Fixation cost nonsense
phillip smith
headed.
this is what I understand the argument to be, for a gene to increase in
frequency other individuals carry the other alleles must die. An this is the
"cost of fixation". Let me count the ways this is wrong
1) This is nothing to do with haldanes dilemma.
2) Evolution works on gene frequencies not populations.
3) Even if we ignore the two above their is an obvious reason why this is
wrong. Say I have a population of 100 sheep and one of them has a mutation
which gives a greater fitness to that individual. Say 1.1 against 1.0. The
fixation cost argument suggests that now all the other sheep are now less
fit and their actual reproductive success will be 1/1.1. The population
should start to shrink as the next generation will have 99 * 1.0/1.1 + 1.1 =
90 + 1.1 = 91.1 so you population has dropped and this is the cost of
fixation. It is immediately obvious why you get a spurious result. Because
we should have calculated relative fitness.
mean w = 99*1.0 +1.1/100 = 1.001
so fitness of normal sheep is 1/1.001 =0.999
and fitness of new sheep = 1.1/1.001 =1.098
so next generation = 99*0.999 = 98.9 + 1.098 = 99.999 = 100
so problem disappears
Is this a mathematical trick?
No because the drop in fitness of those not carrying the new allele is THE
RESULT OF THE INCREASED REPRODUCTION OF THE INDIVIDUALS CARRYING THE NEW
ALLELE! because of fixed population size
An as I mentioned earlier this has nothing to do with haldanes dilemma
--
Phillip Smith
Applied Evolution
Joe Felsenstein
phillip smith <phi...@ihug.co.nz> wrote:
>This fixation cost stuff if I understand it correctly is completely wrong
>headed.
>this is what I understand the argument to be, for a gene to increase in
>frequency other individuals carry the other alleles must die. An this is the
>"cost of fixation". Let me count the ways this is wrong
>
>1) This is nothing to do with haldanes dilemma.
Not sure why you say that -- it sure seems like it to me.
>2) Evolution works on gene frequencies not populations.
But it does get hard on the populations if their gene frequencies give
them very ill-adapted alleles.
>3) Even if we ignore the two above their is an obvious reason why this is
>wrong.
... [mathematics omitted]
>Is this a mathematical trick?
>No because the drop in fitness of those not carrying the new allele is THE
>RESULT OF THE INCREASED REPRODUCTION OF THE INDIVIDUALS CARRYING THE NEW
>ALLELE! because of fixed population size
In this case I have argued, in my paper on Haldane's cost of natural selection
in American Naturalist in 1971, that there is no cost, no dilemma, no
substitutional load.
But there is the other case, which you didn't mention. The sheep are
standing around in the pasture and the environment gets colder. Most of
them are having a hard time, but the few sheep who have a phenotype that
gives them thicker wool aren't bothered. If they have it because of a
genetic variation of the right sort, it will gradually spread through the
population. Until it does there will be extra deaths of scantily-clad sheep.
There is then a cost, a load, and maybe even a dilemma. It is imposed by
the deterioration in environment, not by the gene substitution itself.
So in some cases there is a cost, a load, and maybe a dilemma. The issue
cannot entirely be dismissed by reference to the first case. I hope to
(re-)answer Walter Remine on this (in about a week, when I get some grant
reports and travel done with).
-----
Joe Felsenstein j...@genetics.washington.edu (IP No. 128.95.12.41)
Dept. of Genetics, Univ. of Washington, Box 357360, Seattle, WA 98195-7360 USA
phillip smith
--
Phillip Smith
Applied Evolution
Ph 0064-9-476-2918 fax 0064-9-476-2917
email: phi...@ihug.co.nz
phi...@applied-evolution.co.nz
http://www.applied-evolution.co.nz
> From: j...@evolution.genetics.washington.edu (Joe Felsenstein)
> Organization: University of Washington Genetics
> Newsgroups: sci.bio.evolution
> Date: 26 Oct 2000 16:50:17 -0400
> Subject: Re: Fixation cost nonsense
>
> In article <8sl5pe$sr3$1...@darwin.ediacara.org>,
> phillip smith <phi...@ihug.co.nz> wrote:
>> This fixation cost stuff if I understand it correctly is completely wrong
>> headed.
>> this is what I understand the argument to be, for a gene to increase in
>> frequency other individuals carry the other alleles must die. An this is the
>> "cost of fixation". Let me count the ways this is wrong
>>
>> 1) This is nothing to do with haldanes dilemma.
>
> Not sure why you say that -- it sure seems like it to me.
haldanes dilemma is that the mean fitness of a population is well be low the
hypothetical maximum. I.e. if I have 100 genes at 99 % of maximum
performance the n my fitness is 0.99^1000 gives w of 4.317124741065825e-5.
But it doesn't matter because all living things suffer the same problem so
If my competitors are all at 4.317124741065825e-5 fitness then I'm just as
fit. So I agree there is no dilemma. Natural selection does not deal in
perfection it deals in improvement. Biological systems don't have to perfect
they only have to be good enough to reproduce succesfully
>> 2) Evolution works on gene frequencies not populations.
>
> But it does get hard on the populations if their gene frequencies give
> them very ill-adapted alleles.
ill adapted alleles will be removed and the better adapted alleles will
increase in frequency that is all that matters
As Raup has pointed out Populations can stochastically increase and decrease
infrequency and extinction may just be bad luck. A alleles probability of
surviving into the next generation is a function of its coefficient of
selection, frequency and population size. These effects have been well
described in the literature by Crow and Kimura
>> 3) Even if we ignore the two above their is an obvious reason why this is
>> wrong.
> ... [mathematics omitted]
>> Is this a mathematical trick?
>> No because the drop in fitness of those not carrying the new allele is THE
>> RESULT OF THE INCREASED REPRODUCTION OF THE INDIVIDUALS CARRYING THE NEW
>> ALLELE! because of fixed population size
>
> In this case I have argued, in my paper on Haldane's cost of natural selection
> in American Naturalist in 1971, that there is no cost, no dilemma, no
> substitutional load.
>
> But there is the other case, which you didn't mention. The sheep are
> standing around in the pasture and the environment gets colder. Most of
> them are having a hard time, but the few sheep who have a phenotype that
> gives them thicker wool aren't bothered. If they have it because of a
> genetic variation of the right sort, it will gradually spread through the
> population. Until it does there will be extra deaths of scantily-clad sheep.
This is standard genetic load. There is no difference between this scenario
and one where all the sheep have the thick wool gene and its cold and every
so often mutants occur that don't have the thick wool gene and they die. Its
just a matter of frequency.
I could be argued in your first scenario that the sheep with the wooly gene
mutation now have fewer sheep to compete with in terms of grass, mates etc
and therefore have more off spring. This may or may not balance the losses
in the first generation but eventually the population will expand to a level
beyond which it can not be supported by the environment.
Again this is standard mathusian darwinism. If my competitors die I do
better.
laser...@my-deja.com
In a post titled "Fixation cost nonsense", Phillip Smith concludes that
fixation cost (more commonly called the cost of substitution) is
nonsense. He reaches that mistaken conclusion in a commonplace way, by
garbling the concept through his focus on relative fitness. My previous
posts warned of that classic mistake. The cost of substitution is
scarcely about fitness, fitness is secondary to the issue.
Let me put it another way. The cost concept is about REPRODUCTIVE
EXCESS (sorry for shouting), yet Smith NEVER EVEN MENTIONS reproductive
excess in his argument!!! No wonder he concludes there is no problem,
he thoroughly garbled it.
Nothing could be more fundamental. Yet it draws no comment from other
evolutionists here. Once again this demonstrates my point: Confusion
reigns -- Haldane's Dilemma was never solved, it was merely garbled and
brushed aside.
******
The moderator of this newsgroup (JAH) again pressed into my post his
skewed perspective: (1) People on this newsgroup (in combination with
Robert Williams) were maligning my book -- without even having read it!
Misrepresentation does not get bolder than that. To this point the
moderator issued not a whiff of reprimand to those involved. Instead,
when I posted my objection, the moderator chose to castigate me(!),
claiming my post was a "thinly veiled advertisement" for my book. He
further took the unusual step of publicly threatening to censure me!
His perspective is skewed. (See my previous post for details.)
(2) After my continued objection, the moderator then clarified, saying
his point is that "we are forced to pay" to read my book. Once again he
has a skewed perspective, considering the large amount of taxpayer
funding that evolutionary sciences receive, that we all pay for. It is
difficult to work up the slightest sympathy for his view, considering
that one can obtain my book through libraries FOR FREE!!! This point
was obvious from the start. Once again, his perspective is skewed.
(3) The publisher saw to it that my book has been DONATED to a large
number of libraries for your convenience. When you obtain my book
through a library, you are almost surely using a donated copy. All
together, the moderator's perspective is seriously skewed. It seems an
attempt to distract attention away from the wanton assassination of my
book that was taking place under this moderator. I hope the moderator
will be more even-handed in the future.
-- Walter ReMine
_The Biotic Message_
http://www1.minn.net/~science
Phillip Smith's post:
http://www.deja.com/getdoc.xp?AN=686566541
My previous post:
http://www.deja.com/getdoc.xp?AN=686566528
Sent via Deja.com http://www.deja.com/
Before you buy.
Joe Felsenstein
phillip smith <phi...@ihug.co.nz> wrote:
[originally in response to Walter Remine's views]
>>> 1) This is nothing to do with haldanes dilemma.
[and I replied]
>> Not sure why you say that -- it sure seems like it to me.
[Phillip Smith]
>haldanes dilemma is that the mean fitness of a population is well be low the
>hypothetical maximum. I.e. if I have 100 genes at 99 % of maximum
>performance the n my fitness is 0.99^1000 gives w of 4.317124741065825e-5.
>But it doesn't matter because all living things suffer the same problem so
>If my competitors are all at 4.317124741065825e-5 fitness then I'm just as
>fit. So I agree there is no dilemma. Natural selection does not deal in
>perfection it deals in improvement. Biological systems don't have to perfect
>they only have to be good enough to reproduce succesfully
Haldane's dilemma was a consideration of exactly these issues.
So yes, the subject Remine, and I were discussing, is the one
Haldane was discussing. If you don't think there is a cost, or
load, or associated dilemma in some case, fine, but it is the
same discussion as Haldane. Or can you show he was discussing
something else?
>>> 2) Evolution works on gene frequencies not populations.
>>
>> But it does get hard on the populations if their gene frequencies give
>> them very ill-adapted alleles.
>
>ill adapted alleles will be removed and the better adapted alleles will
>increase in frequency that is all that matters
>As Raup has pointed out Populations can stochastically increase and decrease
>infrequency and extinction may just be bad luck. A alleles probability of
>surviving into the next generation is a function of its coefficient of
>selection, frequency and population size. These effects have been well
>described in the literature by Crow and Kimura
This particular objection seems off the point. It is nice to
hear a general discussion of gene frequency changes but is there
some point this is refuting?
[me in response to Smith saying there is no load when favorable
mutants occur (a point I agreed with) and therefore that there
is no dilemma:]
>> But there is the other case, which you didn't mention. The sheep are
>> standing around in the pasture and the environment gets colder. Most of
>> them are having a hard time, but the few sheep who have a phenotype that
>> gives them thicker wool aren't bothered. If they have it because of a
>> genetic variation of the right sort, it will gradually spread through the
>> population. Until it does there will be extra deaths of scantily-clad sheep.
>
>This is standard genetic load. There is no difference between this scenario
>and one where all the sheep have the thick wool gene and its cold and every
>so often mutants occur that don't have the thick wool gene and they die. Its
>just a matter of frequency.
>I could be argued in your first scenario that the sheep with the wooly gene
>mutation now have fewer sheep to compete with in terms of grass, mates etc
>and therefore have more off spring. This may or may not balance the losses
>in the first generation but eventually the population will expand to a level
>beyond which it can not be supported by the environment.
> Again this is standard mathusian darwinism. If my competitors die I do
>better.
I gave a case where there would be a load, cost, maybe a dilemma.
I think that there is a real distinction between the case where the
environment deteriorates and the sheep have to substitute a gene to
cope, and the case where they undergo deleterious mutation which is
held to a low frequency by selection. In both cases there is a load,
but in the former case it is imposed by environmental changes,
in the latter it is imposed by mutation. This is the distinction
between mutational load and substitutional load. If we can magically
stop the environmental deteriorations, the substitutional load
goes away, the mutational load does not.
And yes, if one waits for the sheep to finish the substitution of the
gene for more wool and population size returns to its previous
value. But the substitutional load, and the dilemma, can occur
if there are repeated environmental deteriorations (at a certain
rate) -- the rate might be high enough that the sheep have too
few offspring to overcome the reduction in survival that these
environmental changes bring about.
So in short, one can't wave away the issue of substitutional
load, or cost of natural selection, by saying that there is none,
or that it's just "standard genetic load", or by a generalized
discussion of gene frequencies.
----
Joe Felsenstein j...@genetics.washington.edu
laser...@my-deja.com
> If we can magically stop the environmental deteriorations,
> the substitutional load goes away, …
The cost of substitution never goes away. Substitution ALWAYS requires
reproductive excess, no matter what the environment. Reproductive
excess is the central focus, the object of concern, of the cost of
substitution.
Here is a simple counter-example to Felsenstein. I here remove all the
complicating factors (to eliminate various forms of confusion and
evasion). Take a haploid, asexual species, in a constant environment.
(A non-deteriorating environment, as per Felsenstein's instructions!)
To make it simpler, assume the population size remains constant. To
make it even simpler, assume all the individuals are identical, except
for one with a new beneficial mutation. If it pleases you, to make it
even simpler, focus on just ONE generation, any generation of your
choosing. That strips away every complication I can think of. Now
focus on the fundamental issue: Assume that due to selection this new
trait increases during the generation you have chosen (it must
eventually rise from 'few' up to 'many' in order for the substitution to
occur). This absolutely requires reproductive excess! Felsenstein is
wrong. For the generation you chose, calculate any of the available
formulas for the cost of substitution -- they do not give zero cost.
Felsenstein is wrong.
My point is larger than Felsenstein and his error. His error has
remained in the literature (and on this newsgroup) since 1971,
unchallenged (explicitly) by any other evolutionist. Yet it is
FUNDAMENTAL! It is a shining example of my thesis -- Haldane's Dilemma
was never solved, instead it was merely garbled, confused, and
eventually brushed aside. Sufficient evidence of that is available on
this newsgroup.
-- Walter ReMine
_The Biotic Message_
http://www1.minn.net/~science
Joe Felsenstein's post:
http://www.deja.com/getdoc.xp?AN=688920207
My previous post:
http://www.deja.com/getdoc.xp?AN=688920165
Joe Felsenstein
>The cost of substitution never goes away. Substitution ALWAYS requires
>reproductive excess, no matter what the environment. Reproductive
>excess is the central focus, the object of concern, of the cost of
>substitution.
>
>Here is a simple counter-example to Felsenstein. I here remove all the
>complicating factors (to eliminate various forms of confusion and
>evasion). Take a haploid, asexual species, in a constant environment.
>(A non-deteriorating environment, as per Felsenstein's instructions!)
>To make it simpler, assume the population size remains constant. To
>make it even simpler, assume all the individuals are identical, except
>for one with a new beneficial mutation. If it pleases you, to make it
>even simpler, focus on just ONE generation, any generation of your
>choosing. That strips away every complication I can think of. Now
>focus on the fundamental issue: Assume that due to selection this new
>trait increases during the generation you have chosen (it must
>eventually rise from 'few' up to 'many' in order for the substitution to
>occur). This absolutely requires reproductive excess! Felsenstein is
>wrong. For the generation you chose, calculate any of the available
>formulas for the cost of substitution -- they do not give zero cost.
>Felsenstein is wrong.
We were over this in the previous round of this argument, two years
and more ago. Yes, a reproductive excess is needed, but the
case of the beneficial mutation is one where the very excess that
is needed is provided by the mutation! If there were no
reproductive excess before (say each individual produced one
offspring each generation), then when the new beneficial mutation
arose with (say) a 10% fitness advantage, this means that the
reproductive excess of its bearers will be ... 10% !
So the population is not put in any peril by this. Does it mean
that we can hope to see the reproductive excess and thus guage
how many beneficial mutations are occuring per unit time? No,
because environmental changes could also be occurring that reduced
the reproductive excess. Thus neither from the point of view of
any threat to the survival of the population, nor from the
point of view of a requirement for observable reproductive excess
tht corresponds to the rate of substitution, do we have any
limit placed on the rate of beneficial gene substitution by the
cost argument.
>My point is larger than Felsenstein and his error. His error has
>remained in the literature (and on this newsgroup) since 1971,
>unchallenged (explicitly) by any other evolutionist. Yet it is
>FUNDAMENTAL! It is a shining example of my thesis -- Haldane's Dilemma
>was never solved, instead it was merely garbled, confused, and
>eventually brushed aside. Sufficient evidence of that is available on
>this newsgroup.
Unfortunately for that grand statement, it's not an error. As was
established in the previous round of debate here.
laser...@my-deja.com
Joe Felsenstein writes:
> Yes, a reproductive excess is needed, but the case of the
> beneficial mutation is one where the very excess that is
> needed is provided by the mutation! If there were no
> reproductive excess before (say each individual produced
> one offspring each generation), then when the new
> beneficial mutation arose with (say) a 10% fitness advantage,
> this means that the reproductive excess of its bearers
> will be ... 10% !
Felsenstein just supplied a counter-example to his own position. He
previously claimed that in a NON-deteriorating environment the cost is
zero, and his above statement shows it isn't. Whether Felsenstein
realizes it or not, he just admitted that the cost of substitution is
not zero (contrary to his standing claim). Substitution requires
reproductive excess, and this is quantified directly in the cost of
substitution. The cost is not zero. This should end the argument right
here. Felsenstein's paper is wrong.
Go no further until you absorb that point. It is fundamental.
******
Felsenstein's main thrust is his attempt to hide the cost. He does that
by introducing several confusion factors. He realizes that reproductive
excess is required and he assumes it is introduced into the population
by the beneficial mutation. (He's making a hidden assumption that the
beneficial mutation increases the actual reproduction rate, which
oftentimes isn't the case, but so far he's not on implausible ground.)
Next comes his acknowledgment that this would produce actual, observable
change in the population -- an ACTUAL, OBSERVABLE increase in
reproduction by 10%! This observable consequence puts his scenario in
jeopardy (which is precisely the point of the cost of substitution -- it
examines the plausibility of a given evolutionary scenario). So he
attempts to hide, destroy, get-rid-of, this observable consequence:
> Does it mean that we can hope to see the reproductive
> excess and thus gauge how many beneficial mutations
> are occurring per unit time? No, because environmental
> changes could also be occurring that reduced the
> reproductive excess.
Felsenstein is trying to have his cake and eat it too. He needs ACTUAL,
OBSERVABLE reproductive excess (to account for a substitution), but he
also tries to simultaneously get rid of it! He cannot have both, it is a
contradiction.
******
Felsenstein throws in another confusion factor, one that is largely
irrelevant and unnecessary -- extinction.
> So the population is not put in any peril by this.
>....
> Thus neither from the point of view of any threat to
> the survival of the population, ….
The cost of substitution puts evolutionary SCENARIOS in peril, not
populations. A given evolutionary scenario incurs a cost of
substitution. If the population cannot plausibly pay the cost of
substitution, then the scenario itself becomes implausible. The
population need not be affected at all by the failure of a given
scenario.
******
Haldane's Dilemma, at its most basic classic importance, suggests that
beneficial substitutions cannot occur at a high enough rate to explain
evolution. Felsenstein aims to solve this by claiming that lots of
HARMFUL change is occurring! (Through the deterioration of the
environment.) That is a contradiction, obvious even to a casual reader.
It should have drawn evolutionists' attention a long time ago.
Haldane's Dilemma was never solved, it was merely garbled and brushed
aside.
-- Walter ReMine
_The Biotic Message_
http://www1.minn.net/~science
Felsenstein's post:
http://www.deja.com/getdoc.xp?AN=691359924
My previous post:
http://www.deja.com/getdoc.xp?AN=689319969
J. W. Edser
Walter ReMine Wrote:-
> > JE:-
> > Could Walter ReMine please define or produce working
> > definitions of absolute and relative finesses.
> WR:-
> That's not necessary, my definitions of fitness are not unusual.
> Moreover, (as I've said many times now) the focus on, and over-emphasis
> of, fitness, is a key reason why the cost of substitution has remained
> garbled for so long. If you're focusing on fitness as the central
> issue, then you've already started to garble the cost of substitution.
JE:-
Your view of
of the cost of substitution is an
absolute costing but Felsenstein's
view is of a relative costing. In this way,
your view of this cost is quite independent
of any selection while Felsenstein's is
dependent on it. In the relative argument,
the cost of substitution is calculated
relative to selection so that a maximum
of the entire substitution cost can be
explained by speculated selective rates.
In our view, zero amounts of this costs can
be so absorbed.
In your view, you seem to me to be suggesting that
not enough units of selection (here defined
as one fertile adult organism) could have ever
been reproduced over the said period of time
to make all the nucleotide substitutions
that are speculated to be necessary, between
a supposed common ancestor and today's modern
forms of chimp and man. Your question then,
just addresses a speculated maximal reproductive
potential endemic to any one species. Is this
a correct interpretation of your view? If
not, please explain what you mean by an
_absolute_ cost of nucleotide substitution?
>snip<
John Edser
Independent Researcher
PO Box 266
Church Pt
NSW 2105
Australia
laser...@my-deja.com
John Edser wrote:
> Your view of the cost of substitution is an
> absolute costing but Felsenstein's view is of a
> relative costing.
That is not a terminology that I use or one commonly in use, so let me
re-cast it. The cost of substitution is given in units of reproductive
excess, which is an absolute quantity, in the sense that it is not
relativized, it is not a figure "relative to", say, the individuals of
some other fitness.
Felsenstein has not used the term "relative costing" either. However,
Edser is correct in the following sense. Felsenstein (as do many
evolutionists concerning this topic) introduces confusion with his focus
on fitness, especially on relative fitness.
> In this way, your view of this cost is quite
> independent of any selection ...
I didn't say that. I said that fitness is SECONDARY, it is not the
primary issue or focus of the cost of substitution. Rather the GROWTH
RATE of the new trait is the primary concern, and fitness is merely ONE
way, a somewhat indirect way, of specifying the growth rate, there are
other ways. Let me put it stronger. The cost of substitution CAN be
calculated completely independently of fitness values, merely by
specifying the growth rate(s) of the new trait. On the other hand, if
one supplies fitness values instead, then you CAN calculate the growth
rate(s) of the new trait, and THEN calculate the cost of substitution.
It is seldom done in this explicit step-wise way, but this explains how
the fitness values are secondary. It is possible (though not always
convenient) to completely tackle issues about the cost of substitution
without ever mentioning fitness values. Again I say, the focus on, and
over-emphasis of, fitness, has caused great confusion about the cost of
substitution ů and Haldane's Dilemma.
> In the relative argument, the cost of substitution
> is calculated relative to selection so that a maximum
> of the entire substitution cost can be explained by
> speculated selective rates. In our view, zero amounts
> of this costs can be so absorbed.
I cannot make any sense of that statement. Nonetheless Edser has
managed to catch the essence of the common confusion. Notice he makes
no mention of REPRODUCTIVE EXCESS (sorry for shouting), yet that is the
central issue! Notice his focus again on "selection" and "selective
rates", in other words, fitness values, which isn't the issue.
> In your view, you seem to me to be suggesting that ů
I will not now entertain further expansions of the argument into wider
issues. That would be an evasive waste of time so long as the
fundamental concept remains garbled, as amply shown on this newsgroup.
Instead, I am demonstrating something here. I am demonstrating
evolutionists' inability to resolve a simple fundamental issue, an issue
that should have been (and could have been) resolved many decades ago.
Felsenstein claims (as he has since 1971) that beneficial mutations have
no inherent cost of substitution -- ZERO cost. My previous posts
identified broad counter-examples. Simply apply the formulas for the
cost of substitution -- it is not zero. Felsenstein is wrong.
It's simple: Apply the formulas and see if the result is zero, or not.
Try it! The inability of evolutionists here to resolve even this
simplest issue testifies that Haldane's Dilemma was never solved, it was
merely garbled and brushed aside.
-- Walter ReMine
_The Biotic Message_
http://www1.minn.net/~science
John Edser's post:
http://www.deja.com/getdoc.xp?AN=691690818
My previous post:
http://www.deja.com/getdoc.xp?AN=691359939
Joe Felsenstein
on Haldane's Dilemma, starts out by accusing me of all kinds of
bad faith and deliberate misleading of people. I'll reply soon to
the scientific points in his posting, but here I just want to react to his
language. Notice:
>Felsenstein's main thrust is his attempt to hide the cost. He does that
>by introducing several confusion factors.
...
>This observable consequence puts his scenario in
>jeopardy (which is precisely the point of the cost of substitution -- it
>examines the plausibility of a given evolutionary scenario). So he
>attempts to hide, destroy, get-rid-of, this observable consequence:
...
>Felsenstein is trying to have his cake and eat it too. He needs ACTUAL,
>OBSERVABLE reproductive excess (to account for a substitution), but he
>also tries to simultaneously get rid of it! He cannot have both, it is a
>contradiction.
...
>Felsenstein throws in another confusion factor, one that is largely
>irrelevant and unnecessary -- extinction.
Remine's message is clear: he is saying that I am attempting to deliberately
confuse and deceive the readers.
I hope we can avoid this level of nasty and immature invective in this
discussion.
I will reply to the substance of Remine's position in a separate posting.
Joe Felsenstein
Walter Remine <laser...@my-deja.com> wrote:
[Rather than quoting most of what Remine said, let me summarize
some possible positions one could take, then indicate what I think
Remine is saying, and why I disagree]
In interpreting Haldane's Dilemma (his cost of natural selection), one may
ask who it is a dilemma for. There seem to be two possible positions:
(1) ... for the organisms. If the environment is deteriorating and they
cope with this by substituting alleles that respond to this,
the response is not instantaneous but takes time to spread
through the population. My 1971 American Naturalist paper
argued that in this case the cost is real, and is paid by
requiring a reproductive excess to avoid the population going
extinct. (If the cost is the reproductive excess necessary to
avoid going extinct, then my position, there and subsequently
that the cost is zero when there is no environmental deterioration
but instead advantageous substitutions is fully justified).
Remine rejects this case as not the one he was invoking:
[Remine]
>The cost of substitution puts evolutionary SCENARIOS in peril, not
>populations. A given evolutionary scenario incurs a cost of
>substitution. If the population cannot plausibly pay the cost of
>substitution, then the scenario itself becomes implausible. The
>population need not be affected at all by the failure of a given
>scenario.
OK, then let's try position
(2) ... for the biologist, who has observed the reproductive excess
of the population and finds it isn't enough to explain the
rate of substitution. The environment is not deteriorating and
the mutants substituting are advantageous, so the population is
happy, but the biologist is disturbed because there isn't enough
reproductive excess to be consistent with the rate of advantageous
mutation.
I believe that this latter is what Remine is talking about.
Thus the issue is observability of the reproductive excess.
I had pointed out, in the exchange here two years ago, that
if there were also changes in the environment occurring that reduced
the absolute fitness (as seen in, for example, the reproductive
excess), then we would not necessarily expect to see this excess.
The favorable mutations could continue to occur, because the
environmental deterioration would affect all genotypes, and
relative fitnesses of different genotypes would still be the same.
This Remine objects to:
>Felsenstein is trying to have his cake and eat it too. He needs ACTUAL,
>OBSERVABLE reproductive excess (to account for a substitution), but he
>also tries to simultaneously get rid of it! He cannot have both, it is a
>contradiction.
Sorry, this makes no sense to me. What exactly prevents the following
two changes from happening?
I. A favorable mutation. Before, everyone was fitness 1. After,
a fraction p of the population has fitness 1+s, AND
II. An environmental change that reduces all of these fitnesses
(I am discussing absolute fitnesses here) so instead of being
1+s : 1 they are now 0.9*(1+s) : 0.9.
I can't see why these contradict each other. Both can happen
and the substitution does not stop.
"Garrett and Catherine" <kat...@enteract.com> raised another, also
relevant, possible reason why fitness increases may not accumulate,
that there is competition for scarce resources leading to a
constant population size. This too has the effect of reducing
absolute fitnesses, but leaving relative fitnesses the same so
that the advantageous mutants keep substituting.
Thus Remine is not talking about a cost paid by the population,
and he seems to be ruling out anything that would make it
difficult to see increases in reproductive excess. But I
can't see any logical justification for doing that.
laser...@my-deja.com
> .... If the cost is the reproductive excess necessary
> to avoid going extinct,
> then my position, ... that the cost
> [of beneficial substitution] is zero when
> there is no environmental deterioration
> ... is fully justified.
[WR: my emphasis]
Felsenstein here has a fully erroneous definition of the cost of
substitution. He defines it as the reproductive excess necessary TO
AVOID GOING EXTINCT. But that isn't, and never was the definition.
Offhand I can't think of any other evolutionist who has published such a
notion.
It is also wrong conceptually. The cost of avoiding extinction, better
called the cost of continuity, is precisely 1. That is, each individual
must, on average, replace themself, or the population will diminish
irreversibly.
The other costs of evolution are IN ADDITION to that. Thus, they are
defined as a reproductive EXCESS, a reproduction in addition to the
obligatory cost of continuity.
Felsenstein arrives at ZERO cost of substitution merely by mis-defining
it. Evolutionary substitutions require new mutation to go from few to
many, but that cannot happen without reproductive excess. Absolutely.
Positively. No exceptions! When Felsenstein says the cost is zero, he
is trying to make an exception. He has yet to address this simple
argument.
Lastly, Felsenstein is wrong by simple math. Apply any of the
traditional formulas for the cost of substitution -- it is not zero. So
even without the concepts, you can see that Felsenstein in wrong.
======================
> What exactly prevents the following two changes from happening?
>
> I. A favorable mutation.
> Before, everyone was fitness 1.
> After, a fraction p of the population has fitness 1+s, AND
> II. An environmental change that reduces all of these
> fitnesses (I am discussing absolute fitnesses here) so
> instead of being 1+s : 1 they are now 0.9*(1+s) : 0.9.
> I can't see why these contradict each other.
> Both can happen and the substitution does not stop.
I accept his challenge. I will turn his example into a counter-example.
Notice that he never mentions REPRODUCTIVE EXCESS (sorry for shouting!),
yet that is the central issue in the cost of substitution. No wonder he
doesn't see the problem. Now let me supply it.
First, put the unfit individuals out of your mind, otherwise they'll
just confuse you. Focus on the fittest individuals. Got that in your
mind? Now suppose the fittest individuals only have a reproduction rate
of 1, one progeny per individual. In other words, there is no
reproductive excess, thus the fittest individuals cannot increase, no
matter how fit you say they are. They are stuck. They cannot increase,
for lack of reproductive excess.
> Walter ReMine, .... starts out by accusing me of all kinds
> of bad faith and deliberate misleading of people.
> ....
> ReMine's message is clear: he is saying that I am attempting
> to deliberately confuse and deceive the readers.
I never said any such thing. I never said his confusions were
deliberate. Rather, I unmask his arguments as though one were unmasking
a magic trick. I specifically point out the various moves -- and the
creation of confusion is often a key move (in magic tricks as well as
bad science arguments). I will not stop doing that.
> Thus ReMine is not talking about a cost paid by the population
That's incorrect. Obviously so. Felsenstein is wasting our time.
-- Walter ReMine
_The Biotic Message_
http://www1.minn.net/~science
Felsenstein's main post:
http://www.deja.com/getdoc.xp?AN=693986776
Felsenstein's trivial post:
http://www.deja.com/getdoc.xp?AN=693986771
My previous post:
http://www.deja.com/getdoc.xp?AN=691689058
William Morse
>Haldane's Dilemma --
>
>Joe Felsenstein writes:
>
>> Yes, a reproductive excess is needed, but the case of the
>> beneficial mutation is one where the very excess that is
>> needed is provided by the mutation! If there were no
>> reproductive excess before (say each individual produced
>> one offspring each generation), then when the new
>> beneficial mutation arose with (say) a 10% fitness advantage,
>> this means that the reproductive excess of its bearers
>> will be ... 10% !
>Felsenstein just supplied a counter-example to his own position. He
>previously claimed that in a NON-deteriorating environment the cost is
>zero, and his above statement shows it isn't. Whether Felsenstein
>realizes it or not, he just admitted that the cost of substitution is
>not zero (contrary to his standing claim). Substitution requires
>reproductive excess, and this is quantified directly in the cost of
>substitution. The cost is not zero. This should end the argument right
>here. Felsenstein's paper is wrong.
>Go no further until you absorb that point. It is fundamental.
Well, I had no problem absorbing Joe's point, but you apparently are unable to
appreciate "fundamental" mathematics. In the above case, there is reproductive
excess but no cost - the hypothetical beneficial mutation allows a 10%
increase in population size. The example given does not contradict Joe's
position.
>Felsenstein's main thrust is his attempt to hide the cost. He does that
>by introducing several confusion factors. He realizes that reproductive
>excess is required and he assumes it is introduced into the population
>by the beneficial mutation. (He's making a hidden assumption that the
>beneficial mutation increases the actual reproduction rate, which
>oftentimes isn't the case, but so far he's not on implausible ground.)
>Next comes his acknowledgment that this would produce actual, observable
>change in the population -- an ACTUAL, OBSERVABLE increase in
>reproduction by 10%! This observable consequence puts his scenario in
>jeopardy (which is precisely the point of the cost of substitution -- it
>examines the plausibility of a given evolutionary scenario). So he
>attempts to hide, destroy, get-rid-of, this observable consequence:
May I point out that the scenario referred to is not his, it is yours. Under
the terms of _your_ scenario, there would be an increase in reproduction. Joe
merely goes on to point out that in the real world, an increase in fitness
will in general not translate directly to an increase in reproductive success.
(snip)
>The cost of substitution puts evolutionary SCENARIOS in peril, not
>populations. A given evolutionary scenario incurs a cost of
>substitution. If the population cannot plausibly pay the cost of
>substitution, then the scenario itself becomes implausible. The
>population need not be affected at all by the failure of a given
>scenario.
>Haldane's Dilemma, at its most basic classic importance, suggests that
>beneficial substitutions cannot occur at a high enough rate to explain
>evolution.
As long as we are going no further until we absorb some fundamental points, I
am going to point out that for the vast majority of species, there is so much
excess reproduction that the idea of a "cost of substitution" is completely
irrelevant. Excess reproduction is central to Darwin's original argument, and
Haldane's Dilemma has nothing whatsoever to do with the question of whether
evolution occurs, at least as I understand the dilemma.
Yours,
Bill Morse
laser...@my-deja.com
Bill Morse writes:
> .... In the above case, there is reproductive excess ....
> The example given does not contradict Joe's position.
Joe Felsenstein gave an example of a substitution in a non-deteriorating
environment, an environment where he previously claimed the cost is ZERO
-- in other words, no reproductive excess is required. And
lo-and-behold, the example -- his own example -- uses reproductive
excess to accomplish the substitution. That means it contradicts Joe's
claim.
> but no cost
That is mistaken. In two ways. First, Morse already admitted the
scenario uses reproductive excess, therefore there is a positive,
NON-ZERO cost. Morse is confused about the concepts. Second, simply
apply any of the published formulas on the cost of substitution -- the
cost is not zero. Morse is wrong.
> ... for the vast majority of species, there is so much
> excess reproduction that the idea of a "cost of substitution"
> is completely irrelevant.
The cost of substitution is about as irrelevant as gravity -- it's
always there, but whether it's a problem depends on the situation. The
problem caused by the cost of substitution -- Haldane's Dilemma -- was
always about species with low reproduction and long generation times --
the higher vertebrates. Nothing Morse said changes that.
> .... Haldane's Dilemma has nothing whatsoever to do
> with the question of whether evolution occurs,
That's a raw statement of faith, about as meaningful here as facing east
and saluting.
-- Walter ReMine
_The Biotic Message_
http://www1.minn.net/~science
Bill Morse's post:
http://www.deja.com/getdoc.xp?AN=696790803
Joe Felsenstein
Walter Remine <laser...@my-deja.com> wrote:
>Joe Felsenstein gave an example of a substitution in a non-deteriorating
>environment, an environment where he previously claimed the cost is ZERO
>-- in other words, no reproductive excess is required. And
>lo-and-behold, the example -- his own example -- uses reproductive
>excess to accomplish the substitution. That means it contradicts Joe's
>claim.
There are several points at issue. Let me list them:
1. Whether cost of substitution is threatens the species with
extinction. This is the issue I addressed in my 1971 paper
-- if the mutations are responses to environmental deteriorations,
the environmental deterioration is a real threat and there needs to be
enough reproductive excess to allow the species to persist while the
evolutionary response is reversing the effect of the deterioriation.
The Haldane calculation is directly relevant here, and my own
calculation arrived at a quantity that Haldane's formula approximates.
This, however, appears not to be the case Remine is discussing.
2. Whether there is a limit placed on how many beneficial mutations
can occur by the observed reproductive excess. A beneficial mutation
can create reproductive excess -- enough so that the population is
certainly not threatened by its occurrence. However if you have
a certain rate of beneficial substitutions, these do _not_ mean that
you will _see_ the resulting reproductive excess. As I have mentioned
several times here, there can also be deteriorations of the environment
occurring during the same period of time, which can offset the rise
in reproductive excess, but not stop the substitution..
This is the case Remine wants to use, to argue that humans could not
have evolved by assimilating beneficial mutations because there is not
enough observed reproductive excess. He has argued, in effect, that
you will see all the reproductive excess.
3. A technical point -- is it true, as Remine says, that one must
*always* have reprodutive excess to have beneficial mutations
substitute? The answer is No. Let me simplify my example even more
in hopes that it will be understood:
Genotype A A'
Fitness 0.9 1.0
Thus we start with a species (all A) that is declining in numbers
(for example, the environment might have deteriorated recently).
The beneficial mutation (A') restores fitness to the point where those
individuals are simply replacing themselves (there is no
observed reproductive excess). In this simple deterministic
model, A' will substitute. Remine has not addressed this case.
4. Whether the cost of substitution argument places any limit on
the rate of neutral substitutions. Answer: it does not.
--
laser...@my-deja.com
> There are several points at issue. Let me list them:
There is one central issue here that comes before all others. The
others cannot be solved until this one is: Can a trait go from 'few' to
'many' with ZERO reproductive excess -- that is, with ZERO cost, as
Felsenstein claims? In particular, can it happen in a non-deteriorating
environment? His (1971) paper mistakenly says it can.
In Felsenstein's latest post he diverts the discussion in four different
directions. NONE of them addresses the issue. Here is the closest one:
> 3. A technical point -- is it true, as ReMine says,
> that one must *always* have reproductive excess to
> have beneficial mutations substitute?
> The answer is No. Let me simplify my example
> even more in hopes that it will be understood:
>
> Genotype A A'
>
> Fitness 0.9 1.0
> Thus we start with a species (all A) that is declining
> in numbers (for example, the environment might have
> deteriorated recently). The beneficial mutation (A')
> restores fitness to the point where those individuals are
> simply replacing themselves (there is no observed
> reproductive excess). In this simple deterministic
> model, A' will substitute. ReMine has not addressed this case.
In Felsenstein's example, the new beneficial mutation starts out at only
a few copies (as a new mutation) which DO NOT INCREASE. They are
"simply replacing themselves". They never go from few to many. Rather,
the rest of the population slowly vanishes. This example precisely
supports what I have been saying -- without reproductive excess the
trait cannot go from few to many. I consistently described the problem
in that way. Felsenstein misrepresented what I said.
Felsenstein is quite aware that the above situation is unsustainable as
an explanation of evolution. He is quite aware that his misuse of the
word "substitute" here makes rubbish out of evolutionary theory. He is
also quite aware that Haldane's equations for the cost of substitution
are defined for a constant population size, and Felsenstein's example is
not remotely a constant population size.
My point about reproductive excess is exceedingly simple. Yet
Felsenstein repeatedly evades and diverts it. More remarkably, other
evolutionists here ALLOW IT -- they allow chronic error to thrive, even
when the error is simple and repeatedly pointed out, as I have done
here. ONCE you see that herding behavior at length, then you understand
why Haldane's Dilemma was never solved. It was merely garbled and
brushed aside.
-- Walter ReMine
_The Biotic Message_
http://www1.minn.net/~science
Felsenstein's post:
http://www.deja.com/getdoc.xp?AN=699275861
My previous post:
http://www.deja.com/ getdoc.xp?AN=697205616
Don Cates
Umm, ISTM that reproductive excess refers to populations, not
individuals. So if the new mutant(s) has(ve) two offspring and the
coresponding number of nonmutants have no offspring (make that three and
one for reproducing pairs), then the population will remain constant but
the new mutation will spread.
eg. Say the mutant gsthers food more quickly and reprodutive rate
depends on nutrition. Where the mutant competes with a nonmutant for
food, it tends to have one more descendent than replacement and the
nonmutant has one less. Where mutants compete with each other, they just
replace themselves. Where nonmutants compete, they just relace
themselves. The population will remain constant but the mutation will
take over the population.
Don Cates
ca...@cc.umanitoba.ca
John Edser
not to have been posted ]
Joe Felsenstein wrote:-
> > Walter Remine wrote:
> >Joe Felsenstein gave an example of a substitution in a non-deteriorating
> >environment, an environment where he previously claimed the cost is ZERO
> >-- in other words, no reproductive excess is required. And
> >lo-and-behold, the example -- his own example -- uses reproductive
> >excess to accomplish the substitution. That means it contradicts Joe's
> >claim.
> JF:
> There are several points at issue. Let me list them:
> 1. Whether cost of substitution is threatens the species with
> extinction. This is the issue I addressed in my 1971 paper
> -- if the mutations are responses to environmental deteriorations,
> the environmental deterioration is a real threat and there needs to be
> enough reproductive excess to allow the species to persist while the
> evolutionary response is reversing the effect of the deterioriation.
> The Haldane calculation is directly relevant here, and my own
> calculation arrived at a quantity that Haldane's formula approximates.
> This, however, appears not to be the case Remine is discussing.
JE:-
Here JF assumes that gene substitution
can happen within a stable organism
population. Here a _trait_ may go from
"few to many" while the _organism_ population
remains stable. Such an event is only
possible assuming reproductive excess
is strictly a derived variable and not
an absolute assumption, i.e. a constant
or a necessarily maximized quantity.
JF fails to show what absolute assumption
he has assumed just to be able to derive
his proposed relative assumption of reproductive
excess. Only the explicit defining of this
one absolute assumption allows any test of
JF's relative view of reproductive excess.
To-date, JF has failed to provide this
explicit definition. This is a bit like Einstein
suggesting mass and time are relative
but refusing to suggest this is only true
re: to the speed of light c being
assumed as an absolute maximum within nature.
Without the missing absolute value c, no
testable science can exist and Einstein's
view cannot be tested. Similarly JF must
nominate his assumed absolute to test
his view. This he will never do so that
his view cannot ever be tested.
> JF:-
> 2. Whether there is a limit placed on how many beneficial mutations
> can occur by the observed reproductive excess.
JE:-
WR suggests that a constant "the cost of continuity"
exists such that no reproductive excess can be paid
for via this constant. This constant is equal
to the population size so a population of _organisms_
must expand to allow any reproductive excess
and thus allow a _trait_ to go from "few to many".
Here is the testable difference between the two
views AFAICS. The cost of substitution is much larger
within WR view because the cost of continuity is
must be removed from any calculation of
reproductive excess. Substitution cannot occur within a
stationary organism population within WR's view
but can occur within JF's view, AFAICS.
ReMine has at least provided one absolute
value his "cost of continuity" but to-date
Felsenstein has only provided "absolute
fitness" defined I believe as the total
population reproductive rate. Such an
absolute value is not "fitness" unless
JF is talking group selection in nature.
How the term, "reproductive excess" is
rationally derived on both sides
just remains the real mystery
here.
I have asked both sides to simply define
the key term "reproductive excess" and
our esteemed moderator has agreed, but
both sides are being deliberately evasive.
This does neither side any credit at all.
> JF:-
> A beneficial mutation
> can create reproductive excess -- enough so that the population is
> certainly not threatened by its occurrence. However if you have
> a certain rate of beneficial substitutions, these do _not_ mean that
> you will _see_ the resulting reproductive excess. As I have mentioned
> several times here, there can also be deteriorations of the environment
> occurring during the same period of time, which can offset the rise
> in reproductive excess, but not stop the substitution..
> This is the case Remine wants to use, to argue that humans could not
> have evolved by assimilating beneficial mutations because there is not
> enough observed reproductive excess. He has argued, in effect, that
> you will see all the reproductive excess.
> 3. A technical point -- is it true, as Remine says, that one must
> *always* have reprodutive excess to have beneficial mutations
> substitute? The answer is No. Let me simplify my example even more
> in hopes that it will be understood:
>
> Genotype A A'
>
> Fitness 0.9 1.0
>
> Thus we start with a species (all A) that is declining in numbers
> (for example, the environment might have deteriorated recently).
> The beneficial mutation (A') restores fitness to the point where those
> individuals are simply replacing themselves (there is no
> observed reproductive excess). In this simple deterministic
> model, A' will substitute. Remine has not addressed this case.
JE:-
By suggesting above that there is
"no observed reproductive excess" Felsenstein
may have redefined reproductive excess.
I believe, he previously implicitly defined
it as the organism reproduction necessary
to maintain the existing population. In
a previous posting he agreed with
ReMine that, using his _implied_ definition,
reproductive excess is always necessary
to avoid extinction. This was a critical
point of agreement. In the above example
for Felsenstein, reproductive excess is 10%.
For ReMine the cost of continuity is 10%
and zero reproductive excess exists.
ReMine may argue that Felsenstein is negating
that reproductive excess must exist. Its just
a babble over terms and definitions AFAICS.
ReMine did address the issue above and
has termed it the cost of continuity as
I mentioned. He regards it as a constant
equal to the size of the population that
must be subtracted from reproductive
excess. Substitution only occurs within
reproductive excess not the cost of
continuity, within ReMines view, AFAICS.
> JF:-
> 4. Whether the cost of substitution argument places any limit on
> the rate of neutral substitutions. Answer: it does not.
JE:-
If and only if you allow
substitutions within the
cost of continuity.
I will leave it to ReMine
to explain why he may think
his "cost of continuity" does
not validly allow any substitution
at all, within it.
William Morse
>Joe Felsenstein gave an example of a substitution in a non-deteriorating
>environment, an environment where he previously claimed the cost is ZERO
>-- in other words, no reproductive excess is required.
No, Walter, get it right. "Reproductive excess" is _not_ equal to "cost of
substitution". Notably, in the case of a growing population, reproductive
excess may have no cost and yet can increase the percentage of substitution.
This was the example we were discussing. We can also discuss the case of a
stable environment and a stable population if you want.
>That is mistaken. In two ways. First, Morse already admitted the
>scenario uses reproductive excess, therefore there is a positive,
>NON-ZERO cost.
No there isn't, as I just explained.
Morse is confused about the concepts. Second, simply
>apply any of the published formulas on the cost of substitution -- the
>cost is not zero. Morse is wrong.
First, no I am not. Second, I will be happy to apply published formulas to a
case where they apply. As Joe Felsenstein has argued recently on this thread,
the cost of substitution is a perfectly legitimate concept, and in many cases
is not zero. But if you want to argue on that basis, you had better be using
examples where it applies, rather than your example where it did not.
>> ... for the vast majority of species, there is so much
>> excess reproduction that the idea of a "cost of substitution"
>> is completely irrelevant.
>The cost of substitution is about as irrelevant as gravity -- it's
>always there, but whether it's a problem depends on the situation. The
>problem caused by the cost of substitution -- Haldane's Dilemma -- was
>always about species with low reproduction and long generation times --
>the higher vertebrates. Nothing Morse said changes that.
I agree entirely about the subject matter of Haldane's Dilemma. You were the
one who claimed:
"Haldane's Dilemma, at its most basic classic importance, suggests that
beneficial substitutions cannot occur at a high enough rate to explain
evolution."
Suddenly we are no longer talking about explaining evolution, but about
explaining observed rates of evolution in a few species of large vertebrates.
>> .... Haldane's Dilemma has nothing whatsoever to do
>> with the question of whether evolution occurs,
>That's a raw statement of faith, about as meaningful here as facing east
>and saluting.
No, that is a simple statement of fact, that you agree with as you state
above. Walter, I will be happy to discuss the limitations imposed on rates of
evolution of large vertebrates because of the necessity for large amounts of
parental investment. Let us not pretend that this is a central question for
the theory of evolution.
Yours,
Bill Morse
John Edser
Walter ReMine Wrote:-
> WR
> There is one central issue here that comes before all others. The
> others cannot be solved until this one is: Can a trait go from 'few' to
> 'many' with ZERO reproductive excess -- that is, with ZERO cost, as
> Felsenstein claims?
JE:-
The answer seems very simple to me, depending
ENTIRELY on how you DEFINE reproductive excess.
Neither Prof. Felsenstein OR Walter ReMine have
_explicitly_ defined the term "reproductive excess"
within this sbe discussion despite repeated
calls for them to do so by myself AND the moderator.
______________________________________________
Possible Points of testability:-
Does ReMine and Prof. Felsenstein require
BOTH the number of organisms
and the number of traits to go from "few to
many" or just ONE of them? If just one of them,
is just the organism unit required or is only the
trait unit so required? If both of them are required
please explain exactly why.
______________________________________________
Once again I bring everybody's attention to my
numerous postings of a model view of ReMines and
Felsenstein's positions, as I understand them
to be, using the woefully _insufficient_ information
provided by BOTH sides. This model was recently posted
to sbe by myself, and invite reasoned criticism of it
to help myself and perhaps many others here understand
and resolve this extremely important dispute, within
the sbe discussion format.
I request, yet again, that ReMine define reproductive
excess WITHIN sbe and also supply the ref: to where
he has _previously_ defined both "reproductive excess"
and "the cost of continuity" within his book (ReMines
book is not available at my library). Without such
references posted to sbe by ReMine _himself_
it is impossible for anybody within sbe to TEST
for THEMSELVES :-
1) What ReMine means by these terms within this sbe
discussion.
2) That ReMine's argument here is the same argument
published previously within his book/papers/etc or just
a revision of it, i.e. is NOT the same argument he is
making now, within this sbe discussion. This is what ReMine
appears to accusing Prof. Felsenstein of providing,
within what he terms Felsenstein's totally evasive sbe
discussion's on Haldane's Dilemma. It is very important
that ReMine prove his hands are clean of the
same impropriety he is accusing Felsenstein.
Of course exactly the same conditions also apply to
to Prof. Felsenstein or anybody else here who is
at all _serious_ about what they have to say.
However, Prof. Felsenstein is a paid public professional
within the evolutionary theory establishment providing
important research papers within Neo Darwinism. ReMine
is, to my knowledge, one private individual outside
of this establishment. It is the taxpayer's within the USA
that pay Felsenstein within their public education system.
They have a right to expect a rational discussion here,
from him, in return. I find it quite inexcusable that
a paid public professional continues to avoid, within this
sbe discussion, providing a simple explicit definition of
"reproductive excess", even when the moderator himself has
called for such a definition (from both sides) when this
thread was originally re-opened. This leads any genuine skeptic
to conclude that Prof. Felsenstein may be deliberately evading
some important issue within this subject matter.
Felsenstein's alleged evasion here is far _more_
serious an issue than ReMines alleged evasion, since
Prof. Felsenstein represents the established view within
this field, and thus by default the massive public
investment that has gone into it and continues to pour
into it.
> > JF:-
> > 3. A technical point -- is it true, as ReMine says,
> > that one must *always* have reproductive excess to
> > have beneficial mutations substitute?
> > The answer is No. Let me simplify my example
> > even more in hopes that it will be understood:
> >
> > Genotype A A'
> >
> > Fitness 0.9 1.0
>
> > Thus we start with a species (all A) that is declining
> > in numbers (for example, the environment might have
> > deteriorated recently). The beneficial mutation (A')
> > restores fitness to the point where those individuals are
> > simply replacing themselves (there is no observed
> > reproductive excess). In this simple deterministic
> > model, A' will substitute. ReMine has not addressed this case.
> WR:-
> In Felsenstein's example, the new beneficial mutation starts out at only
> a few copies (as a new mutation) which DO NOT INCREASE. They are
> "simply replacing themselves". They never go from few to many. Rather,
> the rest of the population slowly vanishes. This example precisely
> supports what I have been saying -- without reproductive excess the
> trait cannot go from few to many. I consistently described the problem
> in that way. Felsenstein misrepresented what I said.
JE:-
ReMine seems here to suggest that his proposed
constant "the cost of continuity (cC)
is NOT equal to the organism population
size (P) 100, but is only equal to the
SUB population of just 10 organisms that actually
carry the trait A, within it. Only these selected
organisms with the said trait A, within the
total population P, actually replace themselves
and constitute his constant cC, within ReMines
argument. Thus for ReMine it seems reproductive excess
(rE) is equal to the population size - the cost
of continuity (cC).
rE = P - Cc
If the original population is 100, mixed
10%A and 90%A' organisms and the beneficial
trait A is represented by just 10 organisms
within that population, then according to
myunderstanding of ReMine's argument:-
cC = 10
P = 100
rE = 100-10
rE = 90
Here the reproductive excess that
is _required_ to keep the population
at a constant organism size of
100, is 90. This is a very
_large_ requirement of rE, i.e. is
NOT a ZERO requirement of rE
as suggested by Felsenstein,
within ReMines understanding of
of this problem, AFAICS.
If only the sub population of
10 organisms with A replaces
itself and is as defined cC, while the
90 organisms with A' just keep declining,
then the population will eventually
fall to just 10 A individuals and stay
there, within this very simplified
model view of nature. Here the substitution
of A for A' has indeed been completed
(the population is now 100% A) but
this population remains at only 1/10 its
original size. The trait A has NOT moved
in any _absolute_ organism numbre sense
(or trait sense) from "few to many". However
it HAS moved in strictly a _relative_
sense from "few to many" since the selected
organism with trait A, now constitutes 100%
of the organisms within the population, where
A' is now exactly 0%. Here however
the original population size has been _reduced_
to the SUB population size of just 10
individuals. This dispute is thus mostly semantic.
Can reproductive excess be defined as an absolute
or just relative value within:-
(i) An over-simplified model view?
(ii) Within what constitutes a testable
theory of nature?
________________________________________
Missing information:-
Is Prof. Felsenstein suggesting that the
entire population (P) remains the same size
i.e. at 100% at no cost-what-so-ever, or
just the sub population of size 10%?
_________________________________________
[Don't expect any reply from Felsenstein
to this critical question since on principle,
JF will never reply to any questions JE asks(!),
so I propose somebody else ask him this same
question, within sbe.]
> WR:-
> Felsenstein is quite aware that the above situation is unsustainable as
> an explanation of evolution. He is quite aware that his misuse of the
> word "substitute" here makes rubbish out of evolutionary theory. He is
> also quite aware that Haldane's equations for the cost of substitution
> are defined for a constant population size, and Felsenstein's example is
> not remotely a constant population size.
JE:-
Certainly, even within William's website,
within his discussion of Haldane's original
propositions, populations are "suddenly" restored
to the "original size" as if by magic. Williams
points out this was a basic "simplification" that
was simply "allowed" by Haldane. As always, the
problem here remains the misuse Vs proper use of gross
simplifications of testable theories of nature
within common gene centric models of selection,
used within Neo Darwinism. Nobody within sbe will
discuss this important question despite repeated
attempts by myself to engage people here on this
topic.
> WR
> My point about reproductive excess is exceedingly simple. Yet
> Felsenstein repeatedly evades and diverts it. More remarkably, other
> evolutionists here ALLOW IT -- they allow chronic error to thrive, even
> when the error is simple and repeatedly pointed out, as I have done
> here. ONCE you see that herding behavior at length, then you understand
> why Haldane's Dilemma was never solved. It was merely garbled and
> brushed aside.
JE:-
I am an evolutionist who has posted
comments on this and related subjects
within sbe, for over 3 years, and I
am _still_ trying to make rational sense
of this one dispute, where the deliberate
withholding and/or the clarification of
quite _basic_ information on BOTH sides, is
all that seems to be prohibiting anybody's
understanding of it! Where is the
essential integrity DEMANDED of all
"scientific discussion" within evolutionary
theory discussion?
John Edser
jimm...@my-deja.com wrote:-
> > > JMcG:-
> > >You have a plainly comical way of arriving at the
> > >conclusion that Darwin's individual centric model is the one and only
> > >correct model. It goes like this:
> > >Darwin thought of his explanation first. Any challengers to
> > >Darwin's model
> > >must refute the assumptions that are attached to Darwin's model or
> > >else
> > >Darwin's model wins. You attempt to cloak this comical way of
> > >arriving at
> > >your conclusion in the terminology of being "testable". In reality
> > >other
> > >less individual centric designations of natural selection are on
> > >equal par
> > >with Darwin's individual centric model in terms of testability.
> > JE:-
> > Really, how interesting. Please name just one
> > of them
> JMcG;-
> Okay, how about the ecosystem as the unit of selection. (Or you could
> choose any other level of biological phenomena. It makes no
> differences. They are all equally testable.)
JE:-
[MgGinn just snipped my request for fitness]
They are NOT all equally testable, this is all
that over 30 years of "group selection" debate
provided! Within that debate it was resolved that
if individuals were selected then groups NEED NOT
BE and if groups were selected then individuals
CANNOT BE. Note the terms "need not be" and "cannot
be" are entirely different propositions where only
one is conclusive, i.e. "cannot be". This is
how the debate was resolved. I can however
resolve it is stronger terms, but lets just
use this historic resolution, for now.
The problem here that McGinn seems not
to wish to understand the difference between
the selection of one _additive_ composite entity
compared to the selection of one _multiplicative_
composite entity. Their selective logic is not the
same, in any way.
An ecosystem is the simple addition of its
parts but an organism is a complex multiple
of its parts in fitness. Both these assumptions
can be tested against nature and confirmed.
The one value that MUST be measured
here on either an additive or multiplicative
scale, is fitness. It is fitness alone that determines
what shall be and what shall not be, selected in
any TESTABLE way. McGinn seems not to be able,
or does not wish to understand, this basic
point. One additive composite of Darwinian fitness
cannot be assumed to selected in competition to
any non additive composite of it, simply because
an additive composite cannot ever have any testable
fitness to measure itself.
Fitness was only _implicitly_ defined
by Darwin. I offer my _explicit_ definition
of his implicit assumption of it:-
___________________________________________
Definition of fitness:-
Darwinian fitness is the total number of
fertile organisms reproduced per parent
lifespan, per population.
___________________________________________
Even if you destroy _entirely_ a whole ecosystem or
one population within it, the only reason that one
ecosystem or a population is NECESSARILY selected
against, is because each Darwinian fertile organism
_reproductive_ within it was selected against, _individually_.
When you suppose one population is selected "against"
in just _survival_ terms, exactly the same result is
obtained if you just assume that only each individual
within it is selected against in _reproductive_ terms.
The survival fitness of the one population is always just
the mere addition of individual reproductive
fitnesses within it! Thus populations "need not be"
ever be supposed, to be selected within selection theory
i.e. group selection is always _entirely_ removed by
Occam's Razor. This was the resolution of the great group
selection 30 year debate. That resolution, though
incomplete, stands today. Group selectionists
have joined forces with the multi-unit-selections under
supposition 2. This non testable supposition is under-
pinned by "post-modernism". This is why evolutionary theory
is now embracing neutral theory, to which McGinn rightly objects.
The whole scientific edifice is under siege and nobody recognizes
the primitive forces at work seeking to destroy it.
Today, creationism seeks a chair within the faculty of
science suggesting it is "just as scientific
as Neo Darwinian evolutionary theory" and within
post-modernism, it is!!
The same false reasoning does NOT apply to fertile
individual units of selection, Darwin's implicit
assumption and his only supposed and testable
unit of fitness. Fertile reproductive individuals
are not just the sum of their parts
like ecosystems are. This suggestion
was suggested by Dawkin's et al
but has failed. Dawkin's one fertile organism
was always just an approximation using the addition of
the fitness of each gene within itself, via Fisher's
1920's over-simple-assumption. Within nature, no
additive genes re: fitness per gene (measured as total
gene reproductions per defined ORGANISM cycle and
not per gene cycle) have never been observed in
nature but are always just assumed within gene centric
models of selection that have been entirely misused, to
predict nature. If they ever were observed then "the
fertile organism unit" of Darwin would be refuted.
Additive gene fitness in NATURE is an EXCLUDED event
within Darwinism. It is only such excluded events that
make the theory at ALL testable. This, amazingly, you
also seem to not want to understand. Any testable view
must have both valid predictions and valid exclusions,
no exceptions.
> JMcG:
> <snipped a bunch of words that made absolutely no sense to me at all>
JE:-
Nothing difficult to understand was
snipped. I simply discussed:
1) Dualism Vs Monism
2) Equality Vs Tautology
since they were entirely necessary
to resolve an argument in a rational
way. (I suggest ref. to Britannica.)
> > Darwin
> > is NOT talking about populations of magic fairies at
> > the bottom of the garden, evolving, he is talking
> > about specific defined entities called fertile
> > organisms, evolving in populations of same. Here
> > only one type of unit of selection is supposed where
> > always, A = A where A is one unit of Darwinian
> > selection. If Darwin exchanged his unit of selection
> > half way then A would not equal A and his view would
> > be invalid and non testable, drivel.
> JmcG:
> I'm not talking about switching.
JE:-
You must be very _explicit_ about
what you are suggesting here,
EITHER:-
1) No units of selection exist.
2) Multiple NON self exclusive
units of selection exists, i.e.
A and B and C and D etc. Here
you can validly exchange units of
selection "midstream" any time
you wish.
3) Multiple self exclusive units
of selection exists, i.e.
A or B or C or D etc. Here
you cannot validly exchange units
of selection "midstream".
Only SINGLE units of selection
are actually proposed within this
view.
Your very last word on this
matter was that your
view excluded ALL units of
selection, i.e. was 1. This is not
a testable view since it
cannot be measured. Previous to
that you suggested that non self
exclusive units were being
used. This is not a testable
view since it cannot
be refuted.
I have only ever suggested that
3. is testable science where in
effect strictly SINGLE units of
selection in nature are alone,
proposed.
> JMcG:-
>I'm saying start with the ecosystem as
> the unit of selection and stay there. Is this not equally testable as
> if we started with the individual and stayed there?
JE:-
Yes it is, if and only if, you refer to
it as one SINGLE unit of selection re: 3.
You do realize that if just one observation
that confirms that just one part of one ecosystem
is selected on an additive basis, this refutes
your view entirely? If you continue with
this line of reasoning you have to define what
is the testable fitness of one ecosystem and then
proceed to measure this fitness within nature.
If you cannot define this fitness or if
you do define it but then cannot measure it in
any practical way, your view remains non testable
against nature and can only be regarded, at
best, as scientific speculation.
If we suppose that McGinn has
indeed defined this missing unit of
fitness, and it is possible to measure it within
nature only then does his view becomes testable.
However the view that one ecosystem is one
unit of selection is simply refuted,
since an excluded event is easily observed.
This excluded event is one sub unit WITHIN
an ecosystem being shown to be selected with its
own additive fitness. This can easily be shown
within nature i.e. single fertile organism
units within one ecosystem's can be measured to
have independent additive fitness. Unless you
move your argument down to proposition 2 from 3, your
view that one ecosystem is one unit of selection
is refuted. Option 2. is always popuar. Most people use
it today invoking "multiple units of selection", since
it always remains _non_ refuted because it cannot
ever be refuted. No excluded events are at
all are derivable from type 2 propositions! This is
why proposition 2 is thrown out as testable science.
Thus you do have a choice; assume your view as a non
testable belief, or acknowledge it as a refuted view.
> > JE:-
> > the unit of
> > Darwinian natural selection is a valid
> > testable assumption of the unit of selection
> > that remains used within all testable views of
> > evolution but remains non refuted to this day.
> JMcG:-
> The same is true if we assume the ecosystem as the unit of selection.
> Right?
JE;-
Wrong, see above.
> > > JMcG:-
> > > Maybe you should stop employing circular logic and mistaking
> > > assumptions
> > > for truths.
JE;-
I suggest AGAIN, very strongly,
that you find out for yourself
the very real difference between
an equality and a tautology before
you use in a loose way, associated
with the term "self reference".
> > JE:-
> > This dispute is EASILY tested but McGinn
> > just refuses to test it. All he has to do
> > is supply any other view of evolution
> > and its definition of fitness.
> I've already chosen the ecosytem. I don't know what you are talking
> about with respect to, "its definition of fitness." Why in the world
> would this be necessary.
JE;-
Their is little point
continuing this discussion
since McGinn does not have
any idea, or just does not want
to know, how selection
theory is/can/could be _measured_
within biology. If one idea is testable
it must be able to be measured!
Fitness is the ONLY measure of
selection and this measure does
not change in principle (but does in
units of selection counted) no matter what
units of selection you may suppose.
You can call fitness another name if
you wish (but that would just confuse,
unnecessarily). As long as fitness
has defined counted and compared,
the units of selection reproduced over
one period of time, into one additive
population grouping of them, fitness can
in principle measure to refutation a
selection theory, allowing that proposition
as testable science.
Tim Tyler
: Joe Felsenstein writes:
:> There are several points at issue. Let me list them:
: There is one central issue here that comes before all others. The
: others cannot be solved until this one is: Can a trait go from 'few' to
: 'many' with ZERO reproductive excess -- that is, with ZERO cost, as
: Felsenstein claims? In particular, can it happen in a non-deteriorating
: environment? His (1971) paper mistakenly says it can.
I'm not really following this thread - but what is the problem with this?
A trait can go from "few" to "many" without any individuals dying
(even in a non-deteriorating environment) - if (for example) the
population consists primarily of the "few" to start with.
[moderator's snippage to comport to charter requirements - JAH]
--
__________ Lotus Artificial Life http://alife.co.uk/ t...@cryogen.com
|im |yler The Mandala Centre http://mandala.co.uk/ Free gift.
laser...@my-deja.com
I disagree with:
> Substitution cannot occur within a
> stationary organism population within WR's view
> For ReMine the cost of continuity is 10%
> The cost of substitution is much larger
> within WR view
-- Walter ReMine
_The Biotic Message_
http://www1.minn.net/~science
Edser's post:
http://www.deja.com/getdoc.xp?AN=701641485
laser...@my-deja.com
> The answer seems very simple to me, depending
> ENTIRELY on how you DEFINE reproductive excess.
I do not detect the definition of "reproductive excess" to be a point of
dispute between Dr. Felsenstein and myself. Our disagreement centers on
the "cost of substitution" and its meaning. [Note: The doctor, at
times, tries to save his failed thesis by using a peculiar
(self-serving?) definition of this key term, a definition no one else
has employed in the literature. At other times, he attempts to shift
away from the case that most clearly shows his error and contradicts his
claims -- a non-deteriorating environment.]
I strongly suspect that people who give the matter some thought will
nearly always agree on the meaning of reproductive excess. However,
Edser does find something about the concept confusing, and repeatedly
brings it up. All right, I'll bite. Simply go to the library and look
up any evolutionary documents on the cost of substitution. Do you find
Edser's questions answered?
* If so, then employ it.
* If not, then you have just proven another facet
of my thesis, that the fundamental issues were garbled,
confused, and allowed to fall into obscurity.
My goal is not to spoon-feed you the answers, you already know precisely
where I have done that. Rather, my goal here is to show that the
answers are not coherently found within current evolutionary thought.
Haldane's Dilemma was never solved. Edser's complaints about the
confusion only demonstrate my point. The absence of evolutionists
stepping in to clear up Edser's confusion also demonstrates my point.
Confusion reigns. Do not blame me for pointing it out.
> Does ReMine ... require BOTH the number of
> organisms and the number of traits to go from
> "few to many" or just ONE of them?
The issue is: What does EVOLUTION require? I bring up the fundamental
point -- Evolution requires new beneficial mutations to go from few
copies to many copies, and this increase must be born (literally) by the
reproductive excess of individual organisms. Contrary to Edser, this is
not a semantic issue. This is a rock hard, unavoidable, quantifiable
issue in evolutionary theory. Going from 'few' to 'many' is required,
and reproductive excess is required to make it happen. No escape.
Edser has been a good sport. He is understandably frustrated by the
lack of clear answers on the cost of substitution and related issues.
That is not my fault. The cost of substitution has been around for 43
years, and (falsely) proclaimed "solved" since the mid-70s. There
should be no confusion now. The answers should be easily available by
now. The fact that they aren't, proves my thesis.
> I find it quite inexcusable that [Dr. Felsenstein]
> a paid public professional continues to avoid, within this
> sbe discussion, providing a simple explicit definition of
> "reproductive excess", ... This leads any genuine skeptic to
> conclude that Prof. Felsenstein may be deliberately evading
> some important issue within this subject matter.
I agree with Edser, that the good doctor Felsenstein is quite evasive.
However, that is not a large concern to me. It is merely a minor clue
to a bigger puzzle. I point out that the evolutionary community has
allowed confusion, error, and contradiction (including Felsenstein's) to
thrive for many decades. It is this LACK OF PURSUIT that is most
curious. Once you see the confusion, the lack of pursuit, and the
importance of the issue, then you will see why Haldane's Dilemma is
worthy of your study.
> I request ... that ReMine ... supply the ref:
> to where he has _previously_ defined both "reproductive
> excess" and "the cost of continuity" within his book
These matters are defined in my book in the chapter titled, "Haldane's
Dilemma" and its appendix. (Some supportive material is discussed in
the chapters, "The Neutral Theory of Evolution" and "Population
Genetics".) It was donated to a large number of libraries for your
convenience.
-- Walter ReMine
_The Biotic Message_
http://www1.minn.net/~science
My post:
http://www.deja.com/getdoc.xp?AN=701023335
Edser's response:
http://www.deja.com/getdoc.xp?AN=701641480
jimm...@my-deja.com
Put up or shut up. You said they are not testable. I say they are.
this is all
> that over 30 years of "group selection" debate
> provided! Within that debate it was resolved that
> if individuals were selected then groups NEED NOT
> BE and if groups were selected then individuals
> CANNOT BE.
Resolved? It was never resolved as you suggest.
Note the terms "need not be" and "cannot
> be" are entirely different propositions where only
> one is conclusive, i.e. "cannot be". This is
> how the debate was resolved.
*This* is how the debate was resolved? You are suggesting semantics
have scietific validity?
I can however
> resolve it is stronger terms, but lets just
> use this historic resolution, for now.
Your ability to come to all encompassing conclusions that have no basis
in reality is amazing.
>
> The problem here that McGinn seems not
> to wish to understand the difference between
> the selection of one _additive_ composite entity
> compared to the selection of one _multiplicative_
> composite entity. Their selective logic is not the
> same, in any way.
Selective logic?
>
> An ecosystem is the simple addition of its
> parts but an organism is a complex multiple
> of its parts in fitness.
What an original way to use mathematics.
Both these assumptions
> can be tested against nature and confirmed.
Of course.
> The one value that MUST be measured
> here on either an additive or multiplicative
> scale, is fitness. It is fitness alone that determines
> what shall be and what shall not be, selected in
> any TESTABLE way. McGinn seems not to be able,
> or does not wish to understand, this basic
> point.
Yes, and my misunderstanding of the logic of your thinking does not
begin and end here.
One additive composite of Darwinian fitness
> cannot be assumed to selected in competition to
> any non additive composite of it, simply because
> an additive composite cannot ever have any testable
> fitness to measure itself.
To measure itself?
>
> Fitness was only _implicitly_ defined
> by Darwin. I offer my _explicit_ definition
> of his implicit assumption of it:-
> ___________________________________________
> Definition of fitness:-
> Darwinian fitness is the total number of
> fertile organisms reproduced per parent
> lifespan, per population.
> ___________________________________________
Isn't this kind of like takeing a wooden box, attaching a wire to it,
and calling it a television?
>
> Even if you destroy _entirely_ a whole ecosystem or
> one population within it, the only reason that one
> ecosystem or a population is NECESSARILY selected
> against, is because each Darwinian fertile organism
> _reproductive_ within it was selected against, _individually_.
> When you suppose one population is selected "against"
> in just _survival_ terms, exactly the same result is
> obtained if you just assume that only each individual
> within it is selected against in _reproductive_ terms.
> The survival fitness of the one population is always just
> the mere addition of individual reproductive
> fitnesses within it! Thus populations "need not be"
> ever be supposed, to be selected within selection theory
> i.e. group selection is always _entirely_ removed by
> Occam's Razor.
And here I was under the impression that Occam's razor was only
applicable in the context of two competing hypotheses that attempt to
explain the *same* phenomena. Silly me.
If one cell in an idividual is selected would this not refute the
individual as the unit of selection. Seems like you have a double
standard. And this is overlooking the fact that I consider this notion
that any unit of selection could be refuted to be plainly comical.
This can easily be shown
> within nature i.e. single fertile organism
> units within one ecosystem's can be measured to
> have independent additive fitness. Unless you
> move your argument down to proposition 2 from 3, your
> view that one ecosystem is one unit of selection
> is refuted. Option 2. is always popuar. Most people use
> it today invoking "multiple units of selection", since
> it always remains _non_ refuted because it cannot
> ever be refuted. No excluded events are at
> all are derivable from type 2 propositions! This is
> why proposition 2 is thrown out as testable science.
> Thus you do have a choice; assume your view as a non
> testable belief, or acknowledge it as a refuted view.
As I've stated before, John, you are mistakenly mixing the terminology
of experimental science with the terminology of theoretical science and
all you are achieving is to decieve yourself that what you are saying
makes any sense to anybody but yourself.
John, you are greatly in need of some asistance with the proper
employment of scientific terminology, methods, traditions. Trust me.
--
Jim McGinn
Human evolution is no longer a mystery:
www.crosswinds.net/sacramento/~jimmcginn/unmissing
laser...@my-deja.com
> ... what is the problem with this?
>
> A trait can go from "few" to "many" without
> any individuals dying (even in a non-deteriorating
> environment) - if (for example) the population
> consists primarily of the "few" to start with.
A trait cannot go from 'few' to 'many' without reproductive excess.
There is no way around it. That is the issue.
-- Walter ReMine
_The Biotic Message_
http://www1.minn.net/~science
My post:
http://www.deja.com/getdoc.xp?AN=701023335
Tyler's response:
http://www.deja.com/getdoc.xp?AN=701641498
laser...@my-deja.com
Bill Morse wrote:
> ... "Reproductive excess" is _not_ equal to "cost of substitution".
The literature already says the cost of substitution is a reproductive
excess. This point is clear, and correct.
> Notably, in the case of a growing population,
> reproductive excess may have no cost ...
That is an ungrammatical statement. Like saying "Ten minus plus divided
by times 10."
Stated correctly: Growth from 'few' to 'many' (as in substitution) has
a cost.
> This was the example we were discussing.
No it wasn't.
>> Second, simply apply any of the published
>> formulas on the cost of substitution -- the
>>cost is not zero. Morse is wrong.
>
> I will be happy to apply published formulas to
> a case where they apply.
Then apply them -- the cost is not zero, contrary to Felsenstein's
claim.
> As Joe Felsenstein has argued recently on this thread,
> the cost of substitution is a perfectly legitimate concept,
> and in many cases is not zero.
This isn't, and never was, the issue. Rather, Felsenstein claims that
beneficial substitutions have no inherent cost of substitution. That is
wrong, both conceptually, and by simple application of the existing
formulas. The cost is not zero.
> Let us not pretend that this is a central question
> for the theory of evolution.
Also, let us not pretend otherwise...
-- Walter ReMine
_The Biotic Message_
http://www1.minn.net/~science
My post:
http://www.deja.com/getdoc.xp?AN=697205616
Bill Morse's response:
http://www.deja.com/getdoc.xp?AN=701641449
Joe Felsenstein
>'many' with ZERO reproductive excess -- that is, with ZERO cost, as
>Felsenstein claims? In particular, can it happen in a non-deteriorating
>environment? His (1971) paper mistakenly says it can.
My 1971 paper says no such thing. It asks whether one needs a reproductive
excess to avoid extinction. In the case of advantageous mutations, they
create their own excess and impose no burden that threatens extinction.
If Remine thinks there is a passage in my paper that says you can go from
few of a mutant to many of that mutant without there being any reproductive
excess let him cite that passage.
>In Felsenstein's latest post he diverts the discussion in four different
>directions. NONE of them addresses the issue. Here is the closest one:
>
>> 3. A technical point -- is it true, as ReMine says,
>> that one must *always* have reproductive excess to
>> have beneficial mutations substitute?
>> The answer is No. Let me simplify my example
>> even more in hopes that it will be understood:
>>
>> Genotype A A'
>>
>> Fitness 0.9 1.0
>
>> Thus we start with a species (all A) that is declining
>> in numbers (for example, the environment might have
>> deteriorated recently). The beneficial mutation (A')
>> restores fitness to the point where those individuals are
>> simply replacing themselves (there is no observed
>> reproductive excess). In this simple deterministic
>> model, A' will substitute. ReMine has not addressed this case.
>
>In Felsenstein's example, the new beneficial mutation starts out at only
>a few copies (as a new mutation) which DO NOT INCREASE. They are
>"simply replacing themselves". They never go from few to many. Rather,
>the rest of the population slowly vanishes. This example precisely
>supports what I have been saying -- without reproductive excess the
>trait cannot go from few to many. I consistently described the problem
>in that way. Felsenstein misrepresented what I said.
Along with a number of recent posters here, I could not conceive that
Remine was talking about numbers, as the sensible thing to talk about is
frequency. "Few" and "many", I thought referred to low and high
frequencies. My example was responsive to that. I see that I was
wrong about what he was talking about. He wants to talk about
numbers. Yes, he is right -- you need a reproductive excess to
increase numbers of any mutant.
He being correct on that point, I have to ask "so what"? In particular,
if we see a species (Remine is thinking of humans) which increases its
fitness by accumulating a bunch of advantageous mutations, as Don Cates
and others have noted, they could be competing more effectively for
limited resources. The accumulation of each mutation would then result,
not in an observable increase of reproductive excess, but would
decrease the fitness of its competitors (by reducing their fertility or
viability). So having a bunch of advantageous mutations substitute does
not lead us to predict such a high reproductive excess that we can
reject this having occurred in the human lineage.
There are other scenarios where the resources are not limiting but as
advantageous mutations are substituted, there are also environmental
deteriorations going on, so the reproductive excess is not getting
improbably high.
So in the case of advantageous mutants, there is no contradiction
between what we know of human reproductive excess and accumulation
of many advantageous mutants in the human lineage. That is Remine's
argument, and he is wrong about that.
(I'll pass over all the passages in Remine's posting where he accuses me
of knowingly spreading falsehoods and misleading the readership. No,
I didn't do any of that.)
----
Christopher Mosley
laser...@my-deja.com wrote:
> the chapters, "The Neutral Theory of Evolution" and "Population
> Genetics".) It was donated to a large number of libraries for your
> convenience.
Just wondering, what exactly is a large # of libraries ? Checked L.A.
Public Library system, N.Y.C. Public Library system , Chicago Public
Library System, Houston Public Library system, Philadelphia Free Library
system -- no ReMine. Well there was a Peggy ReMine, I think.
So, to to get up to speed while waiting for that inter-library loan from
Palookaville check out:
http://home.wxs.nl/~gkorthof/kortho41.htm
[moderator's note: This web page is a serious and reasonably
cordial review, with an eye to refutation, of Walter's book.
As such, I think the pointer is a fair matter to post. Interested
readers should contact the author directly if interested in any
particulars. - JAH]
laser...@my-deja.com
Joe Felsenstein writes:
> If ReMine thinks there is a passage in my paper
> that says you can go from few of a mutant to many
> of that mutant without there being any reproductive
> excess let him cite that passage.
Here is an example from his paper:
"The occurrence and substitution of favorable
mutants in the absence of environmental change
does not impose any cost, but makes the population
better able to bear the cost." (Felsenstein, 1971, p 11)
That statement is dead-center where Felsenstein is wrong. It is also the
clearest case of how he is wrong -- the case where there is no
environmental change. Simply pick an example and apply the published
formulas for the cost of substitution. They show the cost is NOT zero.
So Felsenstein is wrong.
Also, look at the concepts -- reproductive excess is required, so again
Felsenstein is wrong. This point is exceedingly simple.
Felsenstein finally acknowledges my central point:
> Yes, he [ReMine] is right -- you need a reproductive excess
> to increase numbers of any mutant.
Reproductive excess is required, therefore the cost is
positive-non-zero. Why? Because these are two phrases for the same
thing. So again, Felsenstein's thesis is wrong.
He is wrong mathematically -- as shown by application of the published
formulas. He is also wrong conceptually.
=====================================
Felsenstein spends the rest of his post trying to divert the above
issue. He does that by bringing in his favorite confusion factor --
"environmental deterioration" -- along with many more:
"fitness",
"competing more effectively for limited resources",
"would decrease the fitness of its competitors
(by reducing their fertility or viability)",
"the human lineage", and
"There are other scenarios where the resources are not
limiting".
Those are all irrelevant confusion factors that divert your attention
from the above issue.
======
> .... That is ReMine's argument,
> and he is wrong about that.
That is not my argument. Felsenstein garbled it, in another attempt to
divert the issue. Let me re-emphasize this point. You are not likely
to understand Haldane's Dilemma until Felsenstein's classic error is
dispensed with. And so long as Felsenstein's error stands, it
demonstrates my thesis -- Haldane's Dilemma was never solved, it was
merely obscured and brushed aside.
> I'll pass over all the passages in ReMine's posting
> where he accuses me of knowingly spreading falsehoods
> and misleading the readership. ...
Felsenstein decides to "pass over all the passages" because there aren't
any. I never accused him of any such thing.
-- Walter ReMine
_The Biotic Message_
http://www1.minn.net/~science
My previous post:
http://www.deja.com/getdoc.xp?AN=701023335
Felsenstein's post:
http://www.deja.com/getdoc.xp?AN=703078134
John Edser
JMcG Wrote:-
>snip<
> I have issue with the notion that this
> phenomena cannot be described as being an aspect of natural selection.
> Let me put it this way, it's somewhat naive (for want of a better word)
> to assume that the process of natural selection only produces phenomena
> that are beneficial to the survival of biological phenomena WHEN VIEWED
> FROM THE PERSPECTIV OF ONLY ONE [OR A FEW]LEVEL[S] OF SELECTION. There
> are some rather obvious examples that demonstrate the validity of what
> I'm saying on this issue. The first being the fact that lifeforms are
> designed with senescence (the ability to die of old age). Another being
> that many of the cells in our bodies are designed to die. This is a
> natural process. It's a process that has been selected for by the
> process of natural selection.
>snip<
JE:-
Perhaps JMcG may like to propose one TESTABLE
theory i.e. outlining all the valid tests that
are required of nature, where MORE than
"just one (or a few) levels of
selection" is assumed. Perhaps he may
also include a self consistent but simplified
model view of it to help any "nieve" thinkers
understand exactly how such a "theory" works.
We would all be very interested to study
such a view!
Tim Tyler
[snip]
: JE:-
: Perhaps JMcG may like to propose one TESTABLE
: theory i.e. outlining all the valid tests that
: are required of nature, where MORE than
: "just one (or a few) levels of
: selection" is assumed. Perhaps he may
: also include a self consistent but simplified
: model view of it to help any "nieve" thinkers
: understand exactly how such a "theory" works.
: We would all be very interested to study
: such a view!
The existence of species selection and group selection are established in
theory. Group selection has been demonstrated in the lab, while
species-selection is relatively uncontroversial, since species reproduce
(by something akin to cloning) and exhibit subsequent variation.
The controversy is over to what extent these factors are significant in
nature, and whather it is useful to attribute adaptations to them very
frequently.
(Yes, there are also other "levels of selection", which it is profitable
to consider, on smaller scales as well as larger).
--
__________ http://alife.co.uk/ http://mandala.co.uk/
|im |yler t...@cryogen.com http://hex.org.uk/ http://atoms.org.uk/
Kent Paul Dolan
Let's change the rules slightly, to break the problem into smaller bits.
1) There is a population of scantily clad sheep at the carrying
capacity of the environment. No heavily clad genes or sheep are among
them
2) Of necessity, the food budget per sheep is divided among adding
biomass to reach reproductive maturity, staying warm, and reproducing
and nurturing a lamb to the age of independence, other necessities
neglected for simplicity.
3) The climate cools. Still no wooly bear genes, so the food budget
balance has to shift in favor of staying warm (posit "shivering
costs"), reducing energy available for growth, reproduction, and
nurturing.
4) Result: the carrying capacity of the environment drops, sheep
population decreases.
5) [sidetrack] keep changing the environment in deleterious ways,
eventually you are down to no sheep, sheep are extinct.
6) So far nothing to do with substitution of genes, but same results.
7) Back to 4, now woolier sheep mutation occurs, one sheep is able to
dedicate less food budget to (say) shivering, so more to growth,
reproduction, nurturing.
8) This gives woolier sheep a reproductive advantage, over time woolier
sheep grow to maturity faster, reproduce more enthusiastically, nurture
more successfully, increase population to the disbenefit of scantily
clad sheep, latter are displaced from the gene pool.
9) Population returns to a new higher carrying capacity, now consisting
of woolier sheep, which could differ from the original one in either
direction.
10) Now, with the parts separated out like this, please explain to me
what the heck it is that you are arguing about?
11) In what essential way does the situation where the mutation follows
the added environment stress differ from the situation where it is
lying fallow in the population until the environmental stress occurs,
with respect to whatever it is you folks are being rude to one another
about?
How, if all the bad environmental effects to the population can occur
in the absense of the beneficial mutation, does it make sense to credit
them to that mutation when it is fallow in the population?
xanthian, confused as ever.
--
Kent Paul Dolan.
<xant...@well.com> <xant...@aztec.asu.edu>
Joe Felsenstein
I was, in my 1971 paper, considering a scenario very much like this. In mine
the wooliness mutation exactly compensated for the cooler environment.
I argued that the cost of natural selection (substitutional load) is to be
calculated as the reproductive excess that the population needs to prevent
extinction if the environmental changes occur at regular intervals, each
ultimately counteracted by a substitution.
Walter Remine was asserting that I had made an elementary when he said:
> The cost of substitution - Felsenstein's mistake
>
> Joe Felsenstein writes:
>
>> If ReMine thinks there is a passage in my paper
>> that says you can go from few of a mutant to many
>> of that mutant without there being any reproductive
>> excess let him cite that passage.
>
>Here is an example from his paper:
>
> "The occurrence and substitution of favorable
> mutants in the absence of environmental change
> does not impose any cost, but makes the population
> better able to bear the cost." (Felsenstein, 1971, p 11)
In my 1971 paper I did of course have that sentence. Three sentences before
it I said "The cost of gene substitution can be defined as the reproductive
excess necessary to prevent extinction when the population is at a low
density." It should be crystal clear that this definition does not cause
there to be any cost when an advantageous mutation occurs in the absence
of any deterioration of the environment. In that case the mutation
"makes the population better able to bear the cost" (the cost of
other things, of course). I did not define the cost as being the reproductive
excess necessary to allow change of numbers of genes from few to many.
Remine may want to define it that way, but that's his affair and not any
mistake of mine.
You may wonder why all the fuss about whether or not I made such a mistake.
Fundamentally Remine wants to argue that the observed rate of substitution
in the origin of modern humans was too high to be consistent with the
available reproductive excess. He is not arguing that the population would
then have been at risk of extinction, but that there would have been a conflict
between two observables, and therefore that humans did not evolve that way (or
at all). The issue of whether cost must be defined the way Remine wants to
is important in persuading his audience that he has seen truths that us
professionals are deliberately obscuring.
>11) In what essential way does the situation where the mutation follows
>the added environment stress differ from the situation where it is
>lying fallow in the population until the environmental stress occurs,
>with respect to whatever it is you folks are being rude to one another
>about?
In no essential way, and in my 1971 paper you will see that I am in fact
dealing with the case where the mutation is "lying fallow". It is a
deterministic model without mutations but with initial frequencies of each
mutant assumed to be p_0, and the cost is calculated in terms of that initial
frequency.
>How, if all the bad environmental effects to the population can occur
>in the absense of the beneficial mutation, does it make sense to credit
>them to that mutation when it is fallow in the population?
It doesn't. The whole point of my 1971 paper was to say that (pages 4-5)
"there ... is a cost or load in this model, the cost being imposed by the
changes in the environment." I.e., imposed by the environmental deterioration.
Of course, Remine's "cost" is something different.
The sheep example is a fine one for my definition of cost.
laser...@my-deja.com
> Excuse me while I interrupt with a
> large quantity of ignorance.
Dolan's questions are worthy. Yet they would seem to be answered by a
large quantity of ignorance in the literature -- that is, no one, not
anyone, can cite an evolutionary book that supports Felsenstein's thesis
(and thereby resolve Dolan's questions). We've been arguing
Felsenstein's thesis here for several months now, and no one can support
his thesis by citing evolutionary books. AND NO ONE HAS! Not even
Felsenstein! The cost issue is over forty years old, and clear-cut
answers OUGHT to be in the books. But they aren't. Surely this is
curious!
This is because the evolutionary books do not support Felsenstein's
thesis, they contradict it.
For my part, I point out the contradiction and the lack of pursuit by
evolutionists. Here it is again: Felsenstein claims a beneficial
substitution can occur with ZERO cost of substitution (specifically he
says this occurs in a non-deteriorating environment). I point out that
NONE of the published formulas yield a cost of zero. TRY IT! I
challenge you to simply apply the published formulas -- the cost is not
zero. I also point out Felsenstein's conceptual error -- the growth of a
trait from few to many REQUIRES reproductive excess, therefore the cost
of substitution is positive, non-zero, and Felsenstein is wrong.
This matter is simple, embarrassingly so. Yet evolutionists here do not
resolve it -- thus proving my thesis. Haldane's Dilemma was never
solved, it was merely garbled and brushed aside. Sufficient evidence of
this is already present on this newsgroup. Nothing could be more basic
to the problem, yet it drags on and on.
-- Walter ReMine
_The Biotic Message_
http://www1.minn.net/~science
Fellow with Discovery Institute
Dolan's post:
http://www.deja.com/getdoc.xp?AN=715014649
My previous post:
http://www.deja.com/getdoc.xp?AN=694884714
laser...@my-deja.com
Kent Paul Dolan wrote:
> Let's change the rules slightly,
> to break the problem into smaller bits.
Dolan's post needlessly complicates the issue with references to the
"environment".
Here is the issue again, and it is not about the environment. Evolution
requires substitutions, where new traits go from 'few' to 'many' in
number. This process REQUIRES reproductive excess -- absolutely,
positively, no exceptions, and no excuses. Therefore the cost of
substitution is not zero, and established formulas further show this
point.
laser...@my-deja.com
(b) the existing formulas contradict his claims of ZERO cost, (c) the
existing evolutionary literature abundantly (albeit implicitly)
contradicts Felsenstein's thesis, and (d) I predicted that Felsenstein
would eventually attempt to escape into a self-serving "definition" of
cost -- and this is his attempt in his latest post.
That is, his latest post emphasizes HIS definition. Most readers here
have been unaware that Felsenstein's definition is different, in fact it
contradicts the others in the literature. This again demonstrates my
point that confusion on the cost issue is rampant (and is routinely
allowed to remain rampant).
He defines the cost of substitution as follows:
> I argued that the cost of natural selection
> (substitutional load) is to be calculated as the
> reproductive excess that the population needs
> TO PREVENT EXTINCTION
> if the environmental changes occur at regular
> intervals, each ultimately counteracted by a
> substitution.
[WJR: My emphasis]
Thus, his definition is the cost of "preventing extinction" (his words),
not the cost of substitution.
Felsenstein uses his (mis-)definition to claim there is zero cost of
substitution for situations where extinction is not imminent.
> ....
> It should be crystal clear that THIS DEFINITION
> does not cause there to be any cost when an
> advantageous mutation occurs in the absence
> of any deterioration of the environment.
[WJR: my emphasis]
Once again Felsenstein fails to point out to his many readers that his
claim is contradicted by the remaining evolutionary literature. The
confusion just goes on and on.
Felsenstein also specifically rejects the essential defining feature of
substitution -- the growth of the trait from few to many in number.
Here it is:
> I did NOT DEFINE the cost as being the
> reproductive excess necessary to allow change
> of numbers of genes from few to many. ...
> [But that is] not any mistake of mine.
[WJR: my emphasis]
******
> You may wonder why all the fuss about
> whether or not I made such a mistake.
I have been straightforward about this from the beginning. The fuss
about Felsenstein's error is because it has lasted several DECADES
unchallenged, and I am presenting it to DEMONSTRATE to you (by one of
many possible examples) that Haldane's Dilemma was never solved, it was
merely garbled and brushed aside.
> Fundamentally ReMine wants to argue that the
> observed rate of substitution in the origin of
> modern humans was too ...
Fundamentally Felsenstein is trying to change the subject, away from his
long-standing error and the mass confusion about the cost of
substitution. I say, first things first. Before you can understand
Haldane's Dilemma, you must understand the cost of substitution -- and
Felsenstein has thoroughly garbled it.
> .... ReMine wants to [persuade] his audience that he
> has seen truths that us professionals are deliberately
> obscuring.
Felsenstein is misrepresenting me, I never claimed he (or anyone) was
"deliberately obscuring" anything. I challenge him to find an example
of it. Moreover, I point out that "us professionals" (whoever they are)
CONTRADICT FELSENSTEIN!!! Other evolutionists rampantly contradict
Felsenstein, he has no basis for calling on them for support. I call on
evolutionists to resolve this fundamental issue about the cost of
substitution.
-- Walter ReMine
_The Biotic Message_
http://www1.minn.net/~science
Fellow with Discovery Institute
Joe Felsenstein's post:
http://www.deja.com/getdoc.xp?AN=716238724
Don Cates
>
>Kent Paul Dolan wrote:
>> Let's change the rules slightly,
>> to break the problem into smaller bits.
>
>Dolan's post needlessly complicates the issue with references to the
>"environment".
>
>Here is the issue again, and it is not about the environment. Evolution
>requires substitutions, where new traits go from 'few' to 'many' in
>number. This process REQUIRES reproductive excess -- absolutely,
>positively, no exceptions, and no excuses. Therefore the cost of
>substitution is not zero, and established formulas further show this
>point.
I think the environment is very important. I posted the following
several weeks ago and have seen no comment from ReMine.
ISTM that reproductive excess refers to populations, not individuals. So
if the new mutant(s) has(ve) two offspring and the coresponding number
of nonmutants have no offspring (make that three and one for reproducing
pairs), then the population will remain constant but the new mutation
will spread.
eg. Say the mutant gathers food more quickly and reproductive rate
depends on nutrition. Where the mutant competes with a nonmutant for
food, it tends to have one more descendent than replacement and the
nonmutant has one less. Where mutants compete with each other, they just
replace themselves. Where nonmutants compete, they just relace
themselves. The population will remain constant but the mutation will
take over the population.
Don Cates
ca...@cc.umanitoba.ca
Kent Paul Dolan
>Kent Paul Dolan wrote:
>> Let's change the rules slightly,
>> to break the problem into smaller bits.
>Dolan's post needlessly complicates the issue with references to the
>"environment".
Umm, since evolution and reproductive success are all and entirely with
respect to an environment, and to changes in that environment, any
argument that purports to be sound without regard to the environment is
so suspect as to be discardable without further consideration.
>Here is the issue again, and it is not about the environment. Evolution
>requires substitutions, where new traits go from 'few' to 'many' in
>number. This process REQUIRES reproductive excess -- absolutely,
>positively, no exceptions, and no excuses. Therefore the cost of
>substitution is not zero, and established formulas further show this
>point.
Sigh.
Let's break that down into small enough pieces to see where the nonsense
creeps into the text.
1) it is not about the environment
then it is not about evolution, which is only meaningful with respect
to an environment
2) evolution requires substitutions
no, evolution requires _selections_, not substitutions; given a
substantial population, the traits may exist without reproductive
effect: neutral differences, gathered over time by blind chance,
like (perhaps) green versus blue eye color; until and if a shift in
competitive advantage occurs to favor some traits and disfavor
others, at which time they become subject to selections pro or con
such a change is a change in the reproductive environment
3) where new traits go from "few" to "many" in number
it is effectively nonsense to "count" traits, since one observer's
"trait" is another observer's "meaningless difference"; one can only
_select_ traits; i.e., something isn't a meaningful trait from
evolution's viewpoint until it makes a difference, and it doesn't
make a difference until the reproductive environment says it does
moreover, since mutation's probability is essentially continuous
when smeared out over time, mutations helpful, neutral, or harmful
are occuring at a fairly steady rate, so "new traits" are _always_
"going from few to many in number" if that were meaningful in any
case, and so the cost of carrying the constant innovation due to
mutation is a built in cost whether or not any selection pressure is
added or removed by environmental changes
thus, this carrying cost is _independent_ of evolution, not _a part
of_ evolution; populations stuck in a rut still must over-reproduce
the carrying capacity of the environment simply to provide for the
(predominantly adverse) effects of mutation on the survival of
offspring
4) this process REQUIRES reproductive excess ...
shouting your opinions makes them no more true and much less easily
acceptable
reproductive excess is a requirement of an ongoing and inescapable
process of mutation, is independent of whether environmental forces
select for, ignore, or select against such mutations once they have
occurred, and therefore does not agree with your shouted opinions
5) established formulas further show this point
funny about that; I just ran into, and tried out, a bit of
revisionist doggerel called BDM.pl posted in the last couple of days
over in one of the computational evolution newsgroups showing that
the conventional wisdom about mutational bias versus selection bias
wasn't all that wise after all when put to an actual test; things
for me when running a test ran 18 to 2 against the conventional
wisdom in 20 test runs
perhaps not only explicitly quoting but also thoroughly defending
and testing with real data some of these "established formulas"
whose support you claim, carried out in the light of present
knowledge, would be an interesting exercise for you
it would surely be more convincing than your present zero data based
"who can shout loudest" approach to the issues
By the way, you'll notice that I'm not particularly making friends with
your opponent by my version of reality either. I rarely make friends
with anyone with my tactless approach to deflationary universality of
discourse.
Also known as directing the pin toward the hot air source.
Cheers!
xanthian, having a low hot air tolerance lifetime.
===== random archival quality quote =====
As a computing professional, I believe it would be unethical for me to
advise, recommend, or support the use (save possibly for personal
amusement) of any product that is or depends on any Microsoft product.
-- David H. Wolfskill <da...@catwhisker.org>
Kent Paul Dolan
>He defines the cost of substitution as follows:
>> I argued that the cost of natural selection
>> (substitutional load) is to be calculated as the
>> reproductive excess that the population needs
>> TO PREVENT EXTINCTION
>> if the environmental changes occur at regular
>> intervals, each ultimately counteracted by a
>> substitution.
>[WJR: My emphasis]
>Thus, his definition is the cost of "preventing extinction" (his words),
>not the cost of substitution.
>From private communications with you, it seems that you have this
problem reading simple English pervasively. He did not say merely
"is", he said "is to be calculated as", quite different concepts.
For example "four" _is_ a specific integer number, the Peano postulates
successor to "three". One might say that four "is to be calculated as"
the sum of the first prime integer with itself. One might also say
that four "is to be calculated as" the sum of unity with the second
prime integer. The latter two statements stand in opposition to one
another, yet neither changes the essense of the first statement.
Until you can argue against the text rather than your misinterpretation
or misrepresentation of the text, you are quite unlikely to be taken
seriously. Until you can express your arguments as something other
than a personal vendetta, you are unlikely to be taken seriously even
when correct.
Were there the disagreement that you claim in the published literature
to a definition put forth 30 years ago, perhaps you could cite one
single voice beside your own that took on this _specific_ definition
for the _specific_ purpose of refuting it in all that long time, rather
than muzzily claiming _implicit_ disagreement to exist there?
Scientists tend to jump on often repeated obviously unsupportable very
public claims rather more quickly than after a three decade lag, and to
phrase their opposition as to the mistaken concept rather than to the
mistaken author, a skill you have failed to master.
Cheers!
xanthian.
Not a great fan of zealots, today, somehow. Probably low on meds again.
===== random archival quality quote =====
"There are only two kinds of math books. Those you cannot read beyond
the first sentence, and those you cannot read beyond the first page."
-C.N. Yang, as quoted by:
-- Suchandra Thapa <s-t...@uchicago.edu>
laser...@my-deja.com
> Umm, since evolution and reproductive success
> are all and entirely with respect to an environment,
> and to changes in that environment, any argument
> that purports to be sound without regard to the
> environment is so suspect as to be discardable without
> further consideration.
> ....
> 1) it is not about the environment
>
> then it is not about evolution, which is only meaningful
> with respect to an environment
Dolan misrepresented what I said. I said the issue is not about the
environment -- that is, the cost of substitution is not zero regardless
of the environment. The published formulas show this.
> 2) evolution requires substitutions
>
> no, evolution requires _selections_, not substitutions; ...
Dolan claims evolution does not require substitutions. Set your watches
and see how long evolutionists let that go unopposed.
> 3) where new traits go from "few" to "many" in number
>
> it is effectively nonsense to "count" traits,
There go the evolutionary genetics textbooks into the Dumpster. Again,
set your watches and see how long evolutionists let that go unopposed...
> .... so the cost of carrying the constant innovation due
> to mutation is a built in cost whether or not any selection
> pressure is added or removed by environmental changes
> thus, this carrying cost is _independent_ of evolution,
> not _a part of_ evolution; ....
The "cost" he is referring to above is not the cost of substitution.
> ... populations stuck in a rut still must over-reproduce ...
In other words, reproductive excess is required.
-- Walter ReMine
_The Biotic Message_
http://www1.minn.net/~science
Fellow with Discovery Institute
> I rarely make friends with anyone
> with my tactless approach ...
Kent Paul Dolan's post:
http://www.deja.com/getdoc.xp?AN=719863797
My previous post:
http://www.deja.com/getdoc.xp?AN=718062036
laser...@my-deja.com
> I think the environment is very important.
The environment has nothing to do with the issue at hand -- that is, the
cost of substitution is not zero, no matter what the environment. The
published formulas show this.
> [It seems to me] that reproductive excess
> refers to populations, not individuals.
The evolutionary literature on the cost of substitution is not
altogether clear on the matter. I challenge Cates to cite evolutionary
literature that makes it clear.
For a trait to increase from few to many in number, reproductive excess
is required. For example:
> So if the new mutant(s) has(ve) two offspring …
Then there is reproductive excess. Cates is attempting to define it
away.
-- Walter ReMine
_The Biotic Message_
http://www1.minn.net/~science
Fellow with Discovery Institute
Don Cates's post:
http://www.deja.com/getdoc.xp?AN=719161367
My previous post:
http://www.deja.com/getdoc.xp?AN=718062036
Kelly C. Kissane
>
> > So if the new mutant(s) has(ve) two offspring …
>
> Then there is reproductive excess. Cates is attempting to define it
> away.
>
Reading this thread is like driving by a fatal accident. It's grotesque,
but you can't help looking anyway.
Well, I need some explanantion here :
If two parents produce two offspring, how in hell is that reproductive
excess? They replace each other and each other only. Sounds like
replacement, not excess.
--
Kelly C. Kissane
Dept. of EECB
University of Nevada
Reno, NV 89557
John Edser
Walter ReMine Wrote:-
> > I think the environment is very important.
> WR:-
> The environment has nothing to do with the issue at hand -- that is, the
> cost of substitution is not zero, no matter what the environment. The
> published formulas show this.
JE:-
ReMine is partly correct, this issue does concern
how many entities can be reproduced within one
defined population of them. The other missing
part of Haldane's Dilemma is how many genes
within each individual entity within one population
are needed to be substituted within a
common ancestor, to change it into both a man and a
chimp. Would changing an chimp into a man today, require
substituting millions of genes or just a handful
of them? What number of genes are required to
be substituted for the dilemma to be solved? Nobody
is saying. The Neo Darwinists are not saying,
since that may require them to commit themselves to
a much, much smaller number of genes than was first
thought necessary. This in turn would refute the central
tenet of population genetics that assumes each gene must
have only an additive fitness to every other gene.
Not enough gene substitutions could ever be
produced to explain the differences between
man and ape if genes only have an additive fitness
complexity component between themselves within one
organism, since very large numbers of gene substitutions
would be required in this case. On the other
hand if all the genes were non additive in fitness
then a geometrically lessor number of genes would
only now be required.
The supposition of non_ additive gene fitnesses,
could remove Haldane's Dilemma by reducing
the number of substitutions necessary. This would
however refute the "selfish gene" view, also refuting
organism fitness altruism within nature.
This would throw out all the issues of organism
altruism within evolutionary theory including
Triver's "reciprocal altruism" and Hamilton's
view of "included fitness". These are central tenets
of today's very popular, sociobiology, which are
being protected. Walter ReMine is correct to suggest
that today's Neo Darwinists appear to be white-washing
the whole additive gene substitution problem within their
simplified mathematical models. However these models
are just a misrepresentation of the problem.
The non evolutionists always move the goal posts,
re: what defines "one species" so that evolutionists
can never make any definitive case for their view,
by refuting the non evolutionists case, that evolution
cannot cause the transmutation of species.
As soon as one species is observed to evolve into
another it is no longer classed as a separate species
by the non evolutionists! It seems Neo Darwinists have
learnt from this simple trick and move a few goal posts
of their own, when it becomes necessary to protect their
own interests. I refer readers to "Felsenstein's Dilemma"
within sbe (see below).
The most basic problem in Haldane's Dilemma has always
been what _exactly_ are the biological entities
in which the genes reside and what is the type of
relationship that exists between the genes within
them. The entities are organisms and the relationships
between the genes within each organism are non additive.
> > [It seems to me] that reproductive excess
> > refers to populations, not individuals.
> WR:-
> The evolutionary literature on the cost of substitution is not
> altogether clear on the matter. I challenge Cates to cite evolutionary
> literature that makes it clear.
JE:-
The population just provides a limited
domain within which a genetic substitution
must occur within each individual organism. As far as
I am concerned, if reproductive excess is applied
strictly to "populations," then this requires entire
populations to reproduce other populations as separate
entities, and not just the individual's within them.
Conversely, if individuals are the defined units of
reproductive excess then only single individuals need
to reproduce themselves to provide it. The former view
is group selective while the latter is classically
Darwinian. One view entirely excludes the other.
All the confusion lies within the many simplified
population genetics model's of Darwinism. One
individual gene is often made equal to
one individual within a hypothetical haplotype
(a haplotype has only one set of genetic information).
This was clearly the case re: Prof. Felsenstein's
posted model to sbe (again I refer readers to the thread
"Felsenstein's Dilemma" available at Deja News).
Within Prof. Felsenstein's posted model, the population P
had to be substituted. Here allele A' must replace
allele A, totally, within population P. Each
individual haplotype was either A or A'. Felsenstein's
solution only provided a substitution for 0.1 of P leaving,
0.9 of it, unsubstituted. This was the missing reproductive
excess that ReMine quite correctly insisted, _must_
be accounted for within Felsenstein's "solution".
It appears that Prof. Felsenstein simply evaded
this payment by "redefining" exactly what population
had been required to be substituted. Here, only a sub
population that comprised of 0.1 A' alleles was now
being defined by Prof. Felsenstein as the
"fully substituted" population, away from the original
population P, so that the 0.9 reproductive excess
that was required within the original problem
was "successfully" evaded within his "solution".
Prof. Felsenstein simply moved the goal posts to
be closer together for his own benefit. The
non evolutionists move the species goal posts
further away from each other, to suit their
interests. In neither case are the interests of
science ever served.
In many simplified population genetics models, a
population of independent (additive) individuals
now becomes just a population of independent (additive)
genes. This is what any mathematician requires since
non additive genetic epistasis and pleiotrophic effects
produce in turn, non additive complexity and subsequently
highly geared relationships between genes. This is too complex
to model but has the virtue that only a change in a handful
of genes produces a geometric increase in complexity within
a single organism.
Haldane's Dilemma is based on the over-simple assumption
that a many genes must be substituted because all the gene's
within Haldane's population genetics model only ever had
an arithmetic complexity function because they were
all assumed to be "additive" in fitness. Thus many genes
were required to be substituted. The complexity of
an organism is NOT just an additive function re: all
of one organism's genes!
Hadane's dilemma has a lateral solution, if and only if,
the real, known and testable non additive relationships
between genes within biology are _included_ within the
question. These were originally excluded simply because the
mathematics was too complicated. It appears the
mathematics was much more important than the biology.
However the biology has always been way ahead of the mathematics.
Haldane's dilemma is really just a storm in a mathematician's
teacup. It has always been just an artificial problem set
within the domain of just an artificial simplified model
of nature. In principle all that is required to
build a solution is a knowledge of very basic synthetic
biology. No organism is ever the sum of its parts.
This being the case, some parts are very much
more important than other parts. Thus major changes
are easily produced by small differences in the
more important, highly geared, parts. These different
highly geared parts include highly geared genes.
Thus in principle, far fewer highly geared genetic
differences are required to change a man into an ape
from a common ancestor, than Haldane ever thought
possible. This has been confirmed via today's
observations. Within any one genome less than
10% of it seems to code for anything. Of that
critical 10%, over 98% is the same between man
and ape. It seems that only small differences in the
actual number of genes may be all that is required
to turn a man into an ape. This provides a
basic question that is much more interesting than
Haldane's original dilemma: Where does ALL the missing
genetic information really reside and what is it mostly
comprised of?
If any definitive answer to Haldane's Dilemma
is to be produced, the unit of selection issue
must firstly be resolved so we all know what we
are talking about. Are we talking about DNA
substitutions within numbers of reproduced cells
within individuals, numbers of reproduced whole
individuals or numbers of reproduced
whole populations?. There is no will to
tackle this most basic of problems within sbe
(or almost anywhere else). Ever since
the "multi unit" of selection approach became
popular, people just kept switching unequal units
of fitness mid argument, to evade refutation of
their own views. This has become an epidemic of
"goal post moving" irrationality so that the science
of evolutionary theory has become non definitive.
laser...@my-deja.com
> The other missing part of Haldane's Dilemma is
> how many genes ... are needed to be substituted
> within a common ancestor, to change it into both
> a man and a chimp. Would changing a chimp into
> a man today, require substituting millions of genes
> or just a handful of them? What number of genes are
> required to be substituted for the dilemma to be solved?
> Nobody is saying. The Neo Darwinists are not saying,
> since that may require them to commit themselves to a
> much, much smaller number of genes than was first
> thought necessary.
-- Walter ReMine
_The Biotic Message_
http://www1.minn.net/~science
Fellow with Discovery Institute
Edser's post:
http://www.deja.com/getdoc.xp?AN=721588248
My previous post:
http://www.deja.com/getdoc.xp?AN=721025981
John Edser
Kelly C. Kissane wrote:-
> > > DC:-
> > > So if the new mutant(s) has(ve) two offspring …
> > WR:-
> > Then there is reproductive excess. Cates is attempting to define it
> > away.
> KCK:-
> Reading this thread is like driving by a fatal accident. It's grotesque,
> but you can't help looking anyway.
> Well, I need some explanantion here
> If two parents produce two offspring, how in hell is that reproductive
> excess? They replace each other and each other only. Sounds like
> replacement, not excess.
JE:-
I have repeatedly asked for _explicit_
definitions of the term "reproductive
excess" from both ReMine and Felsenstein
since it _does_ matter how you define it,
re: Haldanes Dilemma.
ReMine has replied that he is using a
"standard definition" but refuses to
quote it on sbe. Felsenstein has only
implicitly defined it as the number
of reproduction's necessary to keep
one population constant. Thus the
situation above, where parental organism's
just replace themselves is zero reproductive
excess within Felsenstein's implicit view
of reproductive excess. Since ReMine suggests
that the above has more than zero reproductive
excess, his view must differ to Felsenstein's view.
ReMine insists his view is the "standard view".
If only ReMine would post the definition he has
accepted as the "standard view" of reproductive
excess this matter could be cleared up.
However, Felsenstein is _not_ consistent with
the use of his own implicit use of reproductive
excess. I refer Kelly to the thread, Felsenstein's
Dilemma available at Deja News. In essence Felsenstein
posted a simple model to illustrate his position.
It did so, _very_ clearly. Felsenstein moved the
goal posts mid argument and evaded the reproductive
excess that ReMine correctly insisted must be met
under Felsenstein's own implicit definition of
reproductive excess. Felsenstein refuses to
reply to this simple testable charge, thus I must
assume that Prof. Felsenstein agrees my charge,
was correct.
In common sense terms, all this part of
the debate is about concerns one mutant, in one
gene, in one organism. If it is beneficial within
a finite mixed population of say 100 individuals,
how much reproductive work has to be done
for that one mutant to be substituted in
that population? The answer is quite simple,
99 reproductions must be produced over the
lifespan of the original mutant parent. Thus
the reproductive excess is 0.99. If the
mutant parent dies but reproduces itself
just once and the mutant sub population
remains constant at 0.01, then clearly, in this
case, no substitution within this population
of 100 individuals is at all possible. This
is ReMines point. When he say's that reproductive
excess is ALWAYS required he is simply stating the
obvious, i.e. a single mutant always requires a
reproductive effort over and above its replacement,
to substitute itself within any mixed finite population.
He is entirely correct.
John Edser
Independent Researcher
PO Box 266
Church Point
NSW 2105
Australia
Kelly C. Kissane
>> genes or just a handful of them? What number of genes are
Since sequencing the genome of humans and chimps doesn't equal knowing
what each of the thousands of genes do, the very simple truth is = we
don't know.
If the theory (one of many) that humans are neotenic chimps, we may
learn clues fairly soon. Several researchers are working on domestic dogs
to determine which genes coded for their neotenic wolf features.
> > Nobody is saying.
Nobody, including the ever vocal me, is saying because WE DON'T KNOW WHAT
EACH AND EVERY GENE DOES. Damn, have some patience.
>> The Neo Darwinists are not saying,
> > since that may require them to commit themselves to a
> > much, much smaller number of genes than was first
> > thought necessary.
A snide, very unfair remark. I won't repeat myself, read the above.
Personally, I could care less if it's 10 genes or 1000 or even 100,000.
I'm only interested in the scientific facts, not snide or pompous remarks.
John Edser
Re: Haldane's Dilemma
> >> JE:-
> >> Would changing a chimp into a man today, require substituting
> >> millions of
> >> genes or just a handful of them? What number of genes are
> >> required to be substituted ?
> KCK:-
> Since sequencing the genome of humans and chimps doesn't equal knowing
> what each of the thousands of genes do, the very simple truth is = we
> don't know.
JE:-
A honest answer,
thank you.
> KCK:-
> If the theory (one of many) that humans are neotenic chimps, we may
> learn clues fairly soon. Several researchers are working on domestic dogs
> to determine which genes coded for their neotenic wolf features.
JE:-
In that case, perhaps only a smaller number
of genes would need to be substituted rather
than a larger number?
> > > JE:-
> > > Nobody is saying.
> KCK:-
> Nobody, including the ever vocal me, is saying because WE DON'T KNOW WHAT
> EACH AND EVERY GENE DOES. Damn, have some patience.
JE:
That does not matter within the gene centric
model that produced Haldane's Dilemma since
every gene in that model is regarded as additive
in fitness to every other gene in the gene pool
no matter what it actually does in nature.
> > > The Neo Darwinists are not saying,
> > > since that may require them to commit themselves to a
> > > much, much smaller number of genes than was first
> > > thought necessary.
> KCK:
> A snide, very unfair remark. I won't repeat myself, read the above.
> Personally, I could care less if it's 10 genes or 1000 or even 100,000.
> I'm only interested in the scientific facts, not snide or pompous remarks.
JE:-
The number effects the scope of the problem!
It is much easier to substitute a smaller
number than a larger number.
The facts are all I am interested
in as well, _including_ any theory that
integrates them and makes testable sense
out of them.
If by simply suggesting "that a
much smaller number of genes than was first
thought necessary" constitutes to you a "snide
or pompous" remark (?!?), then I must have really
hit a raw nerve! Here is the nerve I hit:
As I stated previously, the central tenet to all
organism fitness altruism arguments within Neo
Darwinism assumes additive independent gene fitnesses
per genome. These alone were used to calculate
Haldanes Dilemma. If that tenet goes then "selfish
genes" all go along with "reciprocal altruism" and
"inclusive fitness" just to name a few. All of these
constitute the main pillars of today's Darwinism.
It is no mystery to me why most Neo Darwinists are
a little "reluctant" to suggest that by far fewer
genes need to be substituted than they 1st
thought....
John Edser
Independent Researcher
PO Box 266
laser...@my-deja.com
Kelly Kissane writes:
> Well, I need some explanation here :
>
> If two parents produce two offspring, how in hell is
> that reproductive excess? They replace each other
> and each other only. Sounds like replacement, not excess.
Kissane is referring to my post, which is referring to Don Cates's post.
Our discussion made it clear it was ONE parent producing TWO offspring
-- or in the case of mating pairs, it was two parents producing THREE
offspring. In either case, there is reproductive excess. Here is Don
Cates's original wording:
[Cates]
> So if the new mutant(s) has(ve) two offspring
> and the corresponding number of non-mutants
> have no offspring (make that three and one for
> reproducing pairs), then the population will remain
> constant but the new mutation will spread.
Kelly made the above mistake, and not one evolutionist stepped in to
correct it -- Not Felsenstein. Not Cates. Not Kissane. Not one. Yet
the mistake is WAY TOO simple to have stumped all evolutionists here.
There are evolutionist readers here who knew of the error, yet remained
silent. This is another example of what I have been documenting here on
sci.bio.evolution. When it comes to Haldane's Dilemma, errors are not
pursued, instead evolutionists allow errors to thrive. It's not a
conspiracy. And it's not intentional. But it's not random either.
[moderator's note: It's worth pointing out that most pop bio models
deal with the number of females in a population; it's a simplification
and one that allows that "one offspring" replaces the reproducing
(female) adult. This is a matter of usage, not obfuscation. - JAH]
Haldane's Dilemma was never solved, it was merely garbled and brushed
aside.
-- Walter ReMine
_The Biotic Message_
http://www1.minn.net/~science
Fellow with Discovery Institute
Kelly Kissane's post:
http://www.deja.com/getdoc.xp?AN=721586498
My previous post:
http://www.deja.com/getdoc.xp?AN=721025981
Sent via Deja.com
http://www.deja.com/
John Edser
was not posted]
laser...@my-deja.com
diversion directly into my post. He wrote:
> [moderator's note: It's worth pointing out that most
> pop bio models deal with the number of females in
> a population; it's a simplification and one that allows
> that "one offspring" replaces the reproducing (female)
> adult. This is a matter of usage, not obfuscation. - JAH]
Contrary to what the moderator said, my point had nothing whatever to do
with models, terminology, or their "matter of usage". Rather, my point
speaks for itself. Here is my wording again:
[BEGIN QUOTE from my previous post]
Kelly [Kissane] made the above mistake,
and not one evolutionist stepped in to correct it --
Not Felsenstein. Not Cates. Not Kissane. Not one.
Yet the mistake is WAY TOO simple
to have stumped all evolutionists here.
There are evolutionist readers here who knew
of the error, yet remained silent. This is another
example of what I have been documenting
here on sci.bio.evolution. When it comes to Haldane's
Dilemma, errors are not pursued, instead
evolutionists allow errors to thrive. It's not a
conspiracy. And it's not intentional. But it's
not random either.
[END QUOTE]
If the moderator felt his point was worthy, he could have inserted it
into Kissane's post (where the original mistake was made). Or he might
have posted separately, to clear up her mistake. Instead he (like all
evolutionists here) CHOSE TO REMAIN SILENT and allow Kissane's mistake
to thrive. That is, until I posted on the matter, whereupon he used his
position as moderator to reach right inside my post so as to more
effectively divert the issue.
-- Walter ReMine
_The Biotic Message_
http://www1.minn.net/~science
Fellow with Discovery Institute
My previous post:
http://www.deja.com/getdoc.xp?AN=725880600
John Edser
> Kelly Kissane writes:
> > Well, I need some explanation here :
> > If two parents produce two offspring, how in hell is
> > that reproductive excess? They replace each other
> > and each other only. Sounds like replacement, not excess.
> Walter ReMine Writes:
> Kissane is referring to my post, which is referring to Don Cates's post.
> Our discussion made it clear it was ONE parent producing TWO offspring
> -- or in the case of mating pairs, it was two parents producing THREE
> offspring. In either case, there is reproductive excess. Here is Don
> Cates's original wording:
> [Cates]
> > So if the new mutant(s) has(ve) two offspring
> > and the corresponding number of non-mutants
> > have no offspring (make that three and one for
> > reproducing pairs), then the population will remain
> > constant but the new mutation will spread.
> Kelly made the above mistake, and not one evolutionist stepped in to
> correct it -- Not Felsenstein. Not Cates. Not Kissane. Not one. Yet
> the mistake is WAY TOO simple to have stumped all evolutionists here.
JE:-
It really does not matter.
Prof. Felsenstein made the situation
re: "reproductive excess", somewhat clear.
However, he was less clear as to exactly
which population he was referring to
( I refer readers to the thread:
"Felsenstein's Dilemma" available
via Deja News.)
If a gene ONLY maintains its numbers,
then no matter how it exactly does so,
it must require only zero reproductive excess
to do it. This is just a very obvious statement
to make. Unfortunatley, Haldane's Dilemma is all
about unstated, i.e. implicit but obvious truths.
What is less obvious is _exactly_ which population
is being referred to. If one mutant appears
in one population, it can constitute a
separate "population" of exactly 1
individual. If 10 individuals
simultaneously mutate to the same
genetic form, they can constitute
a separate population of 10. If
either mutated population just maintains
its numbers, no matter how they do it,
then they are EITHER, fully
substituted populations of 1 and 10
individuals respectively, in their own
right, or they are just a samll part of
a larger NON substituted population.
To solve the problem re: the substitution
of the entire larger unsubstituted population
from the smaller fully substituted sub population
contained within it, reproductive excess is always
required by the fully substituted sub
population, no exceptions, period. This is
all Walter ReMine is insisting on and he
is prefectly correct to insist on it.
For the smaller sub population to reach
the size of the larger population it is
within, reproduction over and above its
maintenance levels is required, no
matter how it may do it. This is
of course just another very obvious
statement to make. Even more obvious
is that this required reproduction,
within nature, is not of genes but of
fertile individuals carrying these genes.
To solve the problem of
fully substituting a "population" of just 1
mutant, or even 10 mutants, is very easy,
since no problem even exists here; they were
both fully substituted "populations" to start
with.
Incredibly, Prof. Felsenstein moved
within his explanation from
the larger unsubstituted population to
the fully substituted smaller one,
and then back again, for his conclusion
that reproductive excess is not required
to substitute "the population" within
his own posted example, and nobody
even noticed, or its seems even cares,
after it has been pointed out to them.
Prof. Felsenstein had defined explicitly,
"absolute fitness" within sbe. This is the
total of an allele in one population.
Thus, Prof. Felsenstein simply switched
absolute fitnesses within his own posted
model to illegally remove reproductive excess.
In effect he robbed the bank by just
erasing all the debts he had to it;
a very neat trick.
Absolute assumptions cannot be switched
within any rational argument if they are
to remain rational. Felsenstein is not the
only one. Tim Tyler has admitted he commonly
switches units of selection and thinks nothing
of it. Units of selection are absolute assumptions
within evolutionary theory. He is the only Neo
Darwinist here, to admit to such switching. I
congratulate him on his honesty. However, when
challanged, like Felsenstein, he just refused to
discuss it and stalked off, i.e. both of them
have taken the easy way out.
The unit of selection is an absolute assumption
within Neo Darwinism and cannot ever be swiched.
Bob O'Hara switches it all the time but will
never admit that he does. McGinn, a good critic
of Neo Darwinism, refuses to admit that any absolute
assumptions are ever valid. This he stated
"absolutely"! McGinn assumes that all units of
selection evolve "simulaneously" but refuses
to define this "simultaneous unit" since if he
did, he would have produced a definitive, i.e.
absolute assumption. Why all this political dodging
and weaving within a so called scientific disiplin?
The answer is very simple, Neo Darwinism is
_entirely_ a simplified model of Darwinism not
the reverse, but nobody here will ever admit it.
This being the case, all arguments are forced to
switch absolute assumptions, like the units of
selection or modeling absolute fitnesses just to
make them appear rational. If they did not do
so, they would appear non sensical. This switching
takes the place of the required simplification
corrections that science insists any simplified
model must make before it can be validly used to
predict nature. Denial of this simple demmand is
almost complete. Nobody here will even admit that
all additive models of gene freq changes
are one vast over simplification of how genes are
selected in nature. Guy Hoezler admitted that population
genetics was a simplified view of nature. However
he refused to admit that if such simplifications,
are not corrected BEFORE predicting nature, the
view is formally refuted via its very own
over simplifications. Most people here it seems,
wants the luxury of being "right" without ever
supplying the required rigor to test for it.
When challanged on this point they invoke
post modern epistemology to justify it.
In such a situation nothing of science is left.
In reality, depite all the shinning mathematical
chrome, and all kudos give over to it, the Neo
Darwinistic views that have been put forward so far,
which try to include Mendel's genes within Darwin's
theory of natural selection, including population
genetics "the queen of evolutionary theory" have
been extremely CRUDE approximations, that were
then misused. They are nowhere near complete
and in this utterly crude and incomplete state, they have
been misused to predict nature without the required
corrections being made to them, beforehand. The arrogance
of those putting forward and misusing such views is a
lesson in itself as to how even science can be manipulated,
if needs be, for what can only be, politcal purposes.
John Edser
Independent Researcher
PO Box 266
Church Point
NSW 2105
Australia
This is the cancer that is eating the heart
out of evolutionary theory.
John Edser
Independent Researcher
PO Box 266
Don Cates
--------------included mms------------
From: laser...@my-deja.com (laser...@my-deja.com)
Subject: Re: Haldane's Dilemma
Newsgroups: sci.bio.evolution
Date: 2001-01-27 16:26:04 PST
> I think the environment is very important.
The environment has nothing to do with the issue at hand -- that is,
the cost of substitution is not zero, no matter what the environment.
The published formulas show this.
> [It seems to me] that reproductive excess > refers to populations, not individuals.
The evolutionary literature on the cost
of substitution is not altogether clear on the matter.
I challenge Cates to cite evolutionary literature that makes it clear.
For a trait to increase from few to many in number,
reproductive excess is required.
For example:
> So if the new mutant(s) has(ve) two offspring …
Then there is reproductive excess.
Cates is attempting to define it away.
--
Walter ReMine _The Biotic Message_ http://www1.minn.net/~science
Fellow with Discovery Institute
Don Cates's post: http://www.deja.com/getdoc.xp?AN=719161367
My previous post: http://www.deja.com/getdoc.xp?AN=718062036
Sent via Deja.com http://www.deja.com/
---------------end included mss--------------------
This is about the most dishonest bit of unmarked snipage that I have
ever seen. Leave the rest of my comments in and respond to them.
You seem to fitting in well with your Discovery Institute buddies.
Don Cates
ca...@cc.umanitoba.ca