Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Support these premises-2

10 views
Skip to first unread message

Zoe

unread,
Aug 26, 2005, 10:01:56 PM8/26/05
to
Okay, it was a mistake to spend more than a sentence on the
evolutionary tree itself. I should have known that the main thrust of
my post would be ignored.

Enough said about the tree as a metaphor. I am asking evolutionists
to defend the premises upon which they have built their tree.

Let me try again.

Please support these premises:

Premise 1: Similarity is sufficient evidence that life forms should
be grouped together in relationship.

(And, Steven, homology is all about similarity of structure and is an
evolutionary word when used in biology. Inherent in that word is an
evolutionary meaing, the very premise of which I am questioning --
that similarity means relationship. Homology is the supposition that
life forms correspond in basic type of structure and derive from a
common primitive origin; i.e., the wing of a bat and the foreleg of a
mouse are homologous (Webster's). If homology is about similarity,
then it is a weak defense to use "similarity" (dressed up in another
word) as your support for the premise that similarity means
relationship.

To demonstrate premise 1, there can be no exceptions to the similarity
rule. If exceptions are allowed, then exceptions should be allowed to
other theories that differ from yours.

Premise 2: Dissimiliarity is sufficient evidence that life forms
should be grouped together in relationship.

(And, again, Steven, I don't know why you would say that dissimilarity
is not one of the standards used in declaring relationship. I have
been told by posters here that the pattern of similarity/dissimilarity
in life forms is evidence for relationship.)

To demonstrate this, there can be no exceptions to the dissimilarity
rule. If exceptions are allowed, then exceptions should be allowed to
other theories that differ from yours.

Premise 3: A tree-like pattern is sufficient evidence that life forms
are related through a single common ancestor.

(Once again, Steven, your statement that "depending on what traits one
emphasizes and compares, *different* tree-like hierarchies can be
obtained" says it all. As to the traits chosen for the evolutionary
tree, can you defend why those traits are emphasized and are the only
traits that are correct? What makes those choices the correct ones?)

So...since Steven has not yet persuasively defended these premises, I
am asking again: Can anyone else (or Steven again) better defend
these premises?

Harlequin

unread,
Aug 26, 2005, 10:54:17 PM8/26/05
to
Zoe <muz...@aol.com> wrote in
news:t9gvg11b6hi7itp4b...@4ax.com:

> Okay, it was a mistake to spend more than a sentence on the
> evolutionary tree itself. I should have known that the main thrust of
> my post would be ignored.
>
> Enough said about the tree as a metaphor. I am asking evolutionists
> to defend the premises upon which they have built their tree.
>
> Let me try again.
>
> Please support these premises:
>
> Premise 1: Similarity is sufficient evidence that life forms should
> be grouped together in relationship.


You have Premise 2 below with the same sentence with "Dissimiliarity"
instead of "Similiarity."

I will have to disagree with this.

1) Similiarity is _NOT_ sufficent evidence that life forms should
be grouped together in a relationship in evolutionary biology.

2) Similiarity and dissimiliarity cannot be looked separately.

3) Not every form of similiarity is considered relevant by
modern evolutionary biologists. If one uses cladistic methodology
then one only looks at synapomorphies,"shared derived homologies,"
to determine relationships. I am not sure that I know how to
make that language simple enough for you I fear.

>
> (And, Steven, homology is all about similarity of structure and is an
> evolutionary word when used in biology. Inherent in that word is an
> evolutionary meaing,

Actually homology can be defined completely independently of evolution.
Indeed noticing that certain parts on one type of organism correspond
to parts of another organism certainly long predates Darwin.

> the very premise of which I am questioning --
> that similarity means relationship.

I would question it too. It is an oversimplification at best
and a strawman at worst.


> Homology is the supposition that
> life forms correspond in basic type of structure and derive from a
> common primitive origin; i.e., the wing of a bat and the foreleg of a
> mouse are homologous (Webster's). If homology is about similarity,
> then it is a weak defense to use "similarity" (dressed up in another
> word) as your support for the premise that similarity means
> relationship.


While the wing and leg are homologous, I would remind you that
a dictionary can be a dangerous thing to rely on here.


> To demonstrate premise 1, there can be no exceptions to the similarity
> rule. If exceptions are allowed, then exceptions should be allowed to
> other theories that differ from yours.
>
> Premise 2: Dissimiliarity is sufficient evidence that life forms
> should be grouped together in relationship.

This premise is outright false. Dissimiliarity _can_ be evidence for
not having a (close) relationship.


> (And, again, Steven, I don't know why you would say that dissimilarity
> is not one of the standards used in declaring relationship. I have
> been told by posters here that the pattern of similarity/dissimilarity
> in life forms is evidence for relationship.)

I am one of them. And your post here strongly suggests that you
do not understand what we mean by ths.


> To demonstrate this, there can be no exceptions to the dissimilarity
> rule. If exceptions are allowed, then exceptions should be allowed to
> other theories that differ from yours.

What is an "exception" here anyways? If one is only looking
at single characteristics than one would expect many exceptions
since evolution should (and does) reverse many prior trends at
times. One must look at many characteristics.

> Premise 3: A tree-like pattern is sufficient evidence that life forms
> are related through a single common ancestor.
>
> (Once again, Steven, your statement that "depending on what traits one
> emphasizes and compares, *different* tree-like hierarchies can be
> obtained" says it all. As to the traits chosen for the evolutionary
> tree, can you defend why those traits are emphasized and are the only
> traits that are correct? What makes those choices the correct ones?)


I could pick out some very badly choosen characteristics and force
some very bad trees to form. After all anything can be forced into
a "tree."

When cladograms are properly formed, the vast majority of the time
the trees will agree with each other to a degree extremely unlikely
to happen by chance. Unfortunately, you certainly don't have
the background to understand how any of this is done. Again
it comes down to that you need to learn the basic materials
before dealing with more advanced issues. You need to learn
basic biology and mathematics first. This issues involved
are discussed in the "29+ Evidences for Macroevolution" which
you have been pointed to many times before.

> So...since Steven has not yet persuasively defended these premises, I
> am asking again: Can anyone else (or Steven again) better defend
> these premises?


Again we don't want to defend them since they have nothing
to do with the actual study of evolution.


--
Anti-spam: replace "usenet" with "harlequin2"

"So easily are men the dupes of their own prejudice."
-- Percival Lowell, _Mars as the Abode of Life_.
1908. pp. 153-154.

Bobby D. Bryant

unread,
Aug 26, 2005, 11:34:12 PM8/26/05
to
On Sat, 27 Aug 2005, Zoe <muz...@aol.com> wrote:

> Okay, it was a mistake to spend more than a sentence on the
> evolutionary tree itself. I should have known that the main thrust of
> my post would be ignored.

Yeah, bad arguments tend to stop us short.


> Enough said about the tree as a metaphor. I am asking evolutionists
> to defend the premises upon which they have built their tree.
>
> Let me try again.
>
> Please support these premises:
>
> Premise 1: Similarity is sufficient evidence that life forms should
> be grouped together in relationship.

You need to realize that scientists and creationists start from the
same point: an observation that something looks unlikely to be the
result of pure chance. (From there they diverge, with the creationist
jumping immediately to "goddidit" and the scientist looking for an
actual cause for the unlikely something.)

Do you think the genetic and morphological similarities that guide
the construction of the tree are the result of mere chance, or the
result of some cause?

Where the discussion goes from here depends on your answer to that.


--
Bobby Bryant
Austin, Texas

John Harshman

unread,
Aug 27, 2005, 12:44:00 AM8/27/05
to
Zoe wrote:

Steven was talking there about arbitrary hierarchies that you can get
from anything, like, for example, sweaters. He wasn't talking about
biological hierarchies, because those do not depend on what traits one
emphasizes. That's the point of the multiple nested hierarchy.

> So...since Steven has not yet persuasively defended these premises, I
> am asking again: Can anyone else (or Steven again) better defend
> these premises?

Objection, your honor: asked and answered.

Steven J.

unread,
Aug 27, 2005, 1:37:43 AM8/27/05
to

"Zoe" <muz...@aol.com> wrote in message
news:t9gvg11b6hi7itp4b...@4ax.com...

> Okay, it was a mistake to spend more than a sentence on the
> evolutionary tree itself. I should have known that the main thrust of
> my post would be ignored.
>
> Enough said about the tree as a metaphor. I am asking evolutionists
> to defend the premises upon which they have built their tree.
>
> Let me try again.
>
> Please support these premises:
>
> Premise 1: Similarity is sufficient evidence that life forms should
> be grouped together in relationship.
>
Why should we support that premise? It is wrong, and no one is basing an
argument for common descent on it. "Similarity" in some respects is
rejected as a basis for grouping organisms together. For example, despite
similarities in shape and habitat, whales are not grouped apart from other
mammals and together with, e.g. sharks and ichthyosaurs. The box-camera
eyes of octopuses are not seen as a reason to group them with vertebrates,
who have superficially "similar" eyes.

Of course, the similarities of the eyes of cephalopods and vertebrates does
not extend beyond the features necessary to a "box-camera" eye (with lens,
retina, etc.). Famously, vertebrate eyes have an "inverted retina" and
blind spot, and cephalopod eyes do not, among other differences. Likewise,
the common features of whales, sharks, and ichthyosaurs (aside from those
shared by all vertebrates) are adaptions to the way they live and move.

It was for this reason that I brought up "homology" -- detailed similarities
not required by similarity in function. A bat's wing doesn't need all the
similarities it has to a mouse's forelimb, and the detailed correspondence
of the bones in both limbs is remarkable and calls for some explanation.
Note, also, that there are *many* shared homologies between bats and other
mammals. Finding some homologies between bats and mice, but not between
bats and birds, and different homologies between bats and birds, but not
with mice, and still different homologies between bats and pterosaurs (but
not with birds or mice) would not support the idea that these homologies are
explained by inheritance from a common ancestor. Presumably, branching
descent with modification would pass on entire suites of traits together --
and, therefore, entire suites of shared homologies arranged in a consistent
nested hierarchy argues for branching descent with modification.


>
> (And, Steven, homology is all about similarity of structure and is an
> evolutionary word when used in biology. Inherent in that word is an
> evolutionary meaing, the very premise of which I am questioning --
> that similarity means relationship. Homology is the supposition that
> life forms correspond in basic type of structure and derive from a
> common primitive origin; i.e., the wing of a bat and the foreleg of a
> mouse are homologous (Webster's). If homology is about similarity,
> then it is a weak defense to use "similarity" (dressed up in another
> word) as your support for the premise that similarity means
> relationship.
>

Zoe, the word "homology" was coined by the comparative anatomist Sir Richard
Owens, an opponent of Darwin who rejected common descent and adaption by
natural selection. The concept is older than Owen; it was used by Carrolus
Linnaeus as the basis for his taxonomy. Linnaeus, also, had no concept of
evolution (except in some cases within genera). Homologies can be
recognized without reference to common descent. Homology does not mean
merely "similarity;" it distinguishes one sort of similarity (detailed in
many aspects of a feature, not just those needed for common function) from
another sort -- "analogy," or similarity in features required for common
function. The similarity between octopus and human eyes is "analogous;"
that between human eyes and other vertebrate eyes is "homologous."

Note, just to keep things complicated, that a structure can be analogous in
one respect and homologous in another. Bat wings and bird wings are
homologous, as forelimbs: they share the pattern of humerous, radius, carpal
bones, phalanges, etc. But as wings, they are analogous: the shared
forelimb features are adapted as wings in very different ways.


>
> To demonstrate premise 1, there can be no exceptions to the similarity
> rule. If exceptions are allowed, then exceptions should be allowed to
> other theories that differ from yours.
>

I don't think I agree with that. Are exceptions to "creation theory"
explicable? Note that what you wanted was to be able to say that any entity
which possessed some particular, subjectively defined set of properties was
the product of intelligence. It rather makes a difference if nondesigned
entities can possess these properties. Evolution is inferred from a very
large number of features of living things, and the way in which they are
arranged among different living things.


>
> Premise 2: Dissimiliarity is sufficient evidence that life forms
> should be grouped together in relationship.
>
> (And, again, Steven, I don't know why you would say that dissimilarity
> is not one of the standards used in declaring relationship. I have
> been told by posters here that the pattern of similarity/dissimilarity
> in life forms is evidence for relationship.)
>

First of all, Zoe, obviously related things can be dissimilar. You can tell
members of a litter of puppies apart, although they share the same parents.
You can tell different breeds of dogs apart, even though they share common
ancestors. Now, the dissimilarities between, say, a Newfoundland and a
chihuahua are not evidence for common ancestry, but they are clearly
compatible with common ancestry. Moreover, they help us group the different
breeds of dogs together to say which breeds are more closely related to
others, and which are more distantly related. The arguments for common
ancestry (and branching order) of different primates are similar in
principle to those for classifying different breeds in their own nested
hierarchy.

Now, since you seem to have your own idiosyncratic take on "creation
theory," I'm not sure how much weight more ... conventional ... creationist
ideas will have with you. Most creationists hold to an idea of "kinds" above
the species level, which include multiple species (in most cases; humans are
always a kind to themselves) which are indeed related by common descent. A
typical creationist would hold that, e.g. not merely different breeds of
_Felis catus_ (the domestic cat) but different members of _Felis_, perhaps
even different members of the family Felidae (e.g. lions, pumans, bobcats,
and housecats) are members of a single "kind." Of course, the grouping of
species into "kinds" requires this same sort of argument for relationship
from a pattern of similarities and dissimilarities; creationists fail to be
consistent in the application of this principle (e.g. if you group lions and
housecats together, you have no reason not to group humans and chimps
together).

Now, beyond this, recall some of the above points about analogy and
homology. Birds of widely different sizes, habits, and diet share the same
basic avian wing design. Likewise, bats of different sizes, habits and
diets share the same chiropteran wing design. It seems, on the principle of
"common design for common function," that small insect-eating birds and
small insect-eating bats should have more similar wings than, e.g. small
insect-eating birds and larger, fruit-eating birds. Likewise, we might
expect greater homologies between toucans and fruit bats than between fruit
bats and brown bats. Basic wing designs don't seem to be arranged according
to ecological niche, but fall into a pattern suggesting, again, descent with
ad hoc modification. Fruit bats and brown bats have similar wings because
they inherited them from a common ancestor; toucans, hummingbirds, swallows,
and penguins share similar wing designs because, again, a common avian
ancestor had wings like that.

Note that we know about mechanisms of reproduction, mutation, and selection,
and know that these processes can produce suites of organisms that fall into
a consistent nested hierarchy. Again, we know that consistent nested
hierarchies typically result from a process of branching descent with
modification (e.g. languages, breeds of domestic animals, manuscript
copies), and that artifacts typically do not (just to complicate matters,
lateral exchange of traits is possible for all these things -- e.g. gene
transfer in bacteria; it is rare enough in all these cases that it does not
mess up the nested hierarchy, at least for eukaryotes).


>
> To demonstrate this, there can be no exceptions to the dissimilarity
> rule. If exceptions are allowed, then exceptions should be allowed to
> other theories that differ from yours.
>

Since you don't understand the "dissimilarity rule," this condition is
nonsense.


>
> Premise 3: A tree-like pattern is sufficient evidence that life forms
> are related through a single common ancestor.
>

That, again, is not quite my point, although it is at least within an
astronomical unit of my point.


>
> (Once again, Steven, your statement that "depending on what traits one
> emphasizes and compares, *different* tree-like hierarchies can be
> obtained" says it all. As to the traits chosen for the evolutionary
> tree, can you defend why those traits are emphasized and are the only
> traits that are correct? What makes those choices the correct ones?)
>

Zoe, take any traits you like. Some (e.g. the similarity in shape that
links sharks and dolphins, or the similarity in color that links some
parrots and some iguanas) will yield trees that look very, very different
from any phylogenic tree drawn by evolutionists. Of course, a tree based on
overall color and one based on overall shape will yield very different trees
from each other. But draw trees based on comparative limb anatomy, or
comparative skull anatomy, or on various sets of genes, or on various other
sets of genes (after all, you have a lot of sets of genes to compare), or on
proteins (which don't have to form the same tree as genes, since different
sets of codons can code for the same amino acids), and you will find that
they form trees that look very much alike. As I noted in an earlier post,
*all* animals with one bone in the lower jaw and three bones in the middle
ear also have (although these features hardly seem necessary to each other)
mammary glands and a left (but not a right) aortic arch.

Why do not *all* trees yield exactly the same pattern? Well, first of all,
why should they? The most indicative features are those which are not
strongly adaptive to or obvious consequences of a particular lifestyle. For
example, lots of arctic mammals have white fur, presumably as camoflage
against the snow. Whiteness is a sign of adaption to common function, and
therefore would be expected even if the animals did not share common
ancestry (and, indeed, polar bears and arctic foxes don't share a closer
ancestry with each other than they do with, respectively, brown bears and
red foxes). But there's no obvious reason that, e.g. giraffes and mice and
humans should all share seven bones in the neck, or even three bones in the
middle ear (many vertebrates hear well without this feature).


>
> So...since Steven has not yet persuasively defended these premises, I
> am asking again: Can anyone else (or Steven again) better defend
> these premises?
>

To someone, like yourself, determined not to understand or be persuaded, I
rather doubt it. But I took a shot.
>
-- Steven J.


Thore "Tocis" Schmechtig

unread,
Aug 27, 2005, 2:50:17 AM8/27/05
to
Zoe wrote:

> Let me try again.

No. Your bullshit already wasted more than enough bandwidth.

--
Regards

Thore "Tocis" Schmechtig

R. Baldwin

unread,
Aug 27, 2005, 8:21:52 AM8/27/05
to
"Zoe" <muz...@aol.com> wrote in message
news:t9gvg11b6hi7itp4b...@4ax.com...
> Okay, it was a mistake to spend more than a sentence on the
> evolutionary tree itself. I should have known that the main thrust of
> my post would be ignored.
>
> Enough said about the tree as a metaphor. I am asking evolutionists
> to defend the premises upon which they have built their tree.

The tree of life is not built on premises, Zoe. It is built on observations.
Tree diagrams are simply a method for comparing things with distinct
characteristics.

What makes you conclude that the premises you describe below are those used
in biology? Why not ask a biologist?

I would note, by the way, that given your wording, Premises 1 and 2 are
mutually contradictory.

gregwrld

unread,
Aug 27, 2005, 10:33:14 AM8/27/05
to


I'd like to nominate this for POTM. In any event, thanks Steve for an
illuminating artice...g(regwrld)

Victor Eijkhout

unread,
Aug 27, 2005, 1:35:45 PM8/27/05
to
Steven J. <sjt195...@nts.link.net.INVALID> wrote:

> > (And, Steven, homology is all about similarity of structure and is an
> > evolutionary word when used in biology. Inherent in that word is an

> > evolutionary meaing, [...]

> Zoe, the word "homology" was coined by the comparative anatomist Sir Richard
> Owens, an opponent of Darwin who rejected common descent and adaption by
> natural selection. The concept is older than Owen; it was used by Carrolus
> Linnaeus as the basis for his taxonomy. Linnaeus, also, had no concept of
> evolution (except in some cases within genera). Homologies can be
> recognized without reference to common descent.

And yet again we see the creationist making baldfaced assertions not
grounded in any knowledge of what they speak of. It's only my charitable
impulses that makes me attribute it to laziness and ignorance rather
than an active desire to deceive.

Great post btw. Nomination seconded, thirded, wherever we are.

V.
--
email: lastname at cs utk edu
homepage: www cs utk edu tilde lastname

Matthew Isleb

unread,
Aug 27, 2005, 2:20:47 PM8/27/05
to
Pardon the top posting. Just wanted to make the POTM nomination obvious to
whomever looks for these things.

B Richardson

unread,
Aug 27, 2005, 5:00:02 PM8/27/05
to
Zoe wrote:

>Okay, it was a mistake to spend more than a sentence on the
>evolutionary tree itself. I should have known that the main
thrust of
>my post would be ignored.
>
>Enough said about the tree as a metaphor. I am asking
evolutionists
>to defend the premises upon which they have built their tree.
>

Sizeable branches of the tree were taking shape before any
suggestion
of descent simply by the recongnition that larger more inclusive
groups
can be subdivided into smaller less inclusive groups. Mammals,
amphibians
and birds are all vertebrates and can be distinguished from non-
vertebrates.
Mammals have mammary glands, amphibians and birds do not. Birds have
feathers, mammals and amphibians do not. Its really the nesting
that produces the tree that you want to focus on.


You can start with a group characterized by a primitive character

+-------------+
| Vertebrates |
+-------------+

and that group can be subdivided into others. Amphibians
don't have amniotic eggs, birds don't have mammary glands,
mammals don't have feathers. There are plenty of characters
available to categorize something unambiguously into groups
within the larger group of vertebrates.


+-------------------+
| Vertebrates |
| +-------------+ |
| | Birds | |
| +-------------+ |
| |
| +-------------+ |
| | Amphibians | |
| +-------------+ |
| |
| +-------------+ |
| | Mammals | |
| +-------------+ |
+-------------------+

You can subdivide any group of your choosing into yet more groups
and you should be able to see why the term 'nested' is used.

+----------------------------+
| Vertebrates |
| +-------------+ |
| | Birds | |
| +-------------+ |
| |
| +-------------+ |
| | Amphibians | |
| +-------------+ |
| |
| +-----------------------+ |
| | Mammals | |
| | +-------------------+ | |
| | | Placental | | |
| | | +--------------+ | | |
| | | | Artiodactyla | | | |
| | | +--------------+ | | |
| | | +--------------+ | | |
| | | | Carnivora | | | |
| | | +--------------+ | | |
| | +-------------------+ | |
| | +-------------------+ | |
| | | Marsupials | | |
| | +-------------------+ | |
| +-----------------------+ |
+----------------------------+

You can recursively redivide groups into yet more groups that
still have defining characters. You don't have to be a scientist
to distinguish between Carnivora and Artiocactyla, or to even
recursively subdivide them. Below is a slightly more detailed
but still yet rudimentary recursive expansion of the above.


+---------------------------------------------------------+
| Vertebrates |
| +------------+ |
| | Birds | |
| +------------+ |
| |
| +------------+ |
| | Amphibians | |
| +------------+ |
| |
| +-----------------------------------------------+ |
| | Mammals | |
| | +-----------------------------------------+ | |
| | | Eutheria (placental mammals) | | |
| | | +-----------------------------------+ | | |
| | | | Artiodactyla | | | |
| | | | +----------------+ | | | |
| | | | | Cervidae (deer)| | | | |
| | | | +----------------+ | | | |
| | | | | | | |
| | | | +------------------------------+ | | | |
| | | | | Bovidae | | | | |
| | | | | +----------------------+ | | | | |
| | | | | | Bos | | | | | |
| | | | | | +-----------------+ | | | | | |
| | | | | | | cattle | | | | | | |
| | | | | | | +-------------+ | | | | | | |
| | | | | | | | Bos indicus | | | | | | | |
| | | | | | | +-------------+ | | | | | | |
| | | | | | | +-------------+ | | | | | | |
| | | | | | | | Bos taurus | | | | | | | |
| | | | | | | +-------------+ | | | | | | |
| | | | | | +-----------------+ | | | | | |
| | | | | | | | | | | |
| | | | | | +----------+ | | | | | |
| | | | | | | Yak | | | | | | |
| | | | | | +----------+ | | | | | |
| | | | | | | | | | | |
| | | | | | +----------+ | | | | | |
| | | | | | | Banteng | | | | | | |
| | | | | | +----------+ | | | | | |
| | | | | +----------------------+ | | | | |
| | | | | +----------------+ | | | | |
| | | | | | Sheep | | | | | |
| | | | | +----------------+ | | | | |
| | | | | +----------------+ | | | | |
| | | | | | Goats | | | | | |
| | | | | +----------------+ | | | | |
| | | | | +----------------+ | | | | |
| | | | | | Antelopes | | | | | |
| | | | | +----------------+ | | | | |
| | | | +------------------------------+ | | | |
| | | +-----------------------------------+ | | |
| | | | | |
| | | +-----------------------------------+ | | |
| | | | Carnivora | | | |
| | | | +-----------------------+ | | | |
| | | | | Felifornia | | | | |
| | | | | +-------------------+ | | | | |
| | | | | | Felidae | | | | | |
| | | | | | +-------------+ | | | | | |
| | | | | | | Panthera | | | | | | |
| | | | | | | +---------+ | | | | | | |
| | | | | | | | Lion | | | | | | | |
| | | | | | | +---------+ | | | | | | |
| | | | | | | +---------+ | | | | | | |
| | | | | | | | Tiger | | | | | | | |
| | | | | | | +---------+ | | | | | | |
| | | | | | | +---------+ | | | | | | |
| | | | | | | | Leopard | | | | | | | |
| | | | | | | +---------+ | | | | | | |
| | | | | | +-------------+ | | | | | |
| | | | | | | | | | | |
| | | | | | +---------------+ | | | | | |
| | | | | | | Felis | | | | | | |
| | | | | | | +-----------+ | | | | | | |
| | | | | | | | Jungle Cat| | | | | | | |
| | | | | | | +-----------+ | | | | | | |
| | | | | | | +-----------+ | | | | | | |
| | | | | | | | House Cat | | | | | | | |
| | | | | | | +-----------+ | | | | | | |
| | | | | | | +-----------+ | | | | | | |
| | | | | | | | European | | | | | | | |
| | | | | | | | Wild Cat | | | | | | | |
| | | | | | | +-----------+ | | | | | | |
| | | | | | +---------------+ | | | | | |
| | | | | +-------------------+ | | | | |
| | | | | | | | | |
| | | | | +-------------+ | | | | |
| | | | | | Herpestidae | | | | | |
| | | | | | +---------+ | | | | | |
| | | | | | | Fossa | | | | | | |
| | | | | | +---------+ | | | | | |
| | | | | | +---------+ | | | | | |
| | | | | | | Meerkat | | | | | | |
| | | | | | +---------+ | | | | | |
| | | | | | +---------+ | | | | | |
| | | | | | | Mongoose| | | | | | |
| | | | | | +---------+ | | | | | |
| | | | | +-------------+ | | | | |
| | | | +-----------------------+ | | | |
| | | | | | | |
| | | | +-------------------------+ | | | |
| | | | | Canifornia | | | | |
| | | | | +---------------------+ | | | | |
| | | | | | Canidae | | | | | |
| | | | | | +--------------+ | | | | | |
| | | | | | | Canis | | | | | | |
| | | | | | | +---------+ | | | | | | |
| | | | | | | | Dog/Wolf| | | | | | | |
| | | | | | | +---------+ | | | | | | |
| | | | | | | +---------+ | | | | | | |
| | | | | | | | Coyote | | | | | | | |
| | | | | | | +---------+ | | | | | | |
| | | | | | | +---------+ | | | | | | |
| | | | | | | | Jackal | | | | | | | |
| | | | | | | +---------+ | | | | | | |
| | | | | | +--------------+ | | | | | |
| | | | | | +--------------+ | | | | | |
| | | | | | | Vulpes | | | | | | |
| | | | | | | +----------+ | | | | | | |
| | | | | | | | Red Fox | | | | | | | |
| | | | | | | +----------+ | | | | | | |
| | | | | | | +----------+ | | | | | | |
| | | | | | | | Grey Fox | | | | | | | |
| | | | | | | +----------+ | | | | | | |
| | | | | | | +----------+ | | | | | | |
| | | | | | | | Artic Fox| | | | | | | |
| | | | | | | +----------+ | | | | | | |
| | | | | | +--------------+ | | | | | |
| | | | | +---------------------+ | | | | |
| | | | | +---------------------+ | | | | |
| | | | | | Ursidae | | | | | |
| | | | | | +-----------------+ | | | | | |
| | | | | | | Ursus | | | | | | |
| | | | | | | +-------------+ | | | | | | |
| | | | | | | | Polar Bear | | | | | | | |
| | | | | | | +-------------+ | | | | | | |
| | | | | | | +-------------+ | | | | | | |
| | | | | | | | Grizzly Bear| | | | | | | |
| | | | | | | +-------------+ | | | | | | |
| | | | | | +-----------------+ | | | | | |
| | | | | | +-----------------+ | | | | | |
| | | | | | | Helarctos | | | | | | |
| | | | | | | +------------+ | | | | | | |
| | | | | | | | Sun Bear | | | | | | | |
| | | | | | | +------------+ | | | | | | |
| | | | | | +-----------------+ | | | | | |
| | | | | +---------------------+ | | | | |
| | | | +-------------------------+ | | | |
| | | +-----------------------------------+ | | |
| | +-----------------------------------------+ | |
| | +----------------+ | |
| | | Marsupials | | |
| | +----------------+ | |
| +-----------------------------------------------+ |
+---------------------------------------------------------+


Birds and amphibians or any other group can be recursively
subdivided into smaller less inclusive groups.
If categorization is ambiguous from superficial examination,
you resort to comparative anatomy and embryology. When you
explain this, you want the most constraining explanation
possible. Ideally, you don't want a mechanism that *can*
produce this sort of nesting, you want a mechanism that
*must* produce the nesting, and inheritance is the front
runner. You would want you explanation to have consistancy
checks too. Inheritance would be expected to leave things
around that are useless but harmless. Plantaris muscle
in humans, egg tooth on some newly born marsupials, multiple
chambered stomachs on cetaceans.

>
>Let me try again.
>
>Please support these premises:
>
>Premise 1: Similarity is sufficient evidence that life forms
should
>be grouped together in relationship.
>

That premise, I think was yanked out of /dev/null. I think we can
toss it back.

>
>(And, Steven, homology is all about similarity of structure and is
an
>evolutionary word when used in biology. Inherent in that word is
an
>evolutionary meaing, the very premise of which I am questioning --
>that similarity means relationship. Homology is the supposition
that
>life forms correspond in basic type of structure and derive from a
>common primitive origin; i.e., the wing of a bat and the foreleg
of a
>mouse are homologous (Webster's). If homology is about similarity,
>then it is a weak defense to use "similarity" (dressed up in
another
>word) as your support for the premise that similarity means
>relationship.
>
>To demonstrate premise 1, there can be no exceptions to the
similarity
>rule. If exceptions are allowed, then exceptions should be
allowed to
>other theories that differ from yours.
>
>Premise 2: Dissimiliarity is sufficient evidence that life forms
>should be grouped together in relationship.
>

Toss that one too. I think that when you describe the alleged
"premises" in your own words much gets lost in the translation.

You keep getting stuck on similarities/dissimilarities. There's
no predetermined set of (dis)similarities to look for. Think
about the phylogenetic signal. Look at the two groups of strings,
the ancestor for the first is all t's, for the second all c's:


ttcttttatatt
tttttttatatg
gtttgttaattt
gtgttttaattt
tttttttaattt

ccacgcccccct
ccccgcctccct
ccccccgctact
tcccccgctcct
ccacccgcccct

The (dis)similarities in the first group are different from the
second group, but the phylogenetic signal is the same.

Descent with modification is the provisionally accepted explanation
and at the moment the only explanation known that could produce
phylogenetic signal in nucleotide data. Its not that its preferred
over competing explanations, its that no competing explanations
have survived consistancy checks. With the phylogenetic signal in
nucleotide data especially, one can propose that members of a clade
are not related by descent and the phylogenetic signal is a design
artifact. Problem is it wont stay that way. You need a lot of
mutations
to account for all the variation at individual loci and other stuff
like genetic load, those mutations can only act to obscure the
phylogenetic signal if it originally was due to something other
than inheritance. The only way I can think of to get around that
is propose that the signal was maintained by some unknown mechanism
which ceased to operate when the ability to observe what mutations
do to DNA became available.


Zoe

unread,
Aug 27, 2005, 10:07:16 PM8/27/05
to
On 26 Aug 2005 19:54:17 -0700, "Harlequin" <use...@cox.net> wrote:

snip>

>> Please support these premises:
>>
>> Premise 1: Similarity is sufficient evidence that life forms should
>> be grouped together in relationship.
>
>
>You have Premise 2 below with the same sentence with "Dissimiliarity"
>instead of "Similiarity."
>
>I will have to disagree with this.
>
>1) Similiarity is _NOT_ sufficent evidence that life forms should
>be grouped together in a relationship in evolutionary biology.

okay, I'll rephrase.

Please defend the premise that similarity can be justifiably used as a
standard for determining relationship between species.

>2) Similiarity and dissimiliarity cannot be looked separately.

okay, keep them together if you wish. Please justify why
similarity/dissimilarity can be justifiably used as a standard for
determining relationship between species.

There is a game called World of Warcraft in which the characters are
lifelike, can spawn and respawn, and can reproduce copies of
themselves, depending on the goals of the creators of the game. After
a while you begin to recognize templates for the patterns of
similarities and dissimilarties within the game. I am betting that,
based on the observation of reproduction and patterns of
similarity/dissimilarity in the life forms in the game, an
evolutionist "ant" might wander around within that game world and
conclude that the yeti are more closely related to the raptor and thus
have a common ancestor, or that the horde and alliance characters have
a common ancestor, and so on. Coming from the "ant's" perspective,
this might seem a reasonable conclusion based on an arbitrary decision
to call patterns of similarity/dissimilarity evidence for relationship
between species. But coming from the creators of the game's
perspective, nothing could be further from the truth.

>3) Not every form of similiarity is considered relevant by
>modern evolutionary biologists. If one uses cladistic methodology
>then one only looks at synapomorphies,"shared derived homologies,"
>to determine relationships. I am not sure that I know how to
>make that language simple enough for you I fear.

haven't I made MY language simple enough for you to understand? I am
simply asking for a defense of the concept that if you are able to
group life forms according to their similarities/dissimilarities, why
does this ability to do so lead you to the conclusion that, therefore,
these life forms must be related to each other through a common
ancestor?

>> (And, Steven, homology is all about similarity of structure and is an
>> evolutionary word when used in biology. Inherent in that word is an
>> evolutionary meaing,
>
>Actually homology can be defined completely independently of evolution.

you're on a tangent already, Harlequin. I said it is an evolutionary
word WHEN USED IN BIOLOGY. I agree with you that it can be defined
completely independently of evolution. Steven was using the word in
an evolutionary sense.

>Indeed noticing that certain parts on one type of organism correspond
>to parts of another organism certainly long predates Darwin.
>
>> the very premise of which I am questioning --
>> that similarity means relationship.
>
>I would question it too. It is an oversimplification at best
>and a strawman at worst.

okay, so what are the basic tools used to determine relationship
between species if not similarity/dissimilarity? And if similarity is
one of the key tools used, I would like you to support the premise
that such a tool is valid in concluding relationship between species.

>> Homology is the supposition that
>> life forms correspond in basic type of structure and derive from a
>> common primitive origin; i.e., the wing of a bat and the foreleg of a
>> mouse are homologous (Webster's). If homology is about similarity,
>> then it is a weak defense to use "similarity" (dressed up in another
>> word) as your support for the premise that similarity means
>> relationship.
>
>
>While the wing and leg are homologous, I would remind you that
>a dictionary can be a dangerous thing to rely on here.

now really. Dangerous? Hmmm.

>> To demonstrate premise 1, there can be no exceptions to the similarity
>> rule. If exceptions are allowed, then exceptions should be allowed to
>> other theories that differ from yours.
>>
>> Premise 2: Dissimiliarity is sufficient evidence that life forms
>> should be grouped together in relationship.
>
>This premise is outright false. Dissimiliarity _can_ be evidence for
>not having a (close) relationship.

and haven't I been told that it CAN also be evidence for relationship?

>> (And, again, Steven, I don't know why you would say that dissimilarity
>> is not one of the standards used in declaring relationship. I have
>> been told by posters here that the pattern of similarity/dissimilarity
>> in life forms is evidence for relationship.)
>
>I am one of them. And your post here strongly suggests that you
>do not understand what we mean by ths.

well, explain yourself then. You leave it hanging here with the
mysterious suggestion that there is meaning to this, but you're not
saying what it is.

snip rest because I would like direct answers to the above first>

Zoe

unread,
Aug 27, 2005, 10:17:58 PM8/27/05
to
On Sat, 27 Aug 2005 04:44:00 GMT, John Harshman
<jharshman....@pacbell.net> wrote:

snip>

zoe said:

>> (Once again, Steven, your statement that "depending on what traits one
>> emphasizes and compares, *different* tree-like hierarchies can be
>> obtained" says it all. As to the traits chosen for the evolutionary
>> tree, can you defend why those traits are emphasized and are the only
>> traits that are correct? What makes those choices the correct ones?)
>
>Steven was talking there about arbitrary hierarchies that you can get
>from anything, like, for example, sweaters. He wasn't talking about
>biological hierarchies, because those do not depend on what traits one
>emphasizes. That's the point of the multiple nested hierarchy.

I suppose biological hierarchies rose up by themselves and flung
themselves in your face so you didn't have to emphasize any particular
traits? Come on. Since when is the biological field exempt from
arbitrary choices as to what traits to emphasize?

>> So...since Steven has not yet persuasively defended these premises, I
>> am asking again: Can anyone else (or Steven again) better defend
>> these premises?
>
>Objection, your honor: asked and answered.

Overruled, Mr. Harshman. Please sit down.

Zoe

unread,
Aug 27, 2005, 10:16:15 PM8/27/05
to
On Sat, 27 Aug 2005 03:34:12 +0000 (UTC), bdbr...@mail.utexas.edu
(Bobby D. Bryant) wrote:

>On Sat, 27 Aug 2005, Zoe <muz...@aol.com> wrote:

snip>

>> Please support these premises:
>>
>> Premise 1: Similarity is sufficient evidence that life forms should
>> be grouped together in relationship.
>
>You need to realize that scientists and creationists start from the
>same point: an observation that something looks unlikely to be the
>result of pure chance. (From there they diverge, with the creationist
>jumping immediately to "goddidit" and the scientist looking for an
>actual cause for the unlikely something.)
>
>Do you think the genetic and morphological similarities that guide
>the construction of the tree are the result of mere chance, or the
>result of some cause?

well, from the evoluionist's point of view, the construction of the
tree has to be the result of mere chance since the modus operandi of
evolution is random chance mutations, selected for. Selection,
itself, is a random activity since what gets selected are random
events. It is by mere chance that mutations happen to produce life
forms that appear similar.

From a creationist point of view, the construction of the tree is not
a result of mere chance. It is the result of humans sitting down and
classifying already-formed life forms into a tree of their choosing.

B Richardson

unread,
Aug 27, 2005, 10:57:09 PM8/27/05
to
Zoe wrote:
>On Sat, 27 Aug 2005 04:44:00 GMT, John Harshman
><jharshman....@pacbell.net> wrote:
>
>snip>
>
>zoe said:
>
>>> (Once again, Steven, your statement that "depending on what
traits one
>>> emphasizes and compares, *different* tree-like hierarchies can
be
>>> obtained" says it all. As to the traits chosen for the
evolutionary
>>> tree, can you defend why those traits are emphasized and are
the only
>>> traits that are correct? What makes those choices the correct
ones?)
>>
>>Steven was talking there about arbitrary hierarchies that you can
get
>>from anything, like, for example, sweaters. He wasn't talking
about
>>biological hierarchies, because those do not depend on what
traits one
>>emphasizes. That's the point of the multiple nested hierarchy.
>
>I suppose biological hierarchies rose up by themselves and flung
>themselves in your face so you didn't have to emphasize any
particular
>traits? Come on. Since when is the biological field exempt from
>arbitrary choices as to what traits to emphasize?
>

Thats the advantage of working with nucleotides. You can do
phylogenetic analysis without emphasizing traits or making
arbitrary choices. The phylogenetic signal is just there and
there's not anything one can do to make it go away. A dozen
different people could independantly construct the same
phylogenies from nucleotides with unknown function from
unidentified species.

>
>>> So...since Steven has not yet persuasively defended these
premises, I
>>> am asking again: Can anyone else (or Steven again) better
defend
>>> these premises?
>>

John Drayton

unread,
Aug 27, 2005, 11:09:18 PM8/27/05
to
Zoe wrote:

<snip>


> well, from the evoluionist's point of view, the construction of the
> tree has to be the result of mere chance since the modus operandi of
> evolution is random chance mutations, selected for. Selection,
> itself, is a random activity since what gets selected are random
> events.

Are you are seriously claiming that because a process
has an input which is random, the process itself is
random?

<snip>

--
John Drayton

Zoe

unread,
Aug 27, 2005, 11:39:59 PM8/27/05
to
On Sat, 27 Aug 2005 00:37:43 -0500, "Steven J."
<sjt195...@nts.link.net.INVALID> wrote:

>
>"Zoe" <muz...@aol.com> wrote in message
>news:t9gvg11b6hi7itp4b...@4ax.com...

snip>

>> Please support these premises:
>>
>> Premise 1: Similarity is sufficient evidence that life forms should
>> be grouped together in relationship.
>>
>Why should we support that premise? It is wrong, and no one is basing an
>argument for common descent on it.

oh really? Why has it been presented to me as THE argument then?
"The pattern of similarity/dissimilarity that forms a nested hierarchy
is evidence for common descent." Has this not been said before, and
most strongly emphasized by John Harshman himself, your foremost
expert on this matter, as you have conceded?

> "Similarity" in some respects is
>rejected as a basis for grouping organisms together. For example, despite
>similarities in shape and habitat, whales are not grouped apart from other
>mammals and together with, e.g. sharks and ichthyosaurs. The box-camera
>eyes of octopuses are not seen as a reason to group them with vertebrates,
>who have superficially "similar" eyes.

notice the subjectivity here. "Not seen as a reason to group." What
would be a reason? So far none of this ties back to any evidence. It
is merely the arbitrary decision of those with a particular mindset.
In the creationist theory of intelligence, what is seen in nature is
tied back to what is seen in human intelligence. What do you tie your
groupings back to, but an arbitrary decision?

>Of course, the similarities of the eyes of cephalopods and vertebrates does
>not extend beyond the features necessary to a "box-camera" eye (with lens,
>retina, etc.). Famously, vertebrate eyes have an "inverted retina" and
>blind spot, and cephalopod eyes do not, among other differences. Likewise,
>the common features of whales, sharks, and ichthyosaurs (aside from those
>shared by all vertebrates) are adaptions to the way they live and move.

because you say so, or because it has been observed? Anybody can look
at the function of an eye and say, see that other type of eye over
there; this is an adaptation of that. What evidence do you have that
it was an adaptation and not originally the way it is?

>It was for this reason that I brought up "homology" -- detailed similarities
>not required by similarity in function. A bat's wing doesn't need all the
>similarities it has to a mouse's forelimb, and the detailed correspondence
>of the bones in both limbs is remarkable and calls for some explanation.

the explanation the creationist would offer is that intelligence chose
to create it that way. From that point of acknowledgement, science
can proceed with its research on how it operates, how it can be
copied, how new ideas can be derived from how it was made.
Speculation about the history of the eye contributes nothing to
science.

>Note, also, that there are *many* shared homologies between bats and other
>mammals. Finding some homologies between bats and mice, but not between
>bats and birds, and different homologies between bats and birds, but not
>with mice, and still different homologies between bats and pterosaurs (but
>not with birds or mice) would not support the idea that these homologies are
>explained by inheritance from a common ancestor. Presumably, branching
>descent with modification would pass on entire suites of traits together --
>and, therefore, entire suites of shared homologies arranged in a consistent
>nested hierarchy argues for branching descent with modification.

you have not yet answered my direct question as to why
similarity/dissimilarity is a justifiable tool for determining
relationship between species. Has it been observed that when things
have a pattern of similarity and dissimilarity anywhere else than in
life forms, that relationship is the conclusion?

>> (And, Steven, homology is all about similarity of structure and is an
>> evolutionary word when used in biology. Inherent in that word is an
>> evolutionary meaing, the very premise of which I am questioning --
>> that similarity means relationship. Homology is the supposition that
>> life forms correspond in basic type of structure and derive from a
>> common primitive origin; i.e., the wing of a bat and the foreleg of a
>> mouse are homologous (Webster's). If homology is about similarity,
>> then it is a weak defense to use "similarity" (dressed up in another
>> word) as your support for the premise that similarity means
>> relationship.
>>
>Zoe, the word "homology" was coined by the comparative anatomist Sir Richard
>Owens, an opponent of Darwin who rejected common descent and adaption by
>natural selection. The concept is older than Owen; it was used by Carrolus
>Linnaeus as the basis for his taxonomy. Linnaeus, also, had no concept of
>evolution (except in some cases within genera). Homologies can be
>recognized without reference to common descent. Homology does not mean
>merely "similarity;" it distinguishes one sort of similarity (detailed in
>many aspects of a feature, not just those needed for common function) from
>another sort -- "analogy," or similarity in features required for common
>function. The similarity between octopus and human eyes is "analogous;"
>that between human eyes and other vertebrate eyes is "homologous."

you are still maintaining, without defending your position, that
similarity means relationship between species. You are identifying
similarities that are analogous and homologous, but not defending why
such similarities must mean relationship between groups.

snip>

>> To demonstrate premise 1, there can be no exceptions to the similarity
>> rule. If exceptions are allowed, then exceptions should be allowed to
>> other theories that differ from yours.
>>
>I don't think I agree with that. Are exceptions to "creation theory"
>explicable?

there are no exceptions to the identifying characteristics of how
mental activity behaves. Perceived exceptions are explicable.

> Note that what you wanted was to be able to say that any entity
>which possessed some particular, subjectively defined set of properties was
>the product of intelligence.

first of all, the properties are not subjectively defined. They are
observed objectively. And isn't it reasonable that if those
properties are found nowhere else outside of intelligent manipulation,
that they form a reliable standard for identifying mental activity?

> It rather makes a difference if nondesigned
>entities can possess these properties.

you have taken a position that certain entities are nondesigned,
without defending your position that they are nondesigned. You are
starting with a settled determination that certain things are
nondesigned. But it is those very things that you claim are
nondesigned that are being examined.

> Evolution is inferred from a very
>large number of features of living things, and the way in which they are
>arranged among different living things.

the way in which they are arranged are described on the basis of
similarities, right?

>> Premise 2: Dissimiliarity is sufficient evidence that life forms
>> should be grouped together in relationship.
>>
>> (And, again, Steven, I don't know why you would say that dissimilarity
>> is not one of the standards used in declaring relationship. I have
>> been told by posters here that the pattern of similarity/dissimilarity
>> in life forms is evidence for relationship.)
>>
>First of all, Zoe, obviously related things can be dissimilar. You can tell
>members of a litter of puppies apart, although they share the same parents.
>You can tell different breeds of dogs apart, even though they share common
>ancestors. Now, the dissimilarities between, say, a Newfoundland and a
>chihuahua are not evidence for common ancestry, but they are clearly
>compatible with common ancestry. Moreover, they help us group the different
>breeds of dogs together to say which breeds are more closely related to
>others, and which are more distantly related. The arguments for common
>ancestry (and branching order) of different primates are similar in
>principle to those for classifying different breeds in their own nested
>hierarchy.

please note that your example here refers to within the canidae
species. How do you make the leap from canidae to, say, hominids, or
from plants to animals?

>Now, since you seem to have your own idiosyncratic take on "creation
>theory," I'm not sure how much weight more ... conventional ... creationist
>ideas will have with you. Most creationists hold to an idea of "kinds" above
>the species level, which include multiple species (in most cases; humans are
>always a kind to themselves) which are indeed related by common descent. A
>typical creationist would hold that, e.g. not merely different breeds of
>_Felis catus_ (the domestic cat) but different members of _Felis_, perhaps
>even different members of the family Felidae (e.g. lions, pumans, bobcats,
>and housecats) are members of a single "kind." Of course, the grouping of
>species into "kinds" requires this same sort of argument for relationship
>from a pattern of similarities and dissimilarities; creationists fail to be
>consistent in the application of this principle (e.g. if you group lions and
>housecats together, you have no reason not to group humans and chimps
>together).

personally, I think it is an exercise in wasted effort to try to
classify life forms to any degree beyond the most general. The vast
variety of life forms (and more being discovered still) defy
categorization beyond the most general categories.

You may manage to pigeonhole all life forms, but of what use is this
to the advance of science other than to say, we have them categorized
now; see, evolutionary theory is right.

>Now, beyond this, recall some of the above points about analogy and
>homology. Birds of widely different sizes, habits, and diet share the same
>basic avian wing design. Likewise, bats of different sizes, habits and
>diets share the same chiropteran wing design. It seems, on the principle of
>"common design for common function," that small insect-eating birds and
>small insect-eating bats should have more similar wings than, e.g. small
>insect-eating birds and larger, fruit-eating birds. Likewise, we might
>expect greater homologies between toucans and fruit bats than between fruit
>bats and brown bats. Basic wing designs don't seem to be arranged according
>to ecological niche, but fall into a pattern suggesting, again, descent with
>ad hoc modification. Fruit bats and brown bats have similar wings because
>they inherited them from a common ancestor; toucans, hummingbirds, swallows,
>and penguins share similar wing designs because, again, a common avian
>ancestor had wings like that.

and certain ants with a certain mindset, exploring the world of humans
would think that human artifacts that share similar designs must have
evolved.

>Note that we know about mechanisms of reproduction, mutation, and selection,
>and know that these processes can produce suites of organisms that fall into
>a consistent nested hierarchy.

granted, you know about reproduction. But mutation and selection are
not known in the way that reproduction is known. Reproduction is not
speculative. Mutation and selection are speculated to have affected
reproduction in such a way as to cause evolution. These speculations
have not been supported. They are not in the same class as
reproduction.

>Again, we know that consistent nested
>hierarchies typically result from a process of branching descent with
>modification (e.g. languages, breeds of domestic animals, manuscript
>copies), and that artifacts typically do not (just to complicate matters,
>lateral exchange of traits is possible for all these things -- e.g. gene
>transfer in bacteria; it is rare enough in all these cases that it does not
>mess up the nested hierarchy, at least for eukaryotes).

you are promoting nested hierarchies without defending the mechanism
used to develop these hierarchies. I want you to defend why you think
that the similarities/dissimilarities that are teased out to form a
nested hierarchy are sufficient to conclude relationship between
groups.

>> To demonstrate this, there can be no exceptions to the dissimilarity
>> rule. If exceptions are allowed, then exceptions should be allowed to
>> other theories that differ from yours.
>>
>Since you don't understand the "dissimilarity rule," this condition is
>nonsense.

what I understand of your dissimilarity rule is that if the same
structure is used for a different function, then this, too, means
relationship through common descent. Did I understand that
incorrectly?

>> Premise 3: A tree-like pattern is sufficient evidence that life forms
>> are related through a single common ancestor.
>>
>That, again, is not quite my point, although it is at least within an
>astronomical unit of my point.

and can you defend even this astronomical unit of your point?

>> (Once again, Steven, your statement that "depending on what traits one
>> emphasizes and compares, *different* tree-like hierarchies can be
>> obtained" says it all. As to the traits chosen for the evolutionary
>> tree, can you defend why those traits are emphasized and are the only
>> traits that are correct? What makes those choices the correct ones?)
>>
>Zoe, take any traits you like. Some (e.g. the similarity in shape that
>links sharks and dolphins, or the similarity in color that links some
>parrots and some iguanas) will yield trees that look very, very different
>from any phylogenic tree drawn by evolutionists. Of course, a tree based on
>overall color and one based on overall shape will yield very different trees
>from each other. But draw trees based on comparative limb anatomy, or
>comparative skull anatomy, or on various sets of genes, or on various other
>sets of genes (after all, you have a lot of sets of genes to compare), or on
>proteins (which don't have to form the same tree as genes, since different
>sets of codons can code for the same amino acids), and you will find that
>they form trees that look very much alike.

Um-hum. So can you defend the position that because trees look alike
(are similar) that the conclusion HAS to be relationship of different
groups through a common ancestor?

> As I noted in an earlier post,
>*all* animals with one bone in the lower jaw and three bones in the middle
>ear also have (although these features hardly seem necessary to each other)
>mammary glands and a left (but not a right) aortic arch.
>
>Why do not *all* trees yield exactly the same pattern? Well, first of all,
>why should they? The most indicative features are those which are not
>strongly adaptive to or obvious consequences of a particular lifestyle. For
>example, lots of arctic mammals have white fur, presumably as camoflage
>against the snow. Whiteness is a sign of adaption to common function,

whiteness is not always a sign of adaptation to common function. You
know that.

>and
>therefore would be expected even if the animals did not share common
>ancestry (and, indeed, polar bears and arctic foxes don't share a closer
>ancestry with each other than they do with, respectively, brown bears and
>red foxes). But there's no obvious reason that, e.g. giraffes and mice and
>humans should all share seven bones in the neck, or even three bones in the
>middle ear (many vertebrates hear well without this feature).

judging from how mental activity behaves, there are good reasons to
use a similar template for different purposes. Judging from how
random mutations behave, there are no good reasons for similar
templates to be passed on....unless you are willing to acknowledge
that a fully-functioning reproductive system is already in place, and
is then affected by mutations.

>> So...since Steven has not yet persuasively defended these premises, I
>> am asking again: Can anyone else (or Steven again) better defend
>> these premises?
>>
>To someone, like yourself, determined not to understand or be persuaded, I
>rather doubt it. But I took a shot.

thanks for the shot, Steven. It did miss the mark -- again -- but
thanks anyway.

Bobby D. Bryant

unread,
Aug 27, 2005, 11:44:48 PM8/27/05
to
On Sun, 28 Aug 2005, Zoe <muz...@aol.com> wrote:

> On Sat, 27 Aug 2005 03:34:12 +0000 (UTC), bdbr...@mail.utexas.edu
> (Bobby D. Bryant) wrote:
>
>>On Sat, 27 Aug 2005, Zoe <muz...@aol.com> wrote:
>
> snip>
>
>>> Please support these premises:
>>>
>>> Premise 1: Similarity is sufficient evidence that life forms should
>>> be grouped together in relationship.
>>
>>You need to realize that scientists and creationists start from the
>>same point: an observation that something looks unlikely to be the
>>result of pure chance. (From there they diverge, with the creationist
>>jumping immediately to "goddidit" and the scientist looking for an
>>actual cause for the unlikely something.)
>>
>>Do you think the genetic and morphological similarities that guide
>>the construction of the tree are the result of mere chance, or the
>>result of some cause?
>
> well, from the evoluionist's point of view, the construction of the
> tree has to be the result of mere chance since the modus operandi of
> evolution is random chance mutations, selected for. Selection,
> itself, is a random activity since what gets selected are random
> events.

So gravity is random too?

> It is by mere chance that mutations happen to produce life
> forms that appear similar.
>
> From a creationist point of view, the construction of the tree is not
> a result of mere chance. It is the result of humans sitting down and
> classifying already-formed life forms into a tree of their choosing.

By your logic about selection, the construction of the tree would be
random too.

But nevermind; that's tangential to the issue at hand.

And you did't actually answer my quoted question. So, do you think


the genetic and morphological similarities that guide the construction
of the tree are the result of mere chance, or the result of some
cause?

Notice that I'm asking about the similarities, not about what scientists
and creationists do with them. To the options "chance" and "caused",
you can also add "no such similarities exist" as an answer.

And as before, where we go from here depends on how you answer the
question. (No progress can be made if you avoid it.)

John Drayton

unread,
Aug 27, 2005, 11:45:16 PM8/27/05
to
Zoe wrote:

<snip>

> (And, Steven, homology is all about similarity of structure and is an
> evolutionary word when used in biology.

An "evolutionary" word? The word was already in use
in a biological context before Charles Darwin published
The Origin of Species in 1859.

A little research will show you that the word was in
use in biological circles well before TOOS was published,
and was in use by people who were clearly not evolutionists
or Darwinists:

The modern distinction between "homologue" and
"analogue" was established by Richard Owen in 1843.
Owen defined an analogue as "a part or organ in one
animal that has the same function as another part or
organ in a different animal," while a homologue is
"the same organ in different animals under every
variety of form and function." (Lectures on Invertebrate
Animals, pp 374,379, 1843.) Thus the leg of a crab
or insect is "analagous" to the limb of a quadruped,
because they all serve for locomotion, but they are
not "homologous." Homologous organs have similar
structure or bear similar relations to other organs,
whereas analagous organs have similar functions.

Refs:
http://www.ucmp.berkeley.edu/history/owen.html
http://www.christianhubert.com/hypertext/analogy___homology.html

> Inherent in that word is an
> evolutionary meaing, the very premise of which I am questioning --
> that similarity means relationship. Homology is the supposition that
> life forms correspond in basic type of structure and derive from a
> common primitive origin;

Wrong. Homology is a word describing observations about
similarities made *independantly* from, and *prior to*
evolutionary theory, by people who did *not* believe in
a common primitive origin.

It just so happens that evolutionary theory explains
homology better than any other theory.

<snip>

--
John Drayton

Zoe

unread,
Aug 27, 2005, 11:43:35 PM8/27/05
to
Sat, 27 Aug 2005 12:21:52 GMT, "R. Baldwin"
<res0...@nozirevBACKWARDS.net> wrote:

snip>

>The tree of life is not built on premises, Zoe. It is built on observations.

observations trigger premises. One observation is that there are
similarities between life forms. The premise triggered in some minds
is that these similarities must mean relationship between species
through a common ancestor. It is this premise that I want you to
defend, not the observations. Creationists also observe similarities,
you know.

snip tree commentary, for which I will start a new thread.

>I would note, by the way, that given your wording, Premises 1 and 2 are
>mutually contradictory.

don't blame me for that. Ask Harlequin or Harshman or Steven why they
have said just that -- that it is the pattern of similarities and
dissimilarities that are evidence for relationship.

snip>

Steven J.

unread,
Aug 27, 2005, 11:56:32 PM8/27/05
to

"Zoe" <muz...@aol.com> wrote in message
news:q062h11i2to2novd0...@4ax.com...
I am not familiar with _World of Warcraft_. Do these characters spawn
offspring with variant traits (whose variations are inheritable)? Can you
"selectively breed" these characters to change their traits over time? If
the characters do not have these traits, they are not enough like actual
living things for this example to be different from "you can sort cars or
knitware into nested hierarchies."

Note: since I am not familiar with _World of Warcraft_, I cannot say whether
the various creatures -- the yetis, raptors, and whatnot -- fall into a
consistent nested hierarchy, or whether, like centaurs and bucentaurs and
fauns, they fall into different nested hierarchies depending on which traits
one selects for comparison. Science fiction writers occasionally, and game
designers conceivably might, make an effort to make the fauna of their
invented worlds look like the result of common descent with modification.
Perhaps your "evolutionist ant" would not be able to conceive of any
conceivable phylogeny that would make sense of the different creatures in
the simulated world, or perhaps it would infer the the evolutionary history
that the creators fashioned for their world. By analogy, perhaps the
Creator tried to make *this* world look like the result of evolution (in
which case He did an award-worthy job, IMHO).


>
>>3) Not every form of similiarity is considered relevant by
>>modern evolutionary biologists. If one uses cladistic methodology
>>then one only looks at synapomorphies,"shared derived homologies,"
>>to determine relationships. I am not sure that I know how to
>>make that language simple enough for you I fear.
>
> haven't I made MY language simple enough for you to understand? I am
> simply asking for a defense of the concept that if you are able to
> group life forms according to their similarities/dissimilarities, why
> does this ability to do so lead you to the conclusion that, therefore,
> these life forms must be related to each other through a common
> ancestor?
>

Your language is simple enough to make it clear that you either do not
understand or are deliberately misrepresenting the arguments presented to
you.


>
>>> (And, Steven, homology is all about similarity of structure and is an
>>> evolutionary word when used in biology. Inherent in that word is an
>>> evolutionary meaing,
>>
>>Actually homology can be defined completely independently of evolution.
>
> you're on a tangent already, Harlequin. I said it is an evolutionary
> word WHEN USED IN BIOLOGY. I agree with you that it can be defined
> completely independently of evolution. Steven was using the word in
> an evolutionary sense.
>

Homology, when used in biology, can be defined completely independently of
evolution. Richard Owens did just that almost two centuries ago. By
implication, any creationist who used the explanation "common design, not
common descent," admits that homology in biology can be recognized without
assuming common descent.


>
>>Indeed noticing that certain parts on one type of organism correspond
>>to parts of another organism certainly long predates Darwin.
>>
>>> the very premise of which I am questioning --
>>> that similarity means relationship.
>>
>>I would question it too. It is an oversimplification at best
>>and a strawman at worst.
>
> okay, so what are the basic tools used to determine relationship
> between species if not similarity/dissimilarity? And if similarity is
> one of the key tools used, I would like you to support the premise
> that such a tool is valid in concluding relationship between species.
>

http://en.wikipedia.org/wiki/Cladistics
http://www.fossilnews.com/1996/cladistics.html

We know of a mechanism -- inheritance, combined with mutation and
modification through mutation and natural selection, and with extinction of
some intermediate forms -- which can explain why groups of organisms fall
into consistent nested hierarchies. Again, note that these hierarchies are
*consistent* -- it is not as though one set of traits supports one tree of
relationships and another set supports a completely different tree.
Multiple inconsistent hierarchies would be consistent with separate origins
using a common pool of design features (or perhaps with massive lateral
transfer of cooperating genes, but no mechanism is known, other than
advanced genetic engineering, that could accomplish that). A single
hierarchy (combined with apparent vestiges of history, like the vestigial
centromere and telomere in human chromosome 2, or the plantaris tendon)
strongly suggests that the homologies originate through modification of
traits inherited together.


>
>>> Homology is the supposition that
>>> life forms correspond in basic type of structure and derive from a
>>> common primitive origin; i.e., the wing of a bat and the foreleg of a
>>> mouse are homologous (Webster's). If homology is about similarity,
>>> then it is a weak defense to use "similarity" (dressed up in another
>>> word) as your support for the premise that similarity means
>>> relationship.
>>
>>
>>While the wing and leg are homologous, I would remind you that
>>a dictionary can be a dangerous thing to rely on here.
>
> now really. Dangerous? Hmmm.
>

A little knowledge is a dangerous thing; one can assume that having mastered
a few facts, one has mastered the subject.


>
>>> To demonstrate premise 1, there can be no exceptions to the similarity
>>> rule. If exceptions are allowed, then exceptions should be allowed to
>>> other theories that differ from yours.
>>>
>>> Premise 2: Dissimiliarity is sufficient evidence that life forms
>>> should be grouped together in relationship.
>>
>>This premise is outright false. Dissimiliarity _can_ be evidence for
>>not having a (close) relationship.
>
> and haven't I been told that it CAN also be evidence for relationship?
>

No, you were told that it can be evidence that a relationship (supported by
similarities) is not as close as the relationship between two organisms with
greater similarities.


>
>>> (And, again, Steven, I don't know why you would say that dissimilarity
>>> is not one of the standards used in declaring relationship. I have
>>> been told by posters here that the pattern of similarity/dissimilarity
>>> in life forms is evidence for relationship.)
>>
>>I am one of them. And your post here strongly suggests that you
>>do not understand what we mean by ths.
>
> well, explain yourself then. You leave it hanging here with the
> mysterious suggestion that there is meaning to this, but you're not
> saying what it is.
>
> snip rest because I would like direct answers to the above first>
>

-- Steven J.


John Harshman

unread,
Aug 28, 2005, 12:00:03 AM8/28/05
to
Zoe wrote:

> On 26 Aug 2005 19:54:17 -0700, "Harlequin" <use...@cox.net> wrote:
>
>
>>Zoe <muz...@aol.com> wrote in
>>news:t9gvg11b6hi7itp4b...@4ax.com:
>
>
> snip>
>
>>>Please support these premises:
>>>
>>>Premise 1: Similarity is sufficient evidence that life forms should
>>>be grouped together in relationship.
>>
>>
>>You have Premise 2 below with the same sentence with "Dissimiliarity"
>>instead of "Similiarity."
>>
>>I will have to disagree with this.
>>
>>1) Similiarity is _NOT_ sufficent evidence that life forms should
>>be grouped together in a relationship in evolutionary biology.
>
>
> okay, I'll rephrase.
>
> Please defend the premise that similarity can be justifiably used as a
> standard for determining relationship between species.

I don't know how many times it should be necessary to say this:
similarity is not used as such a standard. Not formally, by scientists,
at least.

>>2) Similiarity and dissimiliarity cannot be looked separately.
>
>
> okay, keep them together if you wish. Please justify why
> similarity/dissimilarity can be justifiably used as a standard for
> determining relationship between species.

That would be the nested hierarchy again.

> There is a game called World of Warcraft in which the characters are
> lifelike, can spawn and respawn, and can reproduce copies of
> themselves, depending on the goals of the creators of the game. After
> a while you begin to recognize templates for the patterns of
> similarities and dissimilarties within the game. I am betting that,
> based on the observation of reproduction and patterns of
> similarity/dissimilarity in the life forms in the game, an
> evolutionist "ant" might wander around within that game world and
> conclude that the yeti are more closely related to the raptor and thus
> have a common ancestor, or that the horde and alliance characters have
> a common ancestor, and so on. Coming from the "ant's" perspective,
> this might seem a reasonable conclusion based on an arbitrary decision
> to call patterns of similarity/dissimilarity evidence for relationship
> between species. But coming from the creators of the game's
> perspective, nothing could be further from the truth.

Once again you present an example of what you think, without actual
evidence, might possibly produce a nested hierarchy. I doubt it. And so
that effort was pointless.

>>3) Not every form of similiarity is considered relevant by
>>modern evolutionary biologists. If one uses cladistic methodology
>>then one only looks at synapomorphies,"shared derived homologies,"
>>to determine relationships. I am not sure that I know how to
>>make that language simple enough for you I fear.
>
> haven't I made MY language simple enough for you to understand? I am
> simply asking for a defense of the concept that if you are able to
> group life forms according to their similarities/dissimilarities, why
> does this ability to do so lead you to the conclusion that, therefore,
> these life forms must be related to each other through a common
> ancestor?

Because that's the only process we know of that produces a nested
hierarchy. If you would like to present evidence for a different process
that leads to this pattern, please do so.

>>>(And, Steven, homology is all about similarity of structure and is an
>>>evolutionary word when used in biology. Inherent in that word is an
>>>evolutionary meaing,
>>
>>Actually homology can be defined completely independently of evolution.
>
> you're on a tangent already, Harlequin. I said it is an evolutionary
> word WHEN USED IN BIOLOGY. I agree with you that it can be defined
> completely independently of evolution. Steven was using the word in
> an evolutionary sense.

What's your point here? Homology was originally defined with a
non-evolutionary meaning. Evolution was then advanced as the explanation
for observed homology. Then homology was redefined (comparatively
recently) to explicitly feature evolution. Is that a problem of some kind?

>>Indeed noticing that certain parts on one type of organism correspond
>>to parts of another organism certainly long predates Darwin.
>>
>>
>>>the very premise of which I am questioning --
>>>that similarity means relationship.
>>
>>I would question it too. It is an oversimplification at best
>>and a strawman at worst.
>
> okay, so what are the basic tools used to determine relationship
> between species if not similarity/dissimilarity? And if similarity is
> one of the key tools used, I would like you to support the premise
> that such a tool is valid in concluding relationship between species.

Similarity is not such a tool. As has been explained many times, it's
the pattern of similarities and dissimilarities, i.e. the nested
hierarchy, that is the tool you are groping after so blindly.

>>>Homology is the supposition that
>>>life forms correspond in basic type of structure and derive from a
>>>common primitive origin; i.e., the wing of a bat and the foreleg of a
>>>mouse are homologous (Webster's). If homology is about similarity,
>>>then it is a weak defense to use "similarity" (dressed up in another
>>>word) as your support for the premise that similarity means
>>>relationship.
>>
>>
>>While the wing and leg are homologous, I would remind you that
>>a dictionary can be a dangerous thing to rely on here.
>
> now really. Dangerous? Hmmm.

In your case, any reference that could possibly be misconstrued is
similarly dangerous.

>>>To demonstrate premise 1, there can be no exceptions to the similarity
>>>rule. If exceptions are allowed, then exceptions should be allowed to
>>>other theories that differ from yours.
>>>
>>>Premise 2: Dissimiliarity is sufficient evidence that life forms
>>>should be grouped together in relationship.
>>
>>This premise is outright false. Dissimiliarity _can_ be evidence for
>>not having a (close) relationship.
>
>
> and haven't I been told that it CAN also be evidence for relationship?

Not to my knowledge. Given your tendency to misunderstand just about
everything you read, I doubt it extremely.

>>>(And, again, Steven, I don't know why you would say that dissimilarity
>>>is not one of the standards used in declaring relationship. I have
>>>been told by posters here that the pattern of similarity/dissimilarity
>>>in life forms is evidence for relationship.)
>>
>>I am one of them. And your post here strongly suggests that you
>>do not understand what we mean by ths.
>
> well, explain yourself then. You leave it hanging here with the
> mysterious suggestion that there is meaning to this, but you're not
> saying what it is.

It's been explained hundreds of times before (and I don't think that's
an exaggeration). Including a few times in the post above.

Steven J.

unread,
Aug 28, 2005, 12:01:30 AM8/28/05
to

"Zoe" <muz...@aol.com> wrote in message
news:kj72h11t2kqilbptv...@4ax.com...

> On Sat, 27 Aug 2005 04:44:00 GMT, John Harshman
> <jharshman....@pacbell.net> wrote:
>
> snip>
>
> zoe said:
>
>>> (Once again, Steven, your statement that "depending on what traits one
>>> emphasizes and compares, *different* tree-like hierarchies can be
>>> obtained" says it all. As to the traits chosen for the evolutionary
>>> tree, can you defend why those traits are emphasized and are the only
>>> traits that are correct? What makes those choices the correct ones?)
>>
>>Steven was talking there about arbitrary hierarchies that you can get
>>from anything, like, for example, sweaters. He wasn't talking about
>>biological hierarchies, because those do not depend on what traits one
>>emphasizes. That's the point of the multiple nested hierarchy.
>
> I suppose biological hierarchies rose up by themselves and flung
> themselves in your face so you didn't have to emphasize any particular
> traits? Come on. Since when is the biological field exempt from
> arbitrary choices as to what traits to emphasize?
>
I believe I addressed this in my earlier answer to you. Living organisms
have huge arrays of traits. Even fossil skeletons have huge numbers of
identifiable features that can be compared, though hardly the myriad details
that are provided by soft tissue and, above all, by protein and DNA
sequences. Compare any traits you like. You will find a few -- explicable
as direct adaptions to or results of the organism's environment -- yield a
diverse set of contradictory trees (e.g. dolphins classed with sharks on the
basis of overall shape, or with elephants on the basis of overall color),
while the overwhelming majority of traits result in a set of trees that are
very, very similar to each other. You may recall (though I doubt you
understood) how Harshman and Richardson discussed how the same tree of ape
relationships could be derived by comparing very different sets of DNA from
the various ape species.

>
>>> So...since Steven has not yet persuasively defended these premises, I
>>> am asking again: Can anyone else (or Steven again) better defend
>>> these premises?
>>
>>Objection, your honor: asked and answered.
>
> Overruled, Mr. Harshman. Please sit down.
>
-- Steven J.


John Harshman

unread,
Aug 28, 2005, 12:04:48 AM8/28/05
to
Zoe wrote:

> On Sat, 27 Aug 2005 04:44:00 GMT, John Harshman
> <jharshman....@pacbell.net> wrote:
>
> snip>
>
> zoe said:
>
>
>>>(Once again, Steven, your statement that "depending on what traits one
>>>emphasizes and compares, *different* tree-like hierarchies can be
>>>obtained" says it all. As to the traits chosen for the evolutionary
>>>tree, can you defend why those traits are emphasized and are the only
>>>traits that are correct? What makes those choices the correct ones?)
>>
>>Steven was talking there about arbitrary hierarchies that you can get
>
>>from anything, like, for example, sweaters. He wasn't talking about
>
>>biological hierarchies, because those do not depend on what traits one
>>emphasizes. That's the point of the multiple nested hierarchy.
>
>
> I suppose biological hierarchies rose up by themselves and flung
> themselves in your face so you didn't have to emphasize any particular
> traits? Come on. Since when is the biological field exempt from
> arbitrary choices as to what traits to emphasize?

That's actually one of the big advantages of DNA sequences. The data
*do* fling themselves in your face without you making any choice. All
you do is pick a particular gene to sequence. And that choice is
arbitrary, since (in the human/ape case) we have demonstrated recently
that any gene you pick will give you the same result.

>>>So...since Steven has not yet persuasively defended these premises, I
>>>am asking again: Can anyone else (or Steven again) better defend
>>>these premises?
>>
>>Objection, your honor: asked and answered.
>
>
> Overruled, Mr. Harshman. Please sit down.

Objection. You are playing both judge and prosecutor (jury too, apparently).

Fury said to a mouse,
That he met in the
house, 'Let us
both go to law:
I will prosecute
you.-- Come, I'll
take no denial;
We must have
a trial: For
really this
morning I've
nothing to do.'
Said the mouse
to the cur,
'Such a trial,
dear Sir, With
no jury or
judge, would
be wasting
our breath.'
'I'll be
judge, I'll
be jury,'
Said cunning
old Fury:
'I'll try
the whole
cause, and
condemn
you
to
death.'

John Drayton

unread,
Aug 28, 2005, 12:03:41 AM8/28/05
to
Steven J. wrote:
> "Zoe" <muz...@aol.com> wrote in message
> news:t9gvg11b6hi7itp4b...@4ax.com...

<snip>

> Zoe, the word "homology" was coined by the comparative anatomist Sir Richard
> Owens, an opponent of Darwin who rejected common descent and adaption by
> natural selection. The concept is older than Owen; it was used by Carrolus
> Linnaeus as the basis for his taxonomy. Linnaeus, also, had no concept of
> evolution (except in some cases within genera). Homologies can be
> recognized without reference to common descent. Homology does not mean
> merely "similarity;" it distinguishes one sort of similarity (detailed in
> many aspects of a feature, not just those needed for common function) from
> another sort -- "analogy," or similarity in features required for common
> function. The similarity between octopus and human eyes is "analogous;"
> that between human eyes and other vertebrate eyes is "homologous."

Hmm. If I'd read the entire thread I probably wouldn't
have made my last post, which was just a lesser version
of the above :)

But anyway, it's probably a point that's worth re-iterating.

Zoe seems to be convinced that all these pieces of evidence
that support evolutionary theory have been invented by
evolutionary scientists, and are the result of their
evolutionary preconceptions.

--
John Drayton

Matthew Isleb

unread,
Aug 28, 2005, 1:29:38 AM8/28/05
to
On Sun, 28 Aug 2005 03:39:59 +0000, Zoe wrote:

<snip>

> please note that your example here refers to within the canidae
> species. How do you make the leap from canidae to, say, hominids, or
> from plants to animals?

If you follow evolutionary tree, you will see that it is not a leap at
all, but a matter of going back in time (on paper) to where the different
species diverged. Even tracing the relationship between dog breeds
requires "going back in time." It is just a matter of degree.


>>Now, since you seem to have your own idiosyncratic take on "creation
>>theory," I'm not sure how much weight more ... conventional ... creationist
>>ideas will have with you. Most creationists hold to an idea of "kinds" above
>>the species level, which include multiple species (in most cases; humans are
>>always a kind to themselves) which are indeed related by common descent. A
>>typical creationist would hold that, e.g. not merely different breeds of
>>_Felis catus_ (the domestic cat) but different members of _Felis_, perhaps
>>even different members of the family Felidae (e.g. lions, pumans, bobcats,
>>and housecats) are members of a single "kind." Of course, the grouping of
>>species into "kinds" requires this same sort of argument for relationship
>>from a pattern of similarities and dissimilarities; creationists fail to be
>>consistent in the application of this principle (e.g. if you group lions and
>>housecats together, you have no reason not to group humans and chimps
>>together).
>
> personally, I think it is an exercise in wasted effort to try to
> classify life forms to any degree beyond the most general.

Don't you mean "superficial" and not "general?" The most general
classification of known life forms is "carbon based" or "DNA having" or
somethign like that. Certainly you would want to be more specific than
that. If we only spoke in the most general, we would never know
what the other is talking about. IN the most general terms, a dog is not a
dog, but rather just a life form. Meet Spot, my pet life form. Oh wait,
"Spot" is too specific. There is no way we could possibly be so specific
as to give an individual life form its own category.... "Meet my pet life
form. We call it Life Form."

The vast
> variety of life forms (and more being discovered still) defy
> categorization beyond the most general categories.

You may want to see a neurologist about your difficulties categorizing
things and drawing relationships. You may have some brain damage.

> You may manage to pigeonhole all life forms, but of what use is this
> to the advance of science other than to say, we have them categorized
> now; see, evolutionary theory is right.

It is useful to know the relationships between different life forms for
medical research, for example. If we know the relationship between, say
mice and humans, we can better know how a drug that effects mice in one
way will effect humans. The more detailed we can make the relationships,
the better.

-matthew

Mark VandeWettering

unread,
Aug 28, 2005, 1:53:40 AM8/28/05
to
On 2005-08-28, Zoe <muz...@aol.com> wrote:
> On Sat, 27 Aug 2005 03:34:12 +0000 (UTC), bdbr...@mail.utexas.edu
> (Bobby D. Bryant) wrote:
>
>>On Sat, 27 Aug 2005, Zoe <muz...@aol.com> wrote:
>
> snip>
>
>>> Please support these premises:
>>>
>>> Premise 1: Similarity is sufficient evidence that life forms should
>>> be grouped together in relationship.
>>
>>You need to realize that scientists and creationists start from the
>>same point: an observation that something looks unlikely to be the
>>result of pure chance. (From there they diverge, with the creationist
>>jumping immediately to "goddidit" and the scientist looking for an
>>actual cause for the unlikely something.)
>>
>>Do you think the genetic and morphological similarities that guide
>>the construction of the tree are the result of mere chance, or the
>>result of some cause?
>
> well, from the evoluionist's point of view, the construction of the
> tree has to be the result of mere chance since the modus operandi of
> evolution is random chance mutations, selected for. Selection,
> itself, is a random activity since what gets selected are random
> events. It is by mere chance that mutations happen to produce life
> forms that appear similar.

It's hard to imagine the degree of willful stupidity that would allow
you to participate in this forum for as long as you have and yet remain
so completely clueless about the topics which you discuss.

> From a creationist point of view, the construction of the tree is not
> a result of mere chance. It is the result of humans sitting down and
> classifying already-formed life forms into a tree of their choosing.
>
>>Where the discussion goes from here depends on your answer to that.

Mark

Steven J.

unread,
Aug 28, 2005, 2:37:51 AM8/28/05
to

"Zoe" <muz...@aol.com> wrote in message
news:pcb2h15obr23oblin...@4ax.com...
I attempted to answer that question (before you asked it; in the future,
please try to read for comprehension) in the paragraph below. Having
defined "homology" as "detailed similarity not required for similarity of
function," and noting that, in fact, cephalopod eyes and vertebrate eyes do
not have detailed similarities in respects not necessary to function (cf.
the different retina designs), they are shown not to be homologous. Note
that *human* intelligence, if it were creating both cephalopods and
cetaceans, would be very likely to use the same eye design in both groups --
that is, they would share a homology in eyes with each other.

>
>>Of course, the similarities of the eyes of cephalopods and vertebrates
>>does
>>not extend beyond the features necessary to a "box-camera" eye (with lens,
>>retina, etc.). Famously, vertebrate eyes have an "inverted retina" and
>>blind spot, and cephalopod eyes do not, among other differences.
>>Likewise,
>>the common features of whales, sharks, and ichthyosaurs (aside from those
>>shared by all vertebrates) are adaptions to the way they live and move.
>
> because you say so, or because it has been observed? Anybody can look
> at the function of an eye and say, see that other type of eye over
> there; this is an adaptation of that. What evidence do you have that
> it was an adaptation and not originally the way it is?
>
I was worried about this use of "adaption." "Adaption," in this case,
simply means a feature that enables a species to better survive in a
particular environment. Whether it is a modification of some inherited
ancestral organ, or a designer-modified version of some "common design"
feature, is not critically relevant to the point of whether it is an
adaption.

>
>>It was for this reason that I brought up "homology" -- detailed
>>similarities
>>not required by similarity in function. A bat's wing doesn't need all the
>>similarities it has to a mouse's forelimb, and the detailed correspondence
>>of the bones in both limbs is remarkable and calls for some explanation.
>
> the explanation the creationist would offer is that intelligence chose
> to create it that way. From that point of acknowledgement, science
> can proceed with its research on how it operates, how it can be
> copied, how new ideas can be derived from how it was made.
> Speculation about the history of the eye contributes nothing to
> science.
>
The problem, of course, is that *any* feature can be "explained" by
"intelligence chose to create it that way" (except, of course, the minor
detail of how intelligence implemented its designs). Common descent can
explain why humans and chimpanzees have a GULO pseudogene disabled in the
same way, while guinea pigs have one disabled in a different way; it could
not explain why humans and guinea pigs have GULO pseudogenes disabled the
same way, while chimps have one disabled another way. Common descent could
not explain why gorillas have an octopus-style box-camera eye, while other
primates have the standard vertebrate model. That is, common descent can
explain homologies that fall into a consistent nested hierarchy, because
they ought to fall into such a pattern; "intelligence chose to create it
that way" is equally compatible with any pattern, but there is no
discernable reason for it to prefer consistent nested hierarchies unless it
wanted us to think the diversity of life arose through evolution.

>
>>Note, also, that there are *many* shared homologies between bats and other
>>mammals. Finding some homologies between bats and mice, but not between
>>bats and birds, and different homologies between bats and birds, but not
>>with mice, and still different homologies between bats and pterosaurs (but
>>not with birds or mice) would not support the idea that these homologies
>>are
>>explained by inheritance from a common ancestor. Presumably, branching
>>descent with modification would pass on entire suites of traits
>>together --
>>and, therefore, entire suites of shared homologies arranged in a
>>consistent
>>nested hierarchy argues for branching descent with modification.
>
> you have not yet answered my direct question as to why
> similarity/dissimilarity is a justifiable tool for determining
> relationship between species. Has it been observed that when things
> have a pattern of similarity and dissimilarity anywhere else than in
> life forms, that relationship is the conclusion?
>
Yes. See, e.g. trees of related languages, or trees of manuscripts derived
from copying copies of some lost original.

http://www.danshort.com/ie/iecentum_c.shtml
http://ahds.ac.uk/creating/case-studies/canterbury/

Sets of entities not related by shared ancestry (or "ancestry" -- copies of
the _Canterbury Tales_ do not reproduce themselves, but are reproduced, with
"mutations" or copying errors, by human scribes) do not fall into consistent
nested hierarchies and comparisons of their patterns of similarities and
dissimilarities do not yield consistent trees.

Note that, in general, manuscripts, or languages, or species, will tend to
be more like recent ancestors than like older ancestors separated from them
by more generations. English is more like Middle English than like Old
English, and more like Old English than like proto-Germanic. House cats are
more like the common ancestor of _Felis_ than like the common ancestor of
the Felidae, and more like that ur-felid than like the common ancestor of
carnivores. By the same token, other descendants of that recent common
ancestor will share more homologies, with fewer modifications, than will
fellow descendants only of more remote common ancestors (e.g. Frisian is
more like English than is German, and German is more like English than is
Spanish; house cats are more like bobcats than are lions, and more like
lions than they are like hyenas). Thus, the nested hierarchy is an expected
pattern of branching descent with modification (that is, descent with
modification in which there is more than one way different populations of
descendants can be modified).

I can't make a case for my position until I've explained to you what my
position *is*. You keep asking me to defend positions I've never taken, and
you don't seem to understand why this is an unreasonable demand. For that
matter, until you understand what homologies *are*, you cannot understand
the manner in which homologies and analogies imply common descent. Perhaps
you cannot understand, even when you do understand -- but I'm not sure we've
reached that point yet.


>
> snip>
>
>>> To demonstrate premise 1, there can be no exceptions to the similarity
>>> rule. If exceptions are allowed, then exceptions should be allowed to
>>> other theories that differ from yours.
>>>
>>I don't think I agree with that. Are exceptions to "creation theory"
>>explicable?
>
> there are no exceptions to the identifying characteristics of how
> mental activity behaves. Perceived exceptions are explicable.
>

I do not believe you.


>
>> Note that what you wanted was to be able to say that any entity
>>which possessed some particular, subjectively defined set of properties
>>was
>>the product of intelligence.
>
> first of all, the properties are not subjectively defined. They are
> observed objectively. And isn't it reasonable that if those
> properties are found nowhere else outside of intelligent manipulation,
> that they form a reliable standard for identifying mental activity?
>

You are rather far from demonstrating that these properties objectively
exist (e.g. that DNA is some sort of "control program," except by analogy),
or that they only arise from "mental activity." There is also the problem
that you have not demonstrated -- you have not even hinted at -- any means
by which "mental activity" can actually instantiate design in a material
object, except through some material mechanism.


>
>> It rather makes a difference if nondesigned
>>entities can possess these properties.
>
> you have taken a position that certain entities are nondesigned,
> without defending your position that they are nondesigned. You are
> starting with a settled determination that certain things are
> nondesigned. But it is those very things that you claim are
> nondesigned that are being examined.
>
>> Evolution is inferred from a very
>>large number of features of living things, and the way in which they are
>>arranged among different living things.
>
> the way in which they are arranged are described on the basis of
> similarities, right?
>

Read for comprehension, Zoe. As noted elsewhere, all animals with one bone
in the lower jaw and three bones in the middle ear also have mammary glands
and a left aortic arch. Those are "similarities," but the fact that these
features are correlated is not created by noting the similarities.


>
>>> Premise 2: Dissimiliarity is sufficient evidence that life forms
>>> should be grouped together in relationship.
>>>
>>> (And, again, Steven, I don't know why you would say that dissimilarity
>>> is not one of the standards used in declaring relationship. I have
>>> been told by posters here that the pattern of similarity/dissimilarity
>>> in life forms is evidence for relationship.)
>>>
>>First of all, Zoe, obviously related things can be dissimilar. You can
>>tell
>>members of a litter of puppies apart, although they share the same
>>parents.
>>You can tell different breeds of dogs apart, even though they share common
>>ancestors. Now, the dissimilarities between, say, a Newfoundland and a
>>chihuahua are not evidence for common ancestry, but they are clearly
>>compatible with common ancestry. Moreover, they help us group the
>>different
>>breeds of dogs together to say which breeds are more closely related to
>>others, and which are more distantly related. The arguments for common
>>ancestry (and branching order) of different primates are similar in
>>principle to those for classifying different breeds in their own nested
>>hierarchy.
>
> please note that your example here refers to within the canidae
> species. How do you make the leap from canidae to, say, hominids, or
> from plants to animals?
>

Humans and other primates share the same sort of anatomical and genetic
homologies (in the same sort of consistent nested hierarchy) as canids or
felids. There are fewer shared homologies between animals and plants than
between, e.g. humans and other apes, or even humans and other primates
generally, but there are impressive ones: cells with nuclei and
mitochondria, enzymes like cytochrome-c, a shared genetic code (i.e. the
correspondence of DNA codons to amino acids). Again, the same sort of
analysis that permits grouping manuscripts into families and trees of
descent permits grouping plants and animals within the eukaryotes, nested
within a tree of other eukaryote species (most of the others fall into
kingdoms of single-celled organisms, albeit more genetically distinct from
each other in some cases than humans are from pine trees).


>
>>Now, since you seem to have your own idiosyncratic take on "creation
>>theory," I'm not sure how much weight more ... conventional ...
>>creationist
>>ideas will have with you. Most creationists hold to an idea of "kinds"
>>above
>>the species level, which include multiple species (in most cases; humans
>>are
>>always a kind to themselves) which are indeed related by common descent.
>>A
>>typical creationist would hold that, e.g. not merely different breeds of
>>_Felis catus_ (the domestic cat) but different members of _Felis_, perhaps
>>even different members of the family Felidae (e.g. lions, pumans, bobcats,
>>and housecats) are members of a single "kind." Of course, the grouping of
>>species into "kinds" requires this same sort of argument for relationship
>>from a pattern of similarities and dissimilarities; creationists fail to
>>be
>>consistent in the application of this principle (e.g. if you group lions
>>and
>>housecats together, you have no reason not to group humans and chimps
>>together).
>
> personally, I think it is an exercise in wasted effort to try to
> classify life forms to any degree beyond the most general. The vast
> variety of life forms (and more being discovered still) defy
> categorization beyond the most general categories.
>

This is clearly false, since there is widespread agreement on classification
of different life forms (if they "defied classification," you would expect
that taxonomists could not agree, e.g. that birds should be grouped with
crocodilians in the archosaurs, rather than, e.g. bats). And I think you
mean that it is a wasted effort to classify life forms beyond the most
narrow categories, not the most general. It is "general" categories (e.g.
primates, mammals, animals, etc.) that you seem to object to, not narrow
groups like species and genera.
>
-- [snip of rest; there does not seem any point to rehashing these matters]
>
-- Steven J.


Steven J.

unread,
Aug 28, 2005, 3:03:34 AM8/28/05
to

"Zoe" <muz...@aol.com> wrote in message
news:o272h1l6qciiad6dn...@4ax.com...

> On Sat, 27 Aug 2005 03:34:12 +0000 (UTC), bdbr...@mail.utexas.edu
> (Bobby D. Bryant) wrote:
>
>>On Sat, 27 Aug 2005, Zoe <muz...@aol.com> wrote:
>
> snip>
>
>>> Please support these premises:
>>>
>>> Premise 1: Similarity is sufficient evidence that life forms should
>>> be grouped together in relationship.
>>
>>You need to realize that scientists and creationists start from the
>>same point: an observation that something looks unlikely to be the
>>result of pure chance. (From there they diverge, with the creationist
>>jumping immediately to "goddidit" and the scientist looking for an
>>actual cause for the unlikely something.)
>>
>>Do you think the genetic and morphological similarities that guide
>>the construction of the tree are the result of mere chance, or the
>>result of some cause?
>
> well, from the evoluionist's point of view, the construction of the
> tree has to be the result of mere chance since the modus operandi of
> evolution is random chance mutations, selected for. Selection,
> itself, is a random activity since what gets selected are random
> events. It is by mere chance that mutations happen to produce life
> forms that appear similar.
>
If selection were a random activity, then there would be no tendency for
bacteria exposed to mutagens and antibiotics to evolve antibiotic resistance
(as opposed, e.g. to the ability to digest nylon). If selection were a
random activity ... but enough, already. Zoe, you aren't thinking, here;
you're just stringing words together.

Of course, the really weird thing is that you're sort of right, albeit
surely only by accident. A process of random mutation coupled with random
mutation will, in fact, produce a nested hierarchy. Again, the tree pattern
results because [a] descendants resemble their more recent ancestors more
closely than more distant ancestors, [b] the resemble near cousins (other
descendants of those more recent ancestors) more nearly than distant cousins
(descendants of those more distant ancestors). The treelike character of
the pattern of life is purely the result of inheritance with random errors;
natural selection actually confuses the issue, because of convergence
(analogy; different pathways taken to similar adaptions).


>
> From a creationist point of view, the construction of the tree is not
> a result of mere chance. It is the result of humans sitting down and
> classifying already-formed life forms into a tree of their choosing.
>

Zoe, whale evolution has long been the great conundrum of mammalian
phylogeny. For some decades, the favored hypothesis was that whales were
modified descendants of mesonychids (even-toed, hoofed carnivores that were
not artiodactyls). Then, DNA sequence comparisons (shared endogenous
retroviruses) showed that whales are more similar, at the DNA level, to
hippos than to other artiodactyls (even-toed hoofed mammals with a special
adaption of the rear ankle bones). This made sense only if the ancestors of
whales were themselves artiodactyls; there was no way, otherwise, that they
could be more closely related to hippos than to cows (humans, or horses, for
example, are equally closely related to hippos, pigs, cows, deer, etc.)
Then, fossils were found of primitive whales with legs, and, more important,
the distinctive artiodactyl astralagus in the ankle. So whales were rudely
uprooted from their place in the tree of humans' choosing, and inserted
where the evidence showed they belonged. Note, by the way, an example of
the "twin nested hierarchy" -- as evidence grew more complete, both fossils
and shared retroviruses (two sorts of evidence with no reason, except shared
ancestry, to vary together) supported placement of whales in the
artiodactyls.

Okay, a sperm whale is an even-toed hoofed mammal. I can see why this
evolution thing sometimes makes your head spin. But try, at least, to
understand.


>
>>Where the discussion goes from here depends on your answer to that.
>

-- Steven J.


John Harshman

unread,
Aug 28, 2005, 9:37:29 AM8/28/05
to
Zoe wrote:

> On Sat, 27 Aug 2005 00:37:43 -0500, "Steven J."
> <sjt195...@nts.link.net.INVALID> wrote:
>
>
>>"Zoe" <muz...@aol.com> wrote in message
>>news:t9gvg11b6hi7itp4b...@4ax.com...
>
>
> snip>
>
>>>Please support these premises:
>>>
>>>Premise 1: Similarity is sufficient evidence that life forms should
>>>be grouped together in relationship.
>>>
>>
>>Why should we support that premise? It is wrong, and no one is basing an
>>argument for common descent on it.
>
>
> oh really? Why has it been presented to me as THE argument then?
> "The pattern of similarity/dissimilarity that forms a nested hierarchy
> is evidence for common descent." Has this not been said before, and
> most strongly emphasized by John Harshman himself, your foremost
> expert on this matter, as you have conceded?

Very simply, "the pattern of similarity/dissimilarity" is not the same
thing as "similarity". What is there about this that confuses you?

>>"Similarity" in some respects is
>>rejected as a basis for grouping organisms together. For example, despite
>>similarities in shape and habitat, whales are not grouped apart from other
>>mammals and together with, e.g. sharks and ichthyosaurs. The box-camera
>>eyes of octopuses are not seen as a reason to group them with vertebrates,
>>who have superficially "similar" eyes.
>
> notice the subjectivity here. "Not seen as a reason to group." What
> would be a reason? So far none of this ties back to any evidence. It
> is merely the arbitrary decision of those with a particular mindset.
> In the creationist theory of intelligence, what is seen in nature is
> tied back to what is seen in human intelligence. What do you tie your
> groupings back to, but an arbitrary decision?

In fact, no single character counts as good evidence for relationships.
It's a combination of many characters that makes the evidence. When one
character (for example possession of a camera eye) contradicts the bulk
of evidence (for example presence/absence of a notochord, dorsal hollow
nerve cord, iodine-binding cells in the floor of the pharynx, or on the
other hand spiral cleavage in the embryo, a muscular foot in the head, a
radula, etc.) then we suspect homoplasy in the discordant character.
When further examination finds the discordant similarity to be
superficial, that only confirms our previous conclusion.

To repeat: it's congruence among characters that confirms homology.

>>Of course, the similarities of the eyes of cephalopods and vertebrates does
>>not extend beyond the features necessary to a "box-camera" eye (with lens,
>>retina, etc.). Famously, vertebrate eyes have an "inverted retina" and
>>blind spot, and cephalopod eyes do not, among other differences. Likewise,
>>the common features of whales, sharks, and ichthyosaurs (aside from those
>>shared by all vertebrates) are adaptions to the way they live and move.
>
> because you say so, or because it has been observed? Anybody can look
> at the function of an eye and say, see that other type of eye over
> there; this is an adaptation of that. What evidence do you have that
> it was an adaptation and not originally the way it is?

There really is such evidence. Generally, the evidence is of this form:
because the similarities are found to arise independently on a tree that
best fits all the data, we consider them to result from different
events, and to be transformations of other features that arose earlier
on that tree. In seeking to explain the reasons why similar features
arise multiple times, adaptation is an obvious hypothesis. There are
various means of testing this, which we could go into if you like.

>>It was for this reason that I brought up "homology" -- detailed similarities
>>not required by similarity in function. A bat's wing doesn't need all the
>>similarities it has to a mouse's forelimb, and the detailed correspondence
>>of the bones in both limbs is remarkable and calls for some explanation.
>
> the explanation the creationist would offer is that intelligence chose
> to create it that way.

That's not an explanation. It's vacuous, since it would be an
"explanation" for anything you saw or anything you didn't see. It tells
us nothing at all.

> From that point of acknowledgement, science
> can proceed with its research on how it operates, how it can be
> copied, how new ideas can be derived from how it was made.
> Speculation about the history of the eye contributes nothing to
> science.

What if it's not just speculation?

>>Note, also, that there are *many* shared homologies between bats and other
>>mammals. Finding some homologies between bats and mice, but not between
>>bats and birds, and different homologies between bats and birds, but not
>>with mice, and still different homologies between bats and pterosaurs (but
>>not with birds or mice) would not support the idea that these homologies are
>>explained by inheritance from a common ancestor. Presumably, branching
>>descent with modification would pass on entire suites of traits together --
>>and, therefore, entire suites of shared homologies arranged in a consistent
>>nested hierarchy argues for branching descent with modification.
>
> you have not yet answered my direct question as to why
> similarity/dissimilarity is a justifiable tool for determining
> relationship between species. Has it been observed that when things
> have a pattern of similarity and dissimilarity anywhere else than in
> life forms, that relationship is the conclusion?

Yes, as has also been explained to you several times. Human languages
and hand-copied manuscripts both fit your requirements. Further, simple
simulations of evolution also produce nested hierarchies of just the
sort we see in life.

No, no, and no again. It's not similarities that mean relationships.
It's the nested hierarchy of similarities and differences.

>>>To demonstrate premise 1, there can be no exceptions to the similarity
>>>rule. If exceptions are allowed, then exceptions should be allowed to
>>>other theories that differ from yours.
>>>
>>
>>I don't think I agree with that. Are exceptions to "creation theory"
>>explicable?
>
> there are no exceptions to the identifying characteristics of how
> mental activity behaves. Perceived exceptions are explicable.

Whatever that means.

>>Note that what you wanted was to be able to say that any entity
>>which possessed some particular, subjectively defined set of properties was
>>the product of intelligence.
>
> first of all, the properties are not subjectively defined. They are
> observed objectively. And isn't it reasonable that if those
> properties are found nowhere else outside of intelligent manipulation,
> that they form a reliable standard for identifying mental activity?

It would perhaps be so if what you had written before were not
gibberish. You abandoned those threads for just that reason.

>> It rather makes a difference if nondesigned
>>entities can possess these properties.
>
> you have taken a position that certain entities are nondesigned,
> without defending your position that they are nondesigned. You are
> starting with a settled determination that certain things are
> nondesigned. But it is those very things that you claim are
> nondesigned that are being examined.

They are considered non-designed because they form a nested hierarchy.
Nested hierarchies are not a feature of design, but they are a feature
of common descent.

>> Evolution is inferred from a very
>>large number of features of living things, and the way in which they are
>>arranged among different living things.
>
>
> the way in which they are arranged are described on the basis of
> similarities, right?

No. On the basis of patterns of similarity and dissimilarity. This is
not the same thing.

>>>Premise 2: Dissimiliarity is sufficient evidence that life forms
>>>should be grouped together in relationship.
>>>
>>>(And, again, Steven, I don't know why you would say that dissimilarity
>>>is not one of the standards used in declaring relationship. I have
>>>been told by posters here that the pattern of similarity/dissimilarity
>>>in life forms is evidence for relationship.)
>>>
>>
>>First of all, Zoe, obviously related things can be dissimilar. You can tell
>>members of a litter of puppies apart, although they share the same parents.
>>You can tell different breeds of dogs apart, even though they share common
>>ancestors. Now, the dissimilarities between, say, a Newfoundland and a
>>chihuahua are not evidence for common ancestry, but they are clearly
>>compatible with common ancestry. Moreover, they help us group the different
>>breeds of dogs together to say which breeds are more closely related to
>>others, and which are more distantly related. The arguments for common
>>ancestry (and branching order) of different primates are similar in
>>principle to those for classifying different breeds in their own nested
>>hierarchy.
>
> please note that your example here refers to within the canidae
> species. How do you make the leap from canidae to, say, hominids, or
> from plants to animals?

There is no such leap, so why ask about it? hominids are not descended
from canids, nor animals from plants. However, the same sort of
reasoning that lets us determine the relationships among breeds of dogs
also lets us determine the relationships among hominids, plants, and
animals.

>>Now, since you seem to have your own idiosyncratic take on "creation
>>theory," I'm not sure how much weight more ... conventional ... creationist
>>ideas will have with you. Most creationists hold to an idea of "kinds" above
>>the species level, which include multiple species (in most cases; humans are
>>always a kind to themselves) which are indeed related by common descent. A
>>typical creationist would hold that, e.g. not merely different breeds of
>>_Felis catus_ (the domestic cat) but different members of _Felis_, perhaps
>>even different members of the family Felidae (e.g. lions, pumans, bobcats,
>>and housecats) are members of a single "kind." Of course, the grouping of
>>species into "kinds" requires this same sort of argument for relationship
>
>>from a pattern of similarities and dissimilarities; creationists fail to be
>
>>consistent in the application of this principle (e.g. if you group lions and
>>housecats together, you have no reason not to group humans and chimps
>>together).
>
> personally, I think it is an exercise in wasted effort to try to
> classify life forms to any degree beyond the most general. The vast
> variety of life forms (and more being discovered still) defy
> categorization beyond the most general categories.

This is nonsense. It's possible for you to say this only because you are
completely ignorant of the organisms in question and the data used to
classify them. Go back to the apes again. Notice that you can easily
classify them into the hierarchy
((((human, chimp)gorilla)orangutan)gibbon). This hierarchy is
unambiguous and determinable from any genetic data you care to assemble.
How general is that?

> You may manage to pigeonhole all life forms, but of what use is this
> to the advance of science other than to say, we have them categorized
> now; see, evolutionary theory is right.

Explain why our ability to classify life into a nested hierarchy is not
evidence for evolution. It sure seems like it is to me. Please present
an alternative theory that actually explains the data.

>>Now, beyond this, recall some of the above points about analogy and
>>homology. Birds of widely different sizes, habits, and diet share the same
>>basic avian wing design. Likewise, bats of different sizes, habits and
>>diets share the same chiropteran wing design. It seems, on the principle of
>>"common design for common function," that small insect-eating birds and
>>small insect-eating bats should have more similar wings than, e.g. small
>>insect-eating birds and larger, fruit-eating birds. Likewise, we might
>>expect greater homologies between toucans and fruit bats than between fruit
>>bats and brown bats. Basic wing designs don't seem to be arranged according
>>to ecological niche, but fall into a pattern suggesting, again, descent with
>>ad hoc modification. Fruit bats and brown bats have similar wings because
>>they inherited them from a common ancestor; toucans, hummingbirds, swallows,
>>and penguins share similar wing designs because, again, a common avian
>>ancestor had wings like that.
>
>
> and certain ants with a certain mindset, exploring the world of humans
> would think that human artifacts that share similar designs must have
> evolved.

Again with the unsupported assertion about a hypothetical hierarchy.
This is not evidence to counter the real hierarchy that Steven has just
mentioned.

>>Note that we know about mechanisms of reproduction, mutation, and selection,
>>and know that these processes can produce suites of organisms that fall into
>>a consistent nested hierarchy.
>
>
> granted, you know about reproduction. But mutation and selection are
> not known in the way that reproduction is known. Reproduction is not
> speculative. Mutation and selection are speculated to have affected
> reproduction in such a way as to cause evolution. These speculations
> have not been supported. They are not in the same class as
> reproduction.

Nonsense. We can see mutation and selection just as easily as we can see
reproduction. Again, it's merely your ignorance that lets you make these
silly statements.

>>Again, we know that consistent nested
>>hierarchies typically result from a process of branching descent with
>>modification (e.g. languages, breeds of domestic animals, manuscript
>>copies), and that artifacts typically do not (just to complicate matters,
>>lateral exchange of traits is possible for all these things -- e.g. gene
>>transfer in bacteria; it is rare enough in all these cases that it does not
>>mess up the nested hierarchy, at least for eukaryotes).
>
> you are promoting nested hierarchies without defending the mechanism
> used to develop these hierarchies. I want you to defend why you think
> that the similarities/dissimilarities that are teased out to form a
> nested hierarchy are sufficient to conclude relationship between
> groups.

Are we agreed that this hierarchy is objective, i.e. that it really does
exist and is not arbitrary? If so, then it's been explained to you why
such a hierarchy is a natural (and unavoidable) consequence of common
descent. Further, nobody so far has been able to come up with any other
process that would produce such a hierarchy. It follows from this that
when you see a nested hierarchy, you can be assured that common descent
is behind it.

>>>To demonstrate this, there can be no exceptions to the dissimilarity
>>>rule. If exceptions are allowed, then exceptions should be allowed to
>>>other theories that differ from yours.
>>>
>>
>>Since you don't understand the "dissimilarity rule," this condition is
>>nonsense.
>
> what I understand of your dissimilarity rule is that if the same
> structure is used for a different function, then this, too, means
> relationship through common descent. Did I understand that
> incorrectly?

Yes. The point was that if the same structure is used for different
functions, we can't advance a functional explanation for that
similarity. Thus one alternative to the explanation of common descent
has been falsified.

>>>Premise 3: A tree-like pattern is sufficient evidence that life forms
>>>are related through a single common ancestor.
>>>
>>
>>That, again, is not quite my point, although it is at least within an
>>astronomical unit of my point.
>
>
> and can you defend even this astronomical unit of your point?

Wouldn't it be better for him to defend his actual point instead? And
for you to understand what it is?

>>>(Once again, Steven, your statement that "depending on what traits one
>>>emphasizes and compares, *different* tree-like hierarchies can be
>>>obtained" says it all. As to the traits chosen for the evolutionary
>>>tree, can you defend why those traits are emphasized and are the only
>>>traits that are correct? What makes those choices the correct ones?)
>>>
>>
>>Zoe, take any traits you like. Some (e.g. the similarity in shape that
>>links sharks and dolphins, or the similarity in color that links some
>>parrots and some iguanas) will yield trees that look very, very different
>
>>from any phylogenic tree drawn by evolutionists. Of course, a tree based on
>
>>overall color and one based on overall shape will yield very different trees
>
>>from each other. But draw trees based on comparative limb anatomy, or
>
>>comparative skull anatomy, or on various sets of genes, or on various other
>>sets of genes (after all, you have a lot of sets of genes to compare), or on
>>proteins (which don't have to form the same tree as genes, since different
>>sets of codons can code for the same amino acids), and you will find that
>>they form trees that look very much alike.
>
>
> Um-hum. So can you defend the position that because trees look alike
> (are similar) that the conclusion HAS to be relationship of different
> groups through a common ancestor?

Yes, unless you can think of another explanation for that congruence of
disparate data. Go ahead.

>> As I noted in an earlier post,
>>*all* animals with one bone in the lower jaw and three bones in the middle
>>ear also have (although these features hardly seem necessary to each other)
>>mammary glands and a left (but not a right) aortic arch.
>>
>>Why do not *all* trees yield exactly the same pattern? Well, first of all,
>>why should they? The most indicative features are those which are not
>>strongly adaptive to or obvious consequences of a particular lifestyle. For
>>example, lots of arctic mammals have white fur, presumably as camoflage
>>against the snow. Whiteness is a sign of adaption to common function,
>
> whiteness is not always a sign of adaptation to common function. You
> know that.

I hesitate even to ask what you are trying to say. But note he's talking
specifically about the whiteness of arctic animals, not all whiteness
everywhere.

>>and
>>therefore would be expected even if the animals did not share common
>>ancestry (and, indeed, polar bears and arctic foxes don't share a closer
>>ancestry with each other than they do with, respectively, brown bears and
>>red foxes). But there's no obvious reason that, e.g. giraffes and mice and
>>humans should all share seven bones in the neck, or even three bones in the
>>middle ear (many vertebrates hear well without this feature).
>
> judging from how mental activity behaves, there are good reasons to
> use a similar template for different purposes. Judging from how
> random mutations behave, there are no good reasons for similar
> templates to be passed on....unless you are willing to acknowledge
> that a fully-functioning reproductive system is already in place, and
> is then affected by mutations.

Of course a fully functioning reproductive system is already in place,
and has been for billions of years. Your "similar templates" idea just
will not produce a nested hierarchy, which is the main point. But aside
from that, why would an intelligent designer limit a giraffe neck to 7
vertebrae, when it would work better with 12 or so, like a swan's neck?

>>>So...since Steven has not yet persuasively defended these premises, I
>>>am asking again: Can anyone else (or Steven again) better defend
>>>these premises?
>>>
>>
>>To someone, like yourself, determined not to understand or be persuaded, I
>>rather doubt it. But I took a shot.
>
>
> thanks for the shot, Steven. It did miss the mark -- again -- but
> thanks anyway.

We all keep trying, in the hope that some day you will understand a few
simple points.

Ernest Major

unread,
Aug 28, 2005, 11:10:56 AM8/28/05
to
In message <pcb2h15obr23oblin...@4ax.com>, Zoe
<muz...@aol.com> writes

>>> Premise 1: Similarity is sufficient evidence that life forms should
>>> be grouped together in relationship.
>>>
>>Why should we support that premise? It is wrong, and no one is basing an
>>argument for common descent on it.
>
>oh really? Why has it been presented to me as THE argument then? "The
>pattern of similarity/dissimilarity that forms a nested hierarchy is
>evidence for common descent." Has this not been said before, and most
>strongly emphasized by John Harshman himself, your foremost expert on
>this matter, as you have conceded?

Please note the absence of the terms "pattern" and "nested hierarchy"
from the premise you are asking people to support. It is not mere
similarity that is evidence for common descent; it is the specific
pattern of similarities.
--
alias Ernest Major


--
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.344 / Virus Database: 267.10.16/83 - Release Date: 26/08/2005

John Drayton

unread,
Aug 28, 2005, 8:15:56 PM8/28/05
to
Zoe wrote:
> Sat, 27 Aug 2005 12:21:52 GMT, "R. Baldwin"
> <res0...@nozirevBACKWARDS.net> wrote:
>
> snip>
>
> >The tree of life is not built on premises, Zoe. It is built on observations.
>
> observations trigger premises. One observation is that there are
> similarities between life forms. The premise triggered in some minds
> is that these similarities must mean relationship between species
> through a common ancestor.

I suspect that a lot of people posting here (esp. those
with a math or science background) will be using the
word relationship in the sense of "connection" or
"association", not necessarily in the sense of an
ancestoral relationship.

So for instance when you say:

Premise 1: Similarity is sufficient evidence that
life forms should be grouped together in relationship.

I initially read that in the sense of "association", so to
me it read "Similarity is sufficient evidence for grouping
associated life forms".

I don't think many people would argue with this, and it
puzzled me as to why you thought it was unwarranted.

Now it seems to me that when you say "relationship", you
are using the word in the sense of "ancestoral relationship".

Is that right?

I think very few people would dispute the relationships,
it's the explanation for the relationships that's argued
about.

Evolutionary theory says that these relationships are a
function of common ancestory. I'm yet to hear a good
explanation from creation theory.

> It is this premise that I want you to
> defend, not the observations. Creationists also observe similarities,
> you know.
>
> snip tree commentary, for which I will start a new thread.
>
> >I would note, by the way, that given your wording, Premises 1 and 2 are
> >mutually contradictory.
>
> don't blame me for that. Ask Harlequin or Harshman or Steven why they
> have said just that -- that it is the pattern of similarities and
> dissimilarities that are evidence for relationship.

Your formulation of premises 1 and 2 does not match what
they've actually said. Perhaps you misunderstood what
they said.

Bear in mind that patterns of similarities/differences
were noticed and catalogued well before the advent of
evolutionary theory, so you can't just put them down to
"evolutionary preconceptions".

--
John Drayton

John Harshman

unread,
Aug 28, 2005, 8:46:19 PM8/28/05
to
John Drayton wrote:

> Zoe wrote:
>
>>Sat, 27 Aug 2005 12:21:52 GMT, "R. Baldwin"
>><res0...@nozirevBACKWARDS.net> wrote:
>>
>>snip>
>>
>>>The tree of life is not built on premises, Zoe. It is built on observations.
>>
>>observations trigger premises. One observation is that there are
>>similarities between life forms. The premise triggered in some minds
>>is that these similarities must mean relationship between species
>>through a common ancestor.
>
>
> I suspect that a lot of people posting here (esp. those
> with a math or science background) will be using the
> word relationship in the sense of "connection" or
> "association", not necessarily in the sense of an
> ancestoral relationship.

Perhaps, but not those with a biology background. We are using it just
like Zoe is, for a change. Relationship means common ancestry.

> So for instance when you say:
>
> Premise 1: Similarity is sufficient evidence that
> life forms should be grouped together in relationship.
>
> I initially read that in the sense of "association", so to
> me it read "Similarity is sufficient evidence for grouping
> associated life forms".
>
> I don't think many people would argue with this, and it
> puzzled me as to why you thought it was unwarranted.
>
> Now it seems to me that when you say "relationship", you
> are using the word in the sense of "ancestoral relationship".
>
> Is that right?
>
> I think very few people would dispute the relationships,
> it's the explanation for the relationships that's argued
> about.

I would never put it that way, and I think you are going to confuse her
more than she is already, if that's possible.

B Richardson

unread,
Aug 28, 2005, 9:20:38 PM8/28/05
to
Ernest Major wrote:
>In message <pcb2h15obr23oblin...@4ax.com>, Zoe
><muz...@aol.com> writes
>>>> Premise 1: Similarity is sufficient evidence that life forms should
>>>> be grouped together in relationship.
>>>>
>>>Why should we support that premise? It is wrong, and no one is basing an
>>>argument for common descent on it.
>>
>>oh really? Why has it been presented to me as THE argument then? "The
>>pattern of similarity/dissimilarity that forms a nested hierarchy is
>>evidence for common descent." Has this not been said before, and most
>>strongly emphasized by John Harshman himself, your foremost expert on
>>this matter, as you have conceded?
>
>Please note the absence of the terms "pattern" and "nested hierarchy"
>from the premise you are asking people to support. It is not mere
>similarity that is evidence for common descent; it is the specific
>pattern of similarities.
>
>

Twenty quatloos says you can find at least seven instances
where she has been corrected on this in the last four threads
she's posted to.

Twenty more says she restates the same gross oversimplication
as someone's premise and demand that it be defended.


R. Baldwin

unread,
Aug 28, 2005, 10:26:06 PM8/28/05
to
"Zoe" <muz...@aol.com> wrote in message
news:jgc2h1tp3shc1qkc6...@4ax.com...

> Sat, 27 Aug 2005 12:21:52 GMT, "R. Baldwin"
> <res0...@nozirevBACKWARDS.net> wrote:
>
> snip>
>
>>The tree of life is not built on premises, Zoe. It is built on
>>observations.
>
> observations trigger premises. One observation is that there are
> similarities between life forms. The premise triggered in some minds
> is that these similarities must mean relationship between species
> through a common ancestor. It is this premise that I want you to
> defend, not the observations. Creationists also observe similarities,
> you know.

The fact of similarities between life forms were not the only thing leading
to the conclusion that all life shares common ancestors. The nature of the
similarities was part of what suggested this idea. The process of deliberate
selection through animal and plant husbandry also suggested the idea.

>
> snip tree commentary, for which I will start a new thread.
>
>>I would note, by the way, that given your wording, Premises 1 and 2 are
>>mutually contradictory.
>
> don't blame me for that. Ask Harlequin or Harshman or Steven why they
> have said just that -- that it is the pattern of similarities and
> dissimilarities that are evidence for relationship.

I don't recall them saying what you did. I suspect you have misunderstood
them. It would be correct to say that more similarities are evidence for a
more distant relationship.

John Drayton

unread,
Aug 28, 2005, 10:51:51 PM8/28/05
to
John Harshman wrote:
> John Drayton wrote:
>
> > Zoe wrote:
> >
> >>Sat, 27 Aug 2005 12:21:52 GMT, "R. Baldwin"
> >><res0...@nozirevBACKWARDS.net> wrote:
> >>
> >>snip>
> >>
> >>>The tree of life is not built on premises, Zoe. It is built on observations.
> >>
> >>observations trigger premises. One observation is that there are
> >>similarities between life forms. The premise triggered in some minds
> >>is that these similarities must mean relationship between species
> >>through a common ancestor.
> >
> >
> > I suspect that a lot of people posting here (esp. those
> > with a math or science background) will be using the
> > word relationship in the sense of "connection" or
> > "association", not necessarily in the sense of an
> > ancestoral relationship.
>
> Perhaps, but not those with a biology background. We are using it just
> like Zoe is, for a change. Relationship means common ancestry.

Ok. Perhaps it's just me (I.T. background).


> > So for instance when you say:
> >
> > Premise 1: Similarity is sufficient evidence that
> > life forms should be grouped together in relationship.
> >
> > I initially read that in the sense of "association", so to
> > me it read "Similarity is sufficient evidence for grouping
> > associated life forms".
> >
> > I don't think many people would argue with this, and it
> > puzzled me as to why you thought it was unwarranted.
> >
> > Now it seems to me that when you say "relationship", you
> > are using the word in the sense of "ancestoral relationship".
> >
> > Is that right?
> >
> > I think very few people would dispute the relationships,
> > it's the explanation for the relationships that's argued
> > about.
>
> I would never put it that way, and I think you are going to confuse her
> more than she is already, if that's possible.

How about this:

Very few people would dispute the associations between life
forms, it's the explanation for the associations that's argued
about.

Evolutionary theory says that these associations are a


function of common ancestory. I'm yet to hear a good
explanation from creation theory.

<snip>

--
John Drayton

John Harshman

unread,
Aug 28, 2005, 11:34:32 PM8/28/05
to
John Drayton wrote:

Sure. But in fact creationists tend to argue about both the association
and the explanation. They commonly deny that there is a nested
hierarchy, and also explain it by common design. Zoe is hardly alone
among creationists in making mutually contradictory arguments, and in
fact in making just those very arguments.

R. Baldwin

unread,
Aug 29, 2005, 11:43:28 PM8/29/05
to
"R. Baldwin" <res0...@nozirevBACKWARDS.net> wrote in message
news:2nuQe.16577$Bc2.4423@trnddc06...

Pardon my typo. That should have been dissimilarities.

Bob Pease

unread,
Aug 30, 2005, 12:26:37 AM8/30/05
to

"B Richardson" <br...@nym.hush.com> wrote in message
news:200508290123....@nym.hush.com...

I need the time for serious stuff.
I Plonked Zoe some time ago, as she lives in a world of twisted semantics
and goofy fiats, intentional or not.

RJ Pease


cararia...@gmail.com

unread,
Sep 24, 2012, 5:30:10 PM9/24/12
to Z...@tampabay.rr.com
El viernes, 26 de agosto de 2005 20:01:56 UTC-6, Zoe escribió:
> Okay, it was a mistake to spend more than a sentence on the
> evolutionary tree itself. I should have known that the main thrust of
> my post would be ignored.
>
> Enough said about the tree as a metaphor. I am asking evolutionists
> to defend the premises upon which they have built their tree.
>
> Let me try again.
>
> Please support these premises:
>
> Premise 1: Similarity is sufficient evidence that life forms should
> be grouped together in relationship.
>
> (And, Steven, homology is all about similarity of structure and is an
> evolutionary word when used in biology. Inherent in that word is an
> evolutionary meaing, the very premise of which I am questioning --
> that similarity means relationship. Homology is the supposition that
> life forms correspond in basic type of structure and derive from a
> common primitive origin; i.e., the wing of a bat and the foreleg of a
> mouse are homologous (Webster's). If homology is about similarity,
> then it is a weak defense to use "similarity" (dressed up in another
> word) as your support for the premise that similarity means
> relationship.
>
> To demonstrate premise 1, there can be no exceptions to the similarity
> rule. If exceptions are allowed, then exceptions should be allowed to
> other theories that differ from yours.
>
> Premise 2: Dissimiliarity is sufficient evidence that life forms
> should be grouped together in relationship.
>
> (And, again, Steven, I don't know why you would say that dissimilarity
> is not one of the standards used in declaring relationship. I have
> been told by posters here that the pattern of similarity/dissimilarity
> in life forms is evidence for relationship.)
>
> To demonstrate this, there can be no exceptions to the dissimilarity
> rule. If exceptions are allowed, then exceptions should be allowed to
> other theories that differ from yours.
>
> Premise 3: A tree-like pattern is sufficient evidence that life forms
> are related through a single common ancestor.
>
> (Once again, Steven, your statement that "depending on what traits one
> emphasizes and compares, *different* tree-like hierarchies can be
> obtained" says it all. As to the traits chosen for the evolutionary
> tree, can you defend why those traits are emphasized and are the only
> traits that are correct? What makes those choices the correct ones?)
>
> So...since Steven has not yet persuasively defended these premises, I
> am asking again: Can anyone else (or Steven again) better defend
> these premises?




Hi, my name is Carlos Arias from Costa Rica im very interested about sharing my experiencies with biochemistry and my life i have a good reason to research into a few natural biochemicals in our botanical contries that are increidible helping people into understand our proper Central Nervous System and how to make a maintenance of the same with the nature of our origins, if there is any interested person into help, is more than welcome! Pura vida!

0 new messages