edu.au (John
> Wilkins) wrote:
>
...
More or less. I think that evolutionary systematics as propounded by
Mayr is a chimera of two approaches - gradification, or the putting of
living things into grades, and cladification, or the putting of living
things into historical groups (branches, or clades from the Greek) of
the evolutionary tree). As such it suffers from subjectivity and
amphiboly of terms.
>
> And a clarification, please: Patterson also seems, in several places, to
> use "systematics" (without "evolutionary" tacked on) in a broader sense, to
> include *both* taxonomic methodologies, or am I misreading him?
Systematics is a historical movement or discipline, As such, of course
it includes both methodologies, along with a slew of others now defunct
more or less (e.g., Adanson's "natural method"). There is an
equivocation between the sense of the discipline as a historical thing,
and as a definitional thing, that causes no end of trouble in the
literature.
...
> >I think that assuming evolution *has* occurred but that we aren't given
> >a chronicle before we start doing paleontology and all the other
> >disciplines that investigate the past of life, we can only say that some
> >sequence is more likely than some other, in a Bayesian sort of way, and
> >there's the rub: what priors do we assume to make the likelihoods
> >useful? the obvious, and unhelpful, answer is, we use the best knowledge
> >we now have to assess new claims.
>
> That last is your answer to the problem, not necessarily Patterson's,
> right?
Absolutely. My sympathy for pattern cladism should not be construed as
unalloyed agreement with it.
>
> >Of course, if our priors are wrong, we
> >will assess new claims wrongly. Moreover, *how* do we test new claims of
> >ancestry and history? The pattern cladist account is - by using the
> >*known* data sets and the relations they evidence. In short, we test
> >phylogenies with well-attested cladograms. Hence, cladograms are not, in
> >themselves, phylogenies.
>
> Is this what Patterson is getting at when he says that the 'theory of
> evolution has got between scientists and the data'?
Yes. There was and continues to be individuals being wedded to a
particular model of how evolution *must* have proceeded (whether it is a
panadaptationist account, or relies on sympatric or allopatric
speciation, or whatever) who thus "read" cladograms as fitting that
process. The apothegm that biology follows no rules, including this one,
applies here.
>
> >
> >So, in Patterson's words, what do we *know* about the evolutionary
> >history of any group of organisms? In the most rigid sense of "to know"
> >we know very little indeed. He and the other pattern cladists overstated
> >this, I believe, but the point is firm. And to date, *nobody* of any
> >school has managed to resolve it.
>
> But the IDiots have: 'Throw it out, bassinet, bathwater, baby and all.'
They also throw out the kitchen, house, foundation and driveway. I
gather they want to keep the family TV, though...
...
John Wilkins
> >Subject: Re: Response to Colin Patterson "Quote:" Need Help
> >From: wil...@wehi.edu.au (John Wilkins)
> >Date: 8/28/03 8:36 PM Hawaiian Standard Time
> >Message-id: <1g0gget.zsby1frxv5oN%wil...@wehi.edu.au
>
> >
>
> <snip good stuff>
>
> >--
> >John Wilkins
> >DARK IN HERE, ISN'T IT?
> >wilkins.id.au
> >
>
> ID? Say it isn't so, John.
>
The domain stands for "individual", intelligent, designed or
otherwise...
John Wilkins
> Andrew Criddle wrote:
And as I have said elsehwere, I agree more with Hennig on that score. I
think also it bears noting that pattern cladism of the early 80s is a
very different thing to the pattern cladism of the early noughties, of
Chris Humphries, Malte Ebach and David Williams at the NHM. But the
tendency has been, I think, to stress the objectivity of cladism against
the subjectivity of older impressionistic classification, and then to
overgeneralise. I do try to counter this with Malte. Sometimes, though,
he convinces me, not vice versa...
syvanen
Inflammatory yes, but not stupid nor false. I have read a number of
pattern cladists papers (one by Patterson) and I find the entire
enterprise devoid of common sense. They go out of their way to remove
biological reasoning from their interpretation character sets.
>
> Patterson was a pattern cladist. These are cladists who recognise that
> the nodes in a cladogram are not automatically ancestors. If anyone
> thinks that this is fanatical, they need to investigate some
> mathematical work on minimum message length encoding.
You needn't lecture me on this, my papers 8 and 9 at
http://www.vme.net/hgt/ that describe the star phylogeny are based on
the fact that the internal nodes in that data are meaningless noise
(as is shown in rigorously statistical test).
>
> I'm not going to get into a debate over pattern cladism except to note
> that even "*non*-fanatical" cladists such as Elliot Sober note the
> difficulty in reconstructing the past on the basis on modern
> relationships. Information is lost through insufficient data, horizontal
> transfer, homoplasy and noise.
This is the problem with pure cladistic approaches. They were
incapable of dealing with problems such as horizontal gene transfer
until the phenomena was pointed out to them by molecular biologists
that were uninfected with their ideology. I know from my own
experience. I argued with some of them in the 1980's that they were
overlooking horizontal gene transfer and that further their entire
approach was designed to hide this phenomena. They resisted this
suggestion. Now that it is an accepted fact (no thanks to them) they
can now go back and tinker with their theoretical underpinnings and
try to accomodate hgt. It seems to me that a theory that is
inherently structured to resist what is, in my opinion, the most
important discovery in evolutionary biology since the structure of DNA
and must be modified with alot of hand waving is a totally useless
theory.
> *Process* cladists like Farris are
> perhaps fanatical in that they denigrate anyone who does not think
> cladism is the magic method for finding out the past. Pattern cladists
> have been intemperate in their claims about evolution and
> "neo-Darwinism" and the like, but they have a very real, and so far
> unanswered, point - cladograms are arrangements of datasets, not
> histories,
But histories can be reconstructed using this data. I find cladograms
almost totally useless accept in one very narrow way. Distance trees
have the virtue of inferring time as well relationships and to also
provide a quick visual estimation of error (ie relative length of
internal branches which cladograms obscure horribly).
>and anyone who thinks a cladogram is a history in and of
> itself is reinventing Haeckel's recapitulation theory.
Since nobody does, this is a straw man. But you must agree that once
neotony is properly factored in, ontology does indeed provide us with
a glimpse of our evolutionary history.
Mike Syvanen
ps Though Patterson was not a creationist, he is responsible for his
views and the creationists, in this example, do not need to quote mine
to find a favorable position.
John Harshman
catshark wrote:
Not very satisfactory. We do get an implication in the letter that Peter
Forey provided some sort of information in the April issue. But we don't
know where the transcript came from. It's odd in that it was published
so long after the talk itself, and also so long after Patterson's death.
>>Colin Patterson was an odd guy, but I didn't know he was that odd. Be
>>aware that when Patterson says "evolutionist" in this speech, he means
>>"evolutionary systematist", which is to me quite a different thing.
>>
>>[snip]
>
> It may be a European usage. Peter J. Bowler (from Queen's University in
> Belfast) in his _Evolution: The History of an Idea_ uses "evolutionist" all
> the time and not limited to evolutionary systemists or evolutionary
> biologists.
Let me be clearer. Many people use "evolutionist" to refer to
evolutionary biologists. Patterson was using it in even more restricted
form, to refer to a particular school of systematics, of which the
leading living representative (still) is Ernst Mayr. Evolutionary
systematists are into such things as recognizing paraphyletic groups
based on the degree of morphological divergence of the clades removed
from them, identifying ancestors in the fossil record, basing
relationships on complex scenarios involving hypothetical ancestors and
pathways of evolutionary change, and such like. Most systematists these
days are not evolutionary systematists in that sense, although they are
still evolutionary biologists. Being an "evolutionary systematist" has
nothing to do with whether you believe that evolution occurs and has
occurred in the past.
John Harshman
John Wilkins wrote:
As far as I can tell, all the cladists (in the pejorative sense) claim
to be Popperian, to the extent that there are huge arguments in print
about who's really Popperian and who isn't. I expect when all that dies
down we'll get the same arguments about who's really Bayesian.
> They are ways to represent complex
> data in the most generalisable fashion. From *this* we test scenarios of
> evolutionary history, and much of the time the translation is relatively
> straightforward, or it would be if cladograms were immediately
> resolvable from anything but the most simple of datasets. But lateral
> transfer, hybridisation (which amounts to the same thing) and
> undiscovered homoplasy makes it a non-trivial move to assign historicity
> to a cladogram (that is, to convert it to a phylogenetic dendrogram).
In my work I generally find that the first two don't come up. It's the
third that presents the problem. And depending on what you mean by
non-trivial, that's often not a problem either. Unfortunately, often it
is. Which is why there are still unanswered questions in phylogeny.
But I think we may have some form of agreement about what a cladogram is.
John Wilkins
It already has... Dan Faith? ;-)
Actually, Popper has ceased to be an issue so far as I can tell except
among the players of the 1970s. I was asked (and tempted) to write and
deliver a talk on "Systematics without Popper" but declined due to the
thesis and fear...
>
> > They are ways to represent complex
> > data in the most generalisable fashion. From *this* we test scenarios of
> > evolutionary history, and much of the time the translation is relatively
> > straightforward, or it would be if cladograms were immediately
> > resolvable from anything but the most simple of datasets. But lateral
> > transfer, hybridisation (which amounts to the same thing) and
> > undiscovered homoplasy makes it a non-trivial move to assign historicity
> > to a cladogram (that is, to convert it to a phylogenetic dendrogram).
>
>
> In my work I generally find that the first two don't come up. It's the
> third that presents the problem. And depending on what you mean by
> non-trivial, that's often not a problem either. Unfortunately, often it
> is. Which is why there are still unanswered questions in phylogeny.
>
> But I think we may have some form of agreement about what a cladogram is.
catshark
Ah, my mistake. It did occur to me while I was reading it that Patterson
was putting that meaning to the term but, in trying to follow what else was
going on, it didn't stick. At the time, I thought it was a little odd but
put it down to possibly just a little common terminology slippage.
David Jensen
<unktkvs8ia63h8rl0...@4ax.com>:
I'd be really impressed if it were the latter and you were at Strand.
catshark
<da...@dajensen-family.com> wrote:
[...]
>>It isn't all that much of a book, though, more like a monograph . . . only
>>201 pages, *including* 54 pages of notes, bibliography and index, in not
>>particularly dense typeface. I hesitated for a long time because it was
>>marked $12.50 (US), almost as much as most trade paperbacks (though it is
>>hardback and in good condition). A tad more luck: either the clerk misread
>>the price or gave me a "I've seen your face a *lot*" discount and I got it
>>for $10.
>
>I'd be really impressed if it were the latter and you were at Strand.
No, I don't get into Manhattan that much. But I could take a stab at it.
;-)
David Jensen
>On Tue, 2 Sep 2003 04:22:47 +0000 (UTC), David Jensen
><da...@dajensen-family.com> wrote:
>
>[...]
>
>>>It isn't all that much of a book, though, more like a monograph . . . only
>>>201 pages, *including* 54 pages of notes, bibliography and index, in not
>>>particularly dense typeface. I hesitated for a long time because it was
>>>marked $12.50 (US), almost as much as most trade paperbacks (though it is
>>>hardback and in good condition). A tad more luck: either the clerk misread
>>>the price or gave me a "I've seen your face a *lot*" discount and I got it
>>>for $10.
>>
>>I'd be really impressed if it were the latter and you were at Strand.
>
>No, I don't get into Manhattan that much. But I could take a stab at it.
>;-)
Used book stores are addictive, aren't they.
catshark
<da...@dajensen-family.com> wrote:
>In talk.origins, catshark <cats...@yahoo.com> wrote in
><4ns8lv05p55eitl0k...@4ax.com>:
>
>>On Tue, 2 Sep 2003 04:22:47 +0000 (UTC), David Jensen
>><da...@dajensen-family.com> wrote:
>>
>>[...]
>>
>>>>It isn't all that much of a book, though, more like a monograph . . . only
>>>>201 pages, *including* 54 pages of notes, bibliography and index, in not
>>>>particularly dense typeface. I hesitated for a long time because it was
>>>>marked $12.50 (US), almost as much as most trade paperbacks (though it is
>>>>hardback and in good condition). A tad more luck: either the clerk misread
>>>>the price or gave me a "I've seen your face a *lot*" discount and I got it
>>>>for $10.
>>>
>>>I'd be really impressed if it were the latter and you were at Strand.
>>
>>No, I don't get into Manhattan that much. But I could take a stab at it.
>>;-)
>
>Used book stores are addictive, aren't they.
The place I was talking about is a fairly small shop but stacked to the
ceiling with shelves in narrow aisles and just moving about stirs up enough
dust to make me sneeze . . . I *love* it! It's near my office and I hit it
at least once a week and spend an hour, even though I am mostly checking
for anything new (I'd spend more time but I have to at least pretend to
work). There is another one near where I live that used to be like that
but it is in a touristy town and new owners decided to make it neater. Now
there is half or less the number of books and what is there is consistently
double the price of the other shop. I hardly ever go there now.
Just talking about it makes me crave another fix . . .