General fitness, health and nutrition · Public discussion

Multi Levels Of Selection

John Edser · 2003-12-30T12:21:56+00:00

JE:- This is because selection can only at ONCE: at the FIRST ADDITIVE FITNESS INTERFACE that is supposed within the view. GH:- As you know, I have never agreed with your view on this; but I admit that your language still leaves me with questions about your meaning. Can you briefly and clearly describe what you mean by "selection can only at ONCE". JE:- In any supposed multilevel of units of selection where each level provides a testable fitness, selection can only act at the first additive level. This is equivalent to the first absolute fitness intersection that exists within one evolving population. GH:- This doesn't help me. It is illogical, or I don't grasp your meaning, or both. JE:- OK, lets take it a step at a time. Do you agree that in any supposed multilevel of fitnesses: 1) Fitness associations between a lower level and the next level up can be additive OR non additive. 2) Only additive fitness level associations are independently selected. Here a lower level is independently selectable to any supposed additive level in fitness above itself within a supposed fitness multilevel. 3) Non additive fitness levels are dependently selected. Here a level below is selected _with_ the level above itself within a supposed fitness multilevel. 4) Only _one_ level can constitute the _first_ additive i.e. independent fitness level within any supposed multi level of fitnesses. GH:- If a view includes multiple "additive fitness interfaces", how can you decide which is first? Is it the interface of the smallest particles? Is it the interface that is thought to have manifested first in the history of life? JE:- Logically, within any supposed fitness multi level, one level and just one level only, can be FIRSTLY additive in fitness to the next level above itself. This is first place that selection can act. GH:- This is a repetition of the claim that I asked you to explain. JE:- How can MORE than one level within any supposed multi level of fitnesses validly constitute the FIRST additive fitness interface? JE:- Selection here, determines all subsequent selective acts, i.e. group selection cannot win against individual selection (the first additive in fitness interface within any multilevel in nature) because selection has already acted at a level below _any_ additive composite. GH:- You still have not answered my question directly, but you seem to be implying that "first" means smallest spatial scale in your argument. Is this correct? I cannot fairly judge whether your argument is logical unless your meaning is clear to me on this point. JE:- Since only one fitness level within any supposed multi level can constitute the first additive fitness interface between any two levels (one lower level and the next higher one) and only additive fitnesses allows the independent selection of levels, then only the level that is the lower level at the 1st additive fitness interface, that exists between levels within the supposed multi level can be _firstly_, selected. If like McGinn, you suppose all levels to be simultaneously selected, then the observation of just one multilevel that is not simultaneously selected in nature refutes such a view. Clearly ordinary Darwinian selection refutes it. My prediction is, if you try to select only the higher additive fitness level and just ignore the lower additive fitness level then the higher level must contest the lower just like classical group selection contests selection at the lower Darwinian organism additive fitness level. The logic states that since selection has _already_ operated at the lower organism level firstly, i.e. before it can even start to operate at any higher additive fitness composite level, then selection at a higher level cannot contest and win over selection at the next, lower level. Any enforced selection via artificially selecting at just the higher group level can send an entire population into extinction if it continually lowers fitness at the next, lower, organism level, within the supposed multilevel. In this instance, group level selection has not won over selection at the organism level of selection. The lower organism level of selection has removed this act of group selection be removing all individuals that attempted it, i.e. classical group selection cannot win over Darwinian organism selection if the interface between them is fitness additive. Thus group selection is naturally selected to go with and not against Darwinian organism selection when just an additive fitness interface exists between them. If a non additive fitness interface exists between them then group selection must win, every time, over Darwinian organism selection. JE:- Group selection is predicted never to win against selection at the Darwinian fertile organism level because selection has acted at the Darwinian level before selection at any additive in fitness level above it, can even start. This prediction has been confirmed bot in nature and within simplified models. GH:- As you know, I disagree with you on all counts of this point. JE:- OK, that is fine but I see no contesting logic and no contesting observations being posted. The status quo seems to think it does not have to bother. GH:- While I suspect your logic is flawed, I am trying once again to give you the benefit of the doubt created by my uncertainty regarding your meaning. This is why I am asking for clarification of key words you are using in your argument. Aside from ambiguities in your argument, I do not think that your prediction has ever "been confirmed both in nature and within simplified models." Can you give me at least one example in each category? References would also help. JE:- How many years has group selection been debated? How many more years will such nonsense go on before people understand that selection at any additive fitness composite level cannot contest and win against selection at the additive level below itself for simple, logical reasons. In nature the first additive fitness interface exists between the Darwinian organism level and additive groups of the same units. Here, selection at the Darwinian level always wins. Can you provide any counter examples? You seem to agree that Dawkins was "wrong" to suppose that selection at the gene level can contest and win against selection at Darwinian organism level, every time. Why was Dawkins wrong to suppose this? He was wrong because not a single additive i.e. independent gene fitness has ever been documented within nature, i.e. Dawkins misused an over simplified model that supposed that genes can be independently selected when independent genomic gene fitnesses do not even exist within nature. The model Dawkins misused in this way was Hamilton's. Hamilton deleted all epistatic fitness within his model. In nature, all gene fitnesses are known to be epistatic to one genomic fitness, so Hamilton deleted all known gene fitnesses. Hamilton derived his model from Fisher who very crudely attempted to join Mendel's genes with Darwin's theory. Fisher et al Herculean, mathematical efforts were worthwhile, if and only if, they were not misused. Hamilton misused them and so has everybody else, subsequently. By misuse I mean the model was allowed to contest and win against the theory is was just a simplification from when any such event is just a logical absurdity. Whenever group selection has been used no control over individual selection has been in place i.e. results attributed to group selection could be validly attributed to either. Experimentally, all group selection _appears_ to be able to do is drive an entire group to extinction. Once again, this is only because group selection is losing out to selection at the individual level i.e. the extinction was caused by selection at the organism level removing all group selected individuals. Yet, the common misinterpretation is that just group selection caused the extinction. If group selection appears to go with and enhance selection at the organism lower level then, once again, this is just attributed to group selection when it can be validly attributed to either. For classical group selection to win over the next, lower, additive, level it must be able to demonstrate that individual organism fitnesses can be lowered while additive population units can be observed to keep on expanding in number, i.e. avoid the common observation that supposed group selection appears to just send any population into extinction. [snip] GH:- If there are multiple "additive fitness interfaces", what prevents selection from happening at "interfaces" other than the first one? JE:- Selection is determined at only the first additive fitness interface. Selection at all the other additive interfaces above the first interface cannot win against selection at the first additive interface so here, it is just a sub process of the first selective act, natural selection and cannot win against natural selection. GH:- You did not even begin to answer my question. JE:- The first _independent_ fitness interface that can logically exist is the first additive fitness interface that exists within the multilevel. The first additive fitness interface that exists in nature, exists between the Darwinian organism level and additive groups of same (classical group selection). It is here and only here that selection acts, in dependently and firstly. It is here that Darwin founded his theory and is here that it has been tested for over 150 years. GH:- Dawkins certainly ran with the idea that selection at the level of the gene can potentially explain much of what we see in individual organisms, and I think he took that viewpoint further than Hamilton would ever have done himself. In my estimation, Hamilton had a more balanced and rational view than does Dawkins. This is not to say that I agree with Hamilton on many issues, and I have often indicated in this newsgroup that I think Hamilton's kin selection model (and host/parasite model for that matter) have had unduly strong influences on the field as a whole. I nevertheless had great respect for Hamilton's pioneering ideas, which I think represented important steps forward for evolutionary theory. JE:- Hamilton deleted at least two constants. When they are included rb can never be > c. Do you want me to repost the mathematics that proves this? GH:- There is no need to repost. IMHO, your math failed to prove your point then and that will not change unless you have changed your math. JE:- Everybody here, except a mathematician who was outside the evolutionary theory establishment failed to admit that it is entirely illegal to suppose rb>c if you fail to FIRSTLY say how may rb and c units = one comparable unit of fitness! GH:- In my opinion you have failed to understand my response, and the responses of many others, to your position on this issue. This is why I indicated that I do not want to revisit your math. I am certain that you explained your own position to your mathematician friend without adequately conveying the contrary responses of so many on sbe. If the mathematician was given an opportunity to hear our arguments, I am sure that he/she would agree with us. Can we please stick to the meanings of the words you use in your basic argument for now? JE:- OK I will repost the entire argument. However, firstly, do you agree that mathematically, you cannot validly compare rb with c in any way unless you explicitly suggest how many units of rb = how many units of c no matter what they stand for within biology? JE:- [snip] Why do you want it both ways? You admit that Dawkins' was wrong but you refuse to admit that he was wrong because Hamilton was wrong. GH:- I claim that Dawkins is wrong because he goes too far in attributing too much to selection at the level of the gene, and I claim that Hamilton never made this mistake. I need not "admit" anything relating to this issue as if I had recently promoted flawed arguments with intent to mislead. Your use of rhetoric to impugn me or my arguments is offensive. I don't think I have ever asked you to "admit" anything in our exchanges, although I have hoped to convince you that there have been errors in your arguments. JE:- Saying that "Dawkins is wrong because he goes too far in attributing too much to selection at the level of the gene" is just a subjective impression on your part, i.e. you have contributed nothing _objective_ to this part of the discussion. Do you deny that Dawkins was simply applying Hamiltonian logic? Do you deny that independent genomic gene selection depends entirely on independent genomic gene fitnesses actually existing in nature? Where are they? Not a single, independent genomic gene fitness has ever been observed within nature. This is not "rhetoric" just the facts of the matter, which it appears, everybody here finds unpalatable. Please say why, and by what testable mechanism you can validly suggest, that Dawkins was "wrong"? It appears to be you and not me who is employing "rhetoric". I provided a detailed, objective argument, which you were free to prove wrong, if you can, but you supplied nothing I can test to refutation. JE:- All supposed selfish geneism causing Darwinian organism fitness altruism is based on Hamilton. Dawkins, Wilson etc are just prominent minnows that depend entirely on Hamilton and his entirely _misused_ hypothetical, rule. GH:- OK. The difference in our views is that I do not agree it is wrong to consider how selection at the level of the gene can influence evolution at the level of the whole organism. JE:- What limits do you place on such an "influence" ? GH:- This is what Hamilton did, which certainly was novel at the time, and IMHO it was a very clever way for him to solve a very difficult and prominent problem in evolutionary theory. JE:- He never solved it, he just appeared to solve it. Nobody, himself included, ever asked the simple question: can rb ever be > c within nature? This question was never asked. Why? GH:- Having illustrated the power and IMHO validity of this way of thinking, Dawkins and many others have become so engrossed in this way of thinking that they can't see how it fits into the whole picture. JE:- No, Dawkins just pushed the idea to its obvious conclusion: IF genomic genes can be independently selected THEN organisms will _never_ be able to contest independent gene fitnesses and win. Dawkins was just being logically self consistent. This was to his credit because it allowed the view to be tested. Of course, not a single, independent, genomic gene fitness has ever been observed within nature, i.e. the whole thing was just a fairy story, i.e. a misused oversimplification of just a supposition of independent genomic gene fitnesses existing within nature. This over simplification was produced by Fisher when he suggested that only _additive_ associations between genomic genes are "heritable" and thus selectable, even if non additive associations were actually "inherited". Nobody wants to deal with the incorrect application of Fisher's over simplification of what is "heritable and thus "selectable" because it opens a can or worms. However, the tiny size of the human genome will force everybody to do so. Of course this, preferably, will be much later rather than sooner, to try to save face. In the meantime, millions of taxpayers dollars will be wasted. GH:- I see D.S. Wilson as a different story altogether. Rather than being locked into a very narrow view of reality,I think Wilson has one of the more open minds every among theoretical evolutionary biologists. JE:- yes, so "open minded" nothing he has suggested, so far, is testable. GH:- So, in sum, I see no problem the consistency of my arguments against Dawkins' geneism-only views and Hamilton's or Wilson's view that selection can happen at the level of the gene. JE:- How can you separate selection acting at a different level within a supposed multilevel, by experiment, i.e. how do you _control_ for selection acting at other levels within a multilevel? If selection can validly act at the gene level when not a single, independent gene fitness has ever been observed within nature, what can ever _restrict_ selection happening here?!? Is selection on one multi level unrestricted? Are even the numbers of different levels within one multi level restricted , even within just over simplified models? Regards, John Edser Independent Researcher PO Box 266 Church Pt NSW 2105 Australia [email hidden]

Started by John Edser · 30 December 2003 · Last activity 30 December 2003 · 6 posts · 623 views

← Back to General fitness, health and nutrition

Thread navigation

Jump through the discussion

Go to the original post, the replies on this page, or the latest preserved contribution.

01 / Original postMulti Levels Of SelectionJohn Edser · 30 Dec 2003 02 / This pagePosts 1–6Page 1 of 1 · chronological order 03 / Latest replyPost #6Malcolm · 30 Dec 2003

Thread details

What we know about this thread

Original section: General fitness, health and nutrition
Published: 30 December 2003
Last activity: 30 December 2003
Original author: John Edser
Posts: 6
Discussion status: Public discussion
Total views: 623
Views / 30 days: 0

The navigation and discussion metadata provide context. Posts remain in their original chronological order.

Original post
John Edser
Archived post

30 December 2003, 12:21 UTC #1

Quoted message said:
Quoted message said:
JE:- This is because selection can only at ONCE: at the FIRST ADDITIVE FITNESS INTERFACE that is
supposed within the view.
Quoted message said:
GH:- As you know, I have never agreed with your view on this; but I admit that your language still
leaves me with questions about your meaning. Can you briefly and clearly describe what you mean by
"selection can only at

ONCE".
Quoted message said:
JE:- In any supposed multilevel of units of selection where each level provides a testable
fitness, selection can only act at the first additive level. This is equivalent to the first
absolute fitness intersection that exists within one evolving population.
GH:- This doesn't help me. It is illogical, or I don't grasp your meaning, or both.
JE:- OK, lets take it a step at a time.
Do you agree that in any supposed multilevel of fitnesses:
1) Fitness associations between a lower level and the next level up can be additive OR non additive.
2) Only additive fitness level associations are independently selected. Here a lower level is
independently selectable to any supposed additive level in fitness above itself within a
supposed fitness multilevel.
3) Non additive fitness levels are dependently selected. Here a level below is selected _with_ the
level above itself within a supposed fitness multilevel.
4) Only _one_ level can constitute the _first_ additive i.e. independent fitness level within any
supposed multi level of fitnesses.
Quoted message said:
GH:- If a view includes multiple "additive fitness interfaces", how can you decide which is first?
Is it the interface of the smallest particles? Is it the interface that is thought to have
manifested first in the history

of
Quoted message said:
life?
Quoted message said:
JE:- Logically, within any supposed fitness multi level, one level and just one level only, can
be FIRSTLY additive in fitness to the next level above itself. This is first place that selection
can act.
GH:- This is a repetition of the claim that I asked you to explain.
JE:- How can MORE than one level within any supposed multi level of fitnesses validly constitute the
FIRST additive fitness interface?
Quoted message said:
JE:- Selection here, determines all subsequent selective acts,
i.e. group selection cannot win against individual selection (the first additive in fitness
interface within any multilevel in nature) because selection has already acted at a level
below _any_ additive composite.
GH:- You still have not answered my question directly, but you seem to be implying that "first"
means smallest spatial scale in your argument. Is this correct? I cannot fairly judge whether your
argument is logical unless your meaning is clear to me on this point.
JE:- Since only one fitness level within any supposed multi level can constitute the first additive
fitness interface between any two levels (one lower level and the next higher one) and only additive
fitnesses allows the independent selection of levels, then only the level that is the lower level at
the 1st additive fitness interface, that exists between levels within the supposed multi level can
be _firstly_, selected.
If like McGinn, you suppose all levels to be simultaneously selected, then the observation of just
one multilevel that is not simultaneously selected in nature refutes such a view. Clearly ordinary
Darwinian selection refutes it.
My prediction is, if you try to select only the higher additive fitness level and just ignore the
lower additive fitness level then the higher level must contest the lower just like classical group
selection contests selection at the lower Darwinian organism additive fitness level. The logic
states that since selection has _already_ operated at the lower organism level firstly, i.e. before
it can even start to operate at any higher additive fitness composite level, then selection at a
higher level cannot contest and win over selection at the next, lower level. Any enforced selection
via artificially selecting at just the higher group level can send an entire population into
extinction if it continually lowers fitness at the next, lower, organism level, within the supposed
multilevel. In this instance, group level selection has not won over selection at the organism level
of selection. The lower organism level of selection has removed this act of group selection be
removing all individuals that attempted it, i.e. classical group selection cannot win over Darwinian
organism selection if the interface between them is fitness additive. Thus group selection is
naturally selected to go with and not against Darwinian organism selection when just an additive
fitness interface exists between them. If a non additive fitness interface exists between them then
group selection must win, every time, over Darwinian organism selection.
Quoted message said:
JE:- Group selection is predicted never to win against selection at the Darwinian fertile organism
level because selection has acted at the Darwinian level before selection at any additive in
fitness level above it, can even start. This prediction has been confirmed bot in nature and
within simplified models.
GH:- As you know, I disagree with you on all counts of this point.
JE:- OK, that is fine but I see no contesting logic and no contesting observations being posted. The
status quo seems to think it does not have to bother.
GH:- While I suspect your logic is flawed, I am trying once again to give you the benefit of the
doubt created by my uncertainty regarding your meaning. This is why I am asking for clarification of
key words you are using in your argument. Aside from ambiguities in your argument, I do not think
that your prediction has ever "been confirmed both in nature and within simplified models." Can you
give me at least one example in each category? References would also help.
JE:- How many years has group selection been debated? How many more years will such nonsense go on
before people understand that selection at any additive fitness composite level cannot contest and
win against selection at the additive level below itself for simple, logical reasons. In nature
the first additive fitness interface exists between the Darwinian organism level and additive
groups of the same units. Here, selection at the Darwinian level always wins. Can you provide any
counter examples?
You seem to agree that Dawkins was "wrong" to suppose that selection at the gene level can contest
and win against selection at Darwinian organism level, every time. Why was Dawkins wrong to suppose
this? He was wrong because not a single additive
i.e. independent gene fitness has ever been documented within nature, i.e. Dawkins misused an over
simplified model that supposed that genes can be independently selected when independent
genomic gene fitnesses do not even exist within nature. The model Dawkins misused in this way
was Hamilton's. Hamilton deleted all epistatic fitness within his model. In nature, all gene
fitnesses are known to be epistatic to one genomic fitness, so Hamilton deleted all known gene
fitnesses. Hamilton derived his model from Fisher who very crudely attempted to join Mendel's
genes with Darwin's theory. Fisher et al Herculean, mathematical efforts were worthwhile, if
and only if, they were not misused. Hamilton misused them and so has everybody else,
subsequently. By misuse I mean the model was allowed to contest and win against the theory is
was just a simplification from when any such event is just a logical absurdity.
Whenever group selection has been used no control over individual selection has been in place i.e.
results attributed to group selection could be validly attributed to either. Experimentally, all
group selection _appears_ to be able to do is drive an entire group to extinction. Once again, this
is only because group selection is losing out to selection at the individual level i.e. the
extinction was caused by selection at the organism level removing all group selected individuals.
Yet, the common misinterpretation is that just group selection caused the extinction.
If group selection appears to go with and enhance selection at the organism lower level then, once
again, this is just attributed to group selection when it can be validly attributed to either. For
classical group selection to win over the next, lower, additive, level it must be able to
demonstrate that individual organism fitnesses can be lowered while additive population units can be
observed to keep on expanding in number, i.e. avoid the common observation that supposed group
selection appears to just send any population into extinction.
[snip]
Quoted message said:
GH:- If there are multiple "additive fitness interfaces", what prevents

selection
Quoted message said:
from happening at "interfaces" other than the first one?
Quoted message said:
JE:- Selection is determined at only the first additive fitness interface. Selection at all the
other additive interfaces above the first interface cannot win against selection at the first
additive interface so here, it is just a sub process of the first selective act,
Quoted message said:
natural selection and cannot win against natural selection.
GH:- You did not even begin to answer my question.
JE:- The first _independent_ fitness interface that can logically exist is the first additive
fitness interface that exists within the multilevel. The first additive fitness interface that
exists in nature, exists between the Darwinian organism level and additive groups of same (classical
group selection). It is here and only here that selection acts, in dependently and firstly. It is
here that Darwin founded his theory and is here that it has been tested for over 150 years.
Quoted message said:
Quoted message said:
GH:- Dawkins certainly ran with the idea that selection at the level of

the
Quoted message said:
Quoted message said:
gene can potentially explain much of what we see in individual organisms,

and
Quoted message said:
Quoted message said:
I think he took that viewpoint further than Hamilton would ever have done himself. In my
estimation, Hamilton had a more balanced and rational

view
Quoted message said:
Quoted message said:
than does Dawkins. This is not to say that I agree with Hamilton on many issues, and I have often
indicated in this newsgroup that I think

Hamilton's
Quoted message said:
Quoted message said:
kin selection model (and host/parasite model for that matter) have had

unduly
Quoted message said:
Quoted message said:
strong influences on the field as a whole. I nevertheless had great

respect
Quoted message said:
Quoted message said:
for Hamilton's pioneering ideas, which I think represented important

steps
Quoted message said:
Quoted message said:
forward for evolutionary theory.
Quoted message said:
Quoted message said:
JE:- Hamilton deleted at least two constants. When they are included rb can never be > c. Do you
want me to repost the mathematics that proves this?
Quoted message said:
GH:- There is no need to repost. IMHO, your math failed to prove your point

then
Quoted message said:
and that will not change unless you have changed your math.
Quoted message said:
JE:- Everybody here, except a mathematician who was outside the evolutionary theory establishment
failed to admit that it is entirely illegal to suppose
rb>c if you fail to FIRSTLY say
how may rb and c units = one comparable unit of fitness!
GH:- In my opinion you have failed to understand my response, and the responses of many others, to
your position on this issue. This is why I indicated that I do not want to revisit your math. I am
certain that you explained your own position to your mathematician friend without adequately
conveying the contrary responses of so many on sbe. If the mathematician was given an opportunity to
hear our arguments, I am sure that he/she would agree with us. Can we please stick to the meanings
of the words you use in your basic argument for now?
JE:- OK I will repost the entire argument.
However, firstly, do you agree that mathematically, you cannot validly compare rb with c in any way
unless you explicitly suggest how many units of rb = how many units of c no matter what they stand
for within biology?
Quoted message said:
JE:- [snip] Why do you want it both ways? You admit that Dawkins' was wrong but you refuse to
admit that he was wrong because Hamilton was wrong.
GH:- I claim that Dawkins is wrong because he goes too far in attributing too much to selection at
the level of the gene, and I claim that Hamilton never made this mistake. I need not "admit"
anything relating to this issue as if I had recently promoted flawed arguments with intent to
mislead. Your use of rhetoric to impugn me or my arguments is offensive. I don't think I have ever
asked you to "admit" anything in our exchanges, although I have hoped to convince you that there
have been errors in your arguments.
JE:- Saying that "Dawkins is wrong because he goes too far in attributing too much to selection at
the level of the gene" is just a subjective impression on your part, i.e. you have contributed
nothing _objective_ to this part of the discussion. Do you deny that Dawkins was simply applying
Hamiltonian logic? Do you deny that independent genomic gene selection depends entirely on
independent genomic gene fitnesses actually existing in nature? Where are they? Not a single,
independent genomic gene fitness has ever been observed within nature. This is not "rhetoric" just
the facts of the matter, which it appears, everybody here finds unpalatable.
Please say why, and by what testable mechanism you can validly suggest, that Dawkins was "wrong"? It
appears to be you and not me who is employing "rhetoric". I provided a detailed, objective argument,
which you were free to prove wrong, if you can, but you supplied nothing I can test to refutation.
Quoted message said:
JE:- All supposed selfish geneism causing Darwinian organism fitness altruism is based on
Hamilton. Dawkins, Wilson etc are just prominent minnows that depend entirely on Hamilton and his
entirely _misused_ hypothetical, rule.
GH:- OK. The difference in our views is that I do not agree it is wrong to consider how selection at
the level of the gene can influence evolution at the level of the whole organism.
JE:- What limits do you place on such an "influence" ?
GH:- This is what Hamilton did, which certainly was novel at the time, and IMHO it was a very clever
way for him to solve a very difficult and prominent problem in evolutionary theory.
JE:- He never solved it, he just appeared to solve it. Nobody, himself included, ever asked the
simple question: can rb ever be > c within nature? This question was never asked. Why?
GH:- Having illustrated the power and IMHO validity of this way of thinking, Dawkins and many
others have become so engrossed in this way of thinking that they can't see how it fits into the
whole picture.
JE:- No, Dawkins just pushed the idea to its obvious conclusion: IF genomic genes can be
independently selected THEN organisms will _never_ be able to contest independent gene fitnesses and
win. Dawkins was just being logically self consistent. This was to his credit because it allowed the
view to be tested.
Of course, not a single, independent, genomic gene fitness has ever been observed within nature,
i.e. the whole thing was just a fairy story, i.e. a misused oversimplification of just a supposition
of independent genomic gene fitnesses existing within nature. This over simplification was produced
by Fisher when he suggested that only _additive_ associations between genomic genes are "heritable"
and thus selectable, even if non additive associations were actually "inherited".
Nobody wants to deal with the incorrect application of Fisher's over simplification of what is
"heritable and thus "selectable" because it opens a can or worms. However, the tiny size of the
human genome will force everybody to do so. Of course this, preferably, will be much later rather
than sooner, to try to save face. In the meantime, millions of taxpayers dollars will be wasted.
GH:- I see D.S. Wilson as a different story altogether. Rather than being locked into a very narrow
view of reality,I think Wilson has one of the more open minds every among theoretical evolutionary
biologists.
JE:- yes, so "open minded" nothing he has suggested, so far, is testable.
GH:- So, in sum, I see no problem the consistency of my arguments against Dawkins' geneism-only
views and Hamilton's or Wilson's view that selection can happen at the level of the gene.
JE:- How can you separate selection acting at a different level within a supposed multilevel, by
experiment, i.e. how do you _control_ for selection acting at other levels within a multilevel? If
selection can validly act at the gene level when not a single, independent gene fitness has ever
been observed within nature, what can ever _restrict_ selection happening here?!?
Is selection on one multi level unrestricted? Are even the numbers of different levels within one
multi level restricted , even within just over simplified models?
Regards,
John Edser Independent Researcher
PO Box 266 Church Pt NSW 2105 Australia
[email hidden]

Report
Name And Addres
Archived post

30 December 2003, 12:23 UTC #2

<snip>
This looks like an interesting thread, but i haven't been able to follow any of it, because of the
confusing way in which the previous messages have been quoted. JE, could you start using '>'s to
indicate previous text, adding an extra '>' at each stage, so as to give an indication of the
temporal unfolding of this discussion?
Also, since this is a new thread, could you try giving a quick summary of what is being debated?
Cheers.

Report
Jim McGinn
Archived post

30 December 2003, 12:23 UTC #3

"John Edser" <[email hidden]> wrote in message news:<[email hidden]>...
Quoted message said:
Quoted message said:
Quoted message said:
JE:- This is because selection can only at ONCE: at the FIRST ADDITIVE FITNESS INTERFACE that
is supposed within the view.
I can't make any sense of this.
Quoted message said:

Quoted message said:
GH:- As you know, I have never agreed with your view on this; but I admit that your language
still leaves me with questions about your meaning. Can you briefly and clearly describe what you
mean by "selection can only at

ONCE".
Quoted message said:
JE:- In any supposed multilevel of units of selection where each level provides a testable
fitness, selection can only act at the first additive level. This is equivalent to the first
absolute fitness intersection that exists within one evolving population.
GH:- This doesn't help me. It is illogical, or I don't grasp your meaning, or both.
JE:- OK, lets take it a step at a time.
Do you agree that in any supposed multilevel of fitnesses:
1) Fitness associations between a lower level and the next level up can be additive OR non
additive.
As Guy said above: This doesn't help me. It is illogical, or I don't grasp your meaning, or both.
Quoted message said:

2) Only additive fitness level associations are independently selected. Here a lower level is
independently selectable to any supposed additive level in fitness above itself within a
supposed fitness multilevel.
As Guy said above: This doesn't help me. It is illogical, or I don't grasp your meaning, or both.
Quoted message said:
3) Non additive fitness levels are dependently selected. Here a level below is selected _with_ the
level above itself within a supposed fitness multilevel.
As Guy said above: This doesn't help me. It is illogical, or I don't grasp your meaning, or both.
Quoted message said:
4) Only _one_ level can constitute the _first_ additive i.e. independent fitness level within any
supposed multi level of fitnesses.
As Guy said above: This doesn't help me. It is illogical, or I don't grasp your meaning, or both.
Quoted message said:
Quoted message said:
GH:- If a view includes multiple "additive fitness interfaces", how can you decide which is
first? Is it the interface of the smallest particles? Is it the interface that is thought to
have manifested first in the history

of
Quoted message said:
life?
Quoted message said:
JE:- Logically, within any supposed fitness multi level, one level and just one level only, can
be FIRSTLY additive in fitness to the next level above itself. This is first place that
selection can act.
As Guy said above: This doesn't help me. It is illogical, or I don't grasp your meaning, or both.
Quoted message said:
GH:- This is a repetition of the claim that I asked you to explain.
Without circular logic Edser would have no logic at all.
Quoted message said:
JE:- How can MORE than one level within any supposed multi level of fitnesses validly constitute
the FIRST additive fitness interface?
As Guy said above: This doesn't help me. It is illogical, or I don't grasp your meaning, or both.
Quoted message said:
Quoted message said:
JE:- Selection here, determines all subsequent selective acts,
i.e. group selection cannot win against individual selection (the first additive in fitness
interface within any multilevel in nature) because selection has already acted at a level
below _any_ additive composite.
GH:- You still have not answered my question directly, but you seem to be implying that "first"
means smallest spatial scale in your argument. Is this correct? I cannot fairly judge whether your
argument is logical unless your meaning is clear to me on this point.
Good luck ever getting a straight answer to this question.
Quoted message said:
JE:- Since only one fitness level within any supposed multi level can constitute the first
additive fitness interface between any two levels (one lower level and the next higher one) and
only additive fitnesses allows the independent selection of levels, then only the level that is
the lower level at the 1st additive fitness interface, that exists between levels within the
supposed multi level can be _firstly_, selected.
If like McGinn, you suppose all levels to be simultaneously selected, then the observation of just
one multilevel that is not simultaneously selected in nature refutes such a view. Clearly ordinary
Darwinian selection refutes it.
My prediction is, if you try to select only the higher additive fitness level and just ignore the
lower additive fitness level then the higher level must contest the lower just like classical
group selection contests selection at the lower Darwinian organism additive fitness level. The
logic states that since selection has _already_ operated at the lower organism level firstly, i.e.
before it can even start to operate at any higher additive fitness composite level, then selection
at a higher level cannot contest and win over selection at the next, lower level. Any enforced
selection via artificially selecting at just the higher group level can send an entire population
into extinction if it continually lowers fitness at the next, lower, organism level, within the
supposed multilevel. In this instance, group level selection has not won over selection at the
organism level of selection. The lower organism level of selection has removed this act of group
selection be removing all individuals that attempted it, i.e. classical group selection cannot win
over Darwinian organism selection if the interface between them is fitness additive. Thus group
selection is naturally selected to go with and not against Darwinian organism selection when just
an additive fitness interface exists between them. If a non additive fitness interface exists
between them then group selection must win, every time, over Darwinian organism selection.
Without circular logic Edser would have no logic at all.
Quoted message said:
Quoted message said:
JE:- Group selection is predicted never to win against selection at the Darwinian fertile
organism level because selection has acted at the Darwinian level before selection at any
additive in fitness level above it, can even start. This prediction has been confirmed bot in
nature and within simplified models.
GH:- As you know, I disagree with you on all counts of this point.
JE:- OK, that is fine but I see no contesting logic and no contesting observations being posted.
The status quo seems to think it does not have to bother.
I don't think Guy is arguing from the standpoint of status quo (at least not in this instance). I
think he's just trying to understand the epistemological origin of the absolute truths (supposed
principles) that you seem to grasp out of thin air and then support with "sky hook" (Dennet)
reasoning.
Quoted message said:

GH:- While I suspect your logic is flawed, I am trying once again to give you the benefit of the
doubt created by my uncertainty regarding your meaning. This is why I am asking for clarification
of key words you are using in your argument. Aside from ambiguities in your argument, I do not
think that your prediction has ever "been confirmed both in nature and within simplified models."
Can you give me at least one example in each category? References would also help.
JE:- How many years has group selection been debated? How many more years will such nonsense go on
before people understand that selection at any additive fitness composite level cannot contest and
win against selection at the additive level below itself for simple, logical reasons.
The reasons you've indicated may be simple but they are not logical.
In nature the
Quoted message said:
first additive fitness interface exists between the Darwinian organism level and additive groups
of the same units. Here, selection at the Darwinian level always wins. Can you provide any counter
examples?
Counter to what?
Quoted message said:

You seem to agree that Dawkins was "wrong" to suppose that selection at the gene level can contest
and win against selection at Darwinian organism level, every time. Why was Dawkins wrong to
suppose this?
Because selection is not a level specific phenomenon. (Or is it phenomena?)
He was wrong because not a single additive
Quoted message said:
i.e. independent gene fitness has ever been documented within nature, i.e. Dawkins misused an over
simplified model that supposed that genes can be independently selected when independent
genomic gene fitnesses do not even exist within nature. The model Dawkins misused in this way
was Hamilton's. Hamilton deleted all epistatic fitness within his model.
John, you've mixed and matched terminology from so many different approaches that your thinking is
incomprehensible. In fact I suspect that not even you yourself understand it like you think you do.
In nature, all gene fitnesses
Quoted message said:
are known to be epistatic to one genomic fitness, so Hamilton deleted all known gene fitnesses.
Hamilton derived his model from Fisher who very crudely attempted to join Mendel's genes with
Darwin's theory. Fisher et al Herculean, mathematical efforts were worthwhile, if and only if,
they were not misused. Hamilton misused them and so has everybody else, subsequently. By misuse I
mean the model was allowed to contest and win against the theory is was just a simplification from
when any such event is just a logical absurdity.
Whenever group selection has been used no control over individual selection has been in place i.e.
results attributed to group selection could be validly attributed to either. Experimentally, all
group selection _appears_ to be able to do is drive an entire group to extinction. Once again,
this is only because group selection is losing out to selection at the individual level i.e. the
extinction was caused by selection at the organism level removing all group selected individuals.
Yet, the common misinterpretation is that just group selection caused the extinction.
I can't make any sense of this.
Quoted message said:

If group selection appears to go with and enhance selection at the organism lower level then, once
again, this is just attributed to group selection when it can be validly attributed to either. For
classical group selection to win over the next, lower, additive, level it must be able to
demonstrate that individual organism fitnesses can be lowered while additive population units can
be observed to keep on expanding in number, i.e. avoid the common observation that supposed group
selection appears to just send any population into extinction.
In this passage you seem to have assumed the existence of nonexistent principles. I don't know how
you expect anybody to make sense of this.
Quoted message said:

[snip]
Quoted message said:
GH:- If there are multiple "additive fitness interfaces", what prevents

selection
Quoted message said:
from happening at "interfaces" other than the first one?
Quoted message said:
JE:- Selection is determined at only the first additive fitness interface. Selection at all the
other additive interfaces above the first interface cannot win against selection at the first
additive interface so here, it is just a sub process of the first selective act,
Quoted message said:
Quoted message said:
natural selection and cannot win against natural selection.
GH:- You did not even begin to answer my question.
He never will.
Quoted message said:

JE:- The first _independent_ fitness interface that can logically exist is the first additive
fitness interface that exists within the multilevel. The first additive fitness interface that
exists in nature, exists between the Darwinian organism level and additive groups of same
(classical group selection). It is here and only here that selection acts, in dependently and
firstly. It is here that Darwin founded his theory and is here that it has been tested for over
150 years.
Quoted message said:
Quoted message said:
GH:- Dawkins certainly ran with the idea that selection at the level of

the
Quoted message said:
Quoted message said:
gene can potentially explain much of what we see in individual organisms,

and
Quoted message said:
Quoted message said:
I think he took that viewpoint further than Hamilton would ever have done himself. In my
estimation, Hamilton had a more balanced and rational

view
Quoted message said:
Quoted message said:
than does Dawkins. This is not to say that I agree with Hamilton on many issues, and I have
often indicated in this newsgroup that I think

Hamilton's
Quoted message said:
Quoted message said:
kin selection model (and host/parasite model for that matter) have had

unduly
Quoted message said:
Quoted message said:
strong influences on the field as a whole. I nevertheless had great

respect
Quoted message said:
Quoted message said:
for Hamilton's pioneering ideas, which I think represented important

steps
Quoted message said:
Quoted message said:
forward for evolutionary theory.
Quoted message said:
Quoted message said:
JE:- Hamilton deleted at least two constants. When they are included rb can never be > c. Do
you want me to repost the mathematics that proves this?
Quoted message said:
GH:- There is no need to repost. IMHO, your math failed to prove your point

then
Quoted message said:
and that will not change unless you have changed your math.
Quoted message said:
JE:- Everybody here, except a mathematician who was outside the evolutionary theory
establishment failed to admit that it is entirely illegal to suppose
rb>c
I haven't failed to admit this. I've always known it was nonsense.
if you fail to FIRSTLY say
Quoted message said:
Quoted message said:
how may rb and c units = one comparable unit of fitness!
GH:- In my opinion you have failed to understand my response, and the responses of many others, to
your position on this issue. This is why I indicated that I do not want to revisit your math. I am
certain that you explained your own position to your mathematician friend without adequately
conveying the contrary responses of so many on sbe. If the mathematician was given an opportunity
to hear our arguments, I am sure that he/she would agree with us. Can we please stick to the
meanings of the words you use in your basic argument for now?
JE:- OK I will repost the entire argument.
However, firstly, do you agree that mathematically, you cannot validly compare rb with c in any
way unless you explicitly suggest how many units of rb = how many units of c no matter what they
stand for within biology?
Quoted message said:
JE:- [snip] Why do you want it both ways? You admit that Dawkins' was wrong but you refuse to
admit that he was wrong because Hamilton was wrong.
GH:- I claim that Dawkins is wrong because he goes too far in attributing too much to selection at
the level of the gene, and I claim that Hamilton never made this mistake.
I don't understand how you can overlook the fact that every test of rb=c has failed. In fact it's
impossible to test rb=c because it maintains assumptions which themselves are impossible to confirm
in reality.
I need not "admit" anything relating to this issue as if
Quoted message said:
I had recently promoted flawed arguments with intent to mislead. Your use of rhetoric to impugn me
or my arguments is offensive. I don't think I have ever asked you to "admit" anything in our
exchanges, although I have hoped to convince you that there have been errors in your arguments.
JE:- Saying that "Dawkins is wrong because he goes too far in attributing too much to selection at
the level of the gene" is just a subjective impression on your part, i.e. you have contributed
nothing _objective_ to this part of the discussion. Do you deny that Dawkins was simply applying
Hamiltonian logic? Do you deny that independent genomic gene selection depends entirely on
independent genomic gene fitnesses actually existing in nature? Where are they? Not a single,
independent genomic gene fitness has ever been observed within nature. This is not "rhetoric" just
the facts of the matter, which it appears, everybody here finds unpalatable.
Please say why, and by what testable mechanism you can validly suggest, that Dawkins was "wrong"?
It appears to be you and not me who is employing "rhetoric". I provided a detailed, objective
argument, which you were free to prove wrong, if you can, but you supplied nothing I can test to
refutation.
Quoted message said:
JE:- All supposed selfish geneism causing Darwinian organism fitness altruism is based on
Hamilton. Dawkins, Wilson etc are just prominent minnows that depend entirely on Hamilton and
his entirely _misused_ hypothetical, rule.
GH:- OK. The difference in our views is that I do not agree it is wrong to consider how selection
at the level of the gene can influence evolution at the level of the whole organism.
JE:- What limits do you place on such an "influence" ?
GH:- This is what Hamilton did, which certainly was novel at the time, and IMHO it was a very
clever way for him to solve a very difficult and prominent problem in evolutionary theory.
If it employs erroneous logic then it's cleverness is illusory.
Quoted message said:

JE:- He never solved it, he just appeared to solve it. Nobody, himself included, ever asked the
simple question: can rb ever be > c within nature? This question was never asked. Why?
GH:- Having illustrated the power and IMHO validity of this way of thinking, Dawkins and many
others have become so engrossed in this way of thinking that they can't see how it fits into the
whole picture.
They've fallen for the illusion of conciseness created by the simplicity of rb=c.
Quoted message said:

JE:- No, Dawkins just pushed the idea to its obvious conclusion: IF genomic genes can be
independently selected THEN organisms will _never_ be able to contest independent gene fitnesses
and win. Dawkins was just being logically self consistent. This was to his credit because it
allowed the view to be tested.
Of course, not a single, independent, genomic gene fitness has ever been observed within nature,
i.e. the whole thing was just a fairy story, i.e. a misused oversimplification of just a
supposition of independent genomic gene fitnesses existing within nature. This over simplification
was produced by Fisher when he suggested that only _additive_ associations between genomic genes
are "heritable" and thus selectable, even if non additive associations were actually "inherited".
Nobody wants to deal with the incorrect application of Fisher's over simplification of what is
"heritable and thus "selectable" because it opens a can or worms. However, the tiny size of the
human genome will force everybody to do so. Of course this, preferably, will be much later rather
than sooner, to try to save face. In the meantime, millions of taxpayers dollars will be wasted.
I don't know about this.
Quoted message said:

GH:- I see D.S. Wilson as a different story altogether. Rather than being locked into a very
narrow view of reality,I think Wilson has one of the more open minds every among theoretical
evolutionary biologists.
He's taken a few steps in the right direction. Not enough steps, however, to break away from the
intellectually stultifying grasp of the current paradigm. If I had to estimate, I'd say he's, maybe,
twenty to fifty years behind myself in this respect.
Quoted message said:

JE:- yes, so "open minded" nothing he has suggested, so far, is testable.
GH:- So, in sum, I see no problem the consistency of my arguments against Dawkins' geneism-only
views and Hamilton's or Wilson's view that selection can happen at the level of the gene.
Selection happens at *all* LEVELS OF BIOLOGICAL PHENOMENA simultaneously. Due to the very real
limitations placed upon us by the requirements of observation and communication we are limited to
discussing selection from the perspective of one UNIT OF SELECTION at a time. (Note how carefully I
distinguished between "levels of biological phenomena" and "units of selection". Most of the
conceptual issues disappear once one is careful to maintain this distinction.) (Also note that LoBP
are real whereas LoS, like all units, are not real.)
Quoted message said:

JE:- How can you separate selection acting at a different level within a supposed multilevel, by
experiment, i.e. how do you _control_ for selection acting at other levels within a multilevel? If
selection can validly act at the gene level when not a single, independent gene fitness has ever
been observed within nature, what can ever _restrict_ selection happening here?!?
A nonsense question.
Quoted message said:

Is selection on one multi level unrestricted? Are even the numbers of different levels within one
multi level restricted , even within just over simplified models?
Again, a nonsense question.

Report
Guy Hoelzer
Archived post

30 December 2003, 12:23 UTC #4

in article [email hidden], John Edser at
[email hidden] wrote on 9/20/03 10:06 PM:
Quoted message said:
Quoted message said:
Quoted message said:
JE:- This is because selection can only at ONCE: at the FIRST ADDITIVE FITNESS INTERFACE that is
supposed within the view.
Quoted message said:
GH:- As you know, I have never agreed with your view on this; but I admit that your language
still leaves me with questions about your meaning. Can you briefly and clearly describe what you
mean by "selection can only at

ONCE".
Quoted message said:
JE:- In any supposed multilevel of units of selection where each level provides a testable
fitness, selection can only act at the first additive level. This is equivalent to the first
absolute fitness intersection that exists within one evolving population.
GH:- This doesn't help me. It is illogical, or I don't grasp your meaning, or both.
JE:- OK, lets take it a step at a time.
Do you agree that in any supposed multilevel of fitnesses:
GH:- I will try to address these points, but note that I do not agree with the notion that there is
a coherent "multilevel" of fitness. As I see it, the process of selection can manifest at any level
of organization when and where the necessary and sufficient conditions for the process appear.
Quoted message said:
1) Fitness associations between a lower level and the next level up can be additive OR non
additive.
I do not see a general reason or a scheme for a structure of interactions between fitnesses at
different levels. Indeed, I see the emergence of higher order levels of organization as something
that dissociates fitnesses at different levels.
Quoted message said:
2) Only additive fitness level associations are independently selected. Here a lower level is
independently selectable to any supposed additive level in fitness above itself within a
supposed fitness multilevel.
I do not see how "fitness associations" can be selected at all, unless they constitute their own
higher order entities that can manifest fitnesses of their own.
Quoted message said:
3) Non additive fitness levels are dependently selected. Here a level below is selected _with_ the
level above itself within a supposed fitness multilevel.
I don't know what this might mean.
Quoted message said:
4) Only _one_ level can constitute the _first_ additive i.e. independent fitness level within any
supposed multi level of fitnesses.
Again, I don't know what this means.
[snip]
Quoted message said:
Quoted message said:
JE:- Selection here, determines all subsequent selective acts,
i.e. group selection cannot win against individual selection (the first additive in fitness
interface within any multilevel in nature) because selection has already acted at a level
below _any_ additive composite.
GH:- You still have not answered my question directly, but you seem to be implying that "first"
means smallest spatial scale in your argument. Is this correct? I cannot fairly judge whether your
argument is logical unless your meaning is clear to me on this point.
JE:- Since only one fitness level within any supposed multi level can constitute the first
additive fitness interface between any two levels (one lower level and the next higher one) and
only additive fitnesses allows the independent selection of levels, then only the level that is
the lower level at the 1st additive fitness interface, that exists between levels within the
supposed multi level can be _firstly_, selected.
GH:- You still have not defined what you mean by level or how you can identify which one is first.
Does it have something to do with size or time?
[snip text that I cannot understand until you clear up these fundamental semantics.]
Quoted message said:
GH:- While I suspect your logic is flawed, I am trying once again to give you the benefit of the
doubt created by my uncertainty regarding your meaning. This is why I am asking for clarification
of key words you are using in your argument. Aside from ambiguities in your argument, I do not
think that your prediction has ever "been confirmed both in nature and within simplified models."
Can you give me at least one example in each category? References would also help.
GH:- Your failure to provide any examples or references seems to support my expectation that
none exist.
Quoted message said:
JE:-

[snip]
Quoted message said:
Whenever group selection has been used no control over individual selection has been in place i.e.
results attributed to group selection could be validly attributed to either. Experimentally, all
group selection _appears_ to be able to do is drive an entire group to extinction. Once again,
this is only because group selection is losing out to selection at the individual level i.e. the
extinction was caused by selection at the organism level removing all group selected individuals.
Yet, the common misinterpretation is that just group selection caused the extinction.
GH:- Group selection cannot cause group extinctions any more than individual selection can cause
individual deaths. Effect goes in the other direction.
Quoted message said:
JE:- If group selection appears to go with and enhance selection at the organism lower level then,
once again, this is just attributed to group selection when it can be validly attributed to
either. For classical group selection to win over the next, lower, additive, level it must be able
to demonstrate that individual organism fitnesses can be lowered while additive population units
can be observed to keep on expanding in number, i.e. avoid the common observation that supposed
group selection appears to just send any population into extinction.
GH:- This isn't the way that top-down effects work, so you cannot logically use this framework as a
way to dismiss them. Instead, they work through mechanisms that cause coordinated activities among
the components in ways that the components themselves would not "choose" in the absence of the top-
down effects. Both "carrots and sticks" are often involved. In the end, the higher order entity
typically nourishes its components in the process.
Quoted message said:
Quoted message said:
GH:- If there are multiple "additive fitness interfaces", what prevents selection from happening
at "interfaces" other than the first one?
Quoted message said:
JE:- Selection is determined at only the first additive fitness interface. Selection at all the
other additive interfaces above the first interface cannot win against selection at the first
additive interface so here, it is just a sub process of the first selective act,
Quoted message said:
Quoted message said:
natural selection and cannot win against natural selection.
GH:- You did not even begin to answer my question.
JE:- The first _independent_ fitness interface that can logically exist is the first additive
fitness interface that exists within the multilevel. The first additive fitness interface that
exists in nature, exists between the Darwinian organism level and additive groups of same
(classical group selection). It is here and only here that selection acts, in dependently and
firstly. It is here that Darwin founded his theory and is here that it has been tested for over
150 years.
GH:- I can tell that you tried, but this still does not begin to answer my question. I did not ask
anything about Darwinism or individuals or groups. Care to try again?
Quoted message said:
Quoted message said:
Quoted message said:
GH:- Dawkins certainly ran with the idea that selection at the level of the gene can potentially
explain much of what we see in individual organisms, and I think he took that viewpoint further
than Hamilton would ever have done himself. In my estimation, Hamilton had a more balanced and
rational view than does Dawkins. This is not to say that I agree with Hamilton on many issues,
and I have often indicated in this newsgroup that I think Hamilton's kin selection model (and
host/parasite model for that matter) have had unduly strong influences on the field as a whole.
I nevertheless had great respect for Hamilton's pioneering ideas, which I think represented
important steps forward for evolutionary theory.
Quoted message said:
Quoted message said:
JE:- Hamilton deleted at least two constants. When they are included rb can never be > c. Do you
want me to repost the mathematics that proves this?
Quoted message said:
GH:- There is no need to repost. IMHO, your math failed to prove your point then and that will
not change unless you have changed your math.
Quoted message said:
JE:- Everybody here, except a mathematician who was outside the evolutionary theory establishment
failed to admit that it is entirely illegal to suppose
rb>c if you fail to FIRSTLY say
how may rb and c units = one comparable unit of fitness!
GH:- In my opinion you have failed to understand my response, and the responses of many others, to
your position on this issue. This is why I indicated that I do not want to revisit your math. I am
certain that you explained your own position to your mathematician friend without adequately
conveying the contrary responses of so many on sbe. If the mathematician was given an opportunity
to hear our arguments, I am sure that he/she would agree with us. Can we please stick to the
meanings of the words you use in your basic argument for now?
JE:- OK I will repost the entire argument.
However, firstly, do you agree that mathematically, you cannot validly compare rb with c in any
way unless you explicitly suggest how many units of rb = how many units of c no matter what they
stand for within biology?
GH:- Asked and answered too many times. One unit of rb = one unit of c. That is what the = sign
means. Before you misuse this point, recall that Hamilton's rule is a threshold, a sort of phase
transition point, not a prediction that equivalence should be expected. If I think that water will
be solid when colder than 0 degrees C, and liquid when warmer than 0 degree C, I could state a rule
that says the state of water changes when T=0 degrees C. My rule would not be refuted by showing an
example of water existing at 20 degrees C. The threshold point for the evolution of altruism is what
Hamilton's rule is about.
Quoted message said:
Quoted message said:
JE:- [snip] Why do you want it both ways? You admit that Dawkins' was wrong but you refuse to
admit that he was wrong because Hamilton was wrong.
GH:- I claim that Dawkins is wrong because he goes too far in attributing too much to selection at
the level of the gene, and I claim that Hamilton never made this mistake. I need not "admit"
anything relating to this issue as if I had recently promoted flawed arguments with intent to
mislead. Your use of rhetoric to impugn me or my arguments is offensive. I don't think I have ever
asked you to "admit" anything in our exchanges, although I have hoped to convince you that there
have been errors in your arguments.
JE:- Saying that "Dawkins is wrong because he goes too far in attributing too much to selection at
the level of the gene" is just a subjective impression on your part, i.e. you have contributed
nothing _objective_ to this part of the discussion. Do you deny that Dawkins was simply applying
Hamiltonian logic?
GH:- Simply - yes.
Quoted message said:
JE:- Do you deny that independent genomic gene selection depends entirely on independent genomic
gene fitnesses actually existing in nature?
GH:- No.
Quoted message said:
JE:- Where are they? Not a single, independent genomic gene fitness has ever been observed within
nature. This is not "rhetoric" just the facts of the matter, which it appears, everybody here finds
unpalatable.
GH:- I and others have repeated provided you with such examples. The fact that you deny this and
continue claiming that such examples don't exist undermines your credibility. Consider again meiotic
drive genes.
Quoted message said:
JE:- Please say why, and by what testable mechanism you can validly suggest, that Dawkins was
"wrong"? It appears to be you and not me who is employing "rhetoric". I provided a detailed,
objective argument, which you were free to prove wrong, if you can, but you supplied nothing I can
test to refutation.
GH:- This is more rehashing. To be brief, Dawkins is wrong because mechanisms causing the
coordination of gene effects evolved along with the origin of complex genomes. Mitosis and meiosis
are the best examples. These reproductive processes generally enforce equivalent fitnesses on gene
copies throughout the genome. Selfish genetic elements can only succeed at the expense of individual
fitness by breaking out of the egalitarian rules of the organism. This is exactly how meiotic drive
genes do it. Dawkins is wrong only to the extent that he argues aggressively selfish genes are the
rule rather than the exception.
Quoted message said:
Quoted message said:
JE:- All supposed selfish geneism causing Darwinian organism fitness altruism is based on
Hamilton. Dawkins, Wilson etc are just prominent minnows that depend entirely on Hamilton and his
entirely _misused_ hypothetical, rule.
GH:- OK. The difference in our views is that I do not agree it is wrong to consider how selection
at the level of the gene can influence evolution at the level of the whole organism.
JE:- What limits do you place on such an "influence" ?
GH:- None. The potential for form and extent of the influence is entirely mechanism dependent. On
the other hand, I think that organisms have evolved highly successful mechanisms constraining the
potential for bottom-up control (biased manipulation) of organismal evolution, so I do not expect to
find instances of strong bottom-up control.
Quoted message said:
GH:- This is what Hamilton did, which certainly was novel at the time, and IMHO it was a very
clever way for him to solve a very difficult and prominent problem in evolutionary theory.
JE:- He never solved it, he just appeared to solve it. Nobody, himself included, ever asked the
simple question: can rb ever be > c within nature? This question was never asked. Why?
GH:- Asked and answered - YES. Can we please try to stick to your model? I know what I think about
Hamilton's views. I am trying to determine whether your alternative is logical.
Quoted message said:
GH:- Having illustrated the power and IMHO validity of this way of thinking, Dawkins and many
others have become so engrossed in this way of thinking that they can't see how it fits into the
whole picture.
JE:- No, Dawkins just pushed the idea to its obvious conclusion: IF genomic genes can be
independently selected THEN organisms will _never_ be able to contest independent gene fitnesses
and win.
GH:- Neither Dawkins nor Hamilton has ever made this argument, and they would have nothing to gain
from keeping it secret. I am still waiting for you to make it, so at the moment I still cannot see
any logic in this point.
[snip]
Quoted message said:
GH:- So, in sum, I see no problem the consistency of my arguments against Dawkins' geneism-only
views and Hamilton's or Wilson's view that selection can happen at the level of the gene.
JE:- How can you separate selection acting at a different level within a supposed multilevel, by
experiment, i.e. how do you _control_ for selection acting at other levels within a multilevel?
GH:- You can easily simulate this situation to illustrate its logic, but you can only try to be as
careful as you can in biological experiments. There can often be evolutionary effects at several
levels from selection acting at a particular level. This must be obvious to anyone who thinks that
selection can only operate at a single level, because he/she must be attributing evolution at other
levels to selection at the "first" level.
Quoted message said:
JE:- If selection can validly act at the gene level when not a single, independent gene fitness
has ever been observed within nature, what can ever _restrict_ selection happening here?!?
GH:- It has been observed many times in nature.
Selection happening here can be restricted through the evolution of mechanisms at a higher level of
organization that result in a correlation among fitnesses within a population. For example, meiosis
and mitosis generally result in each gene having the same probability of replication and passage
into the next organismal generation. The evolution of these mechanisms restricted the potential for
selection at the gene level compared with the open competition and scope for selfish genism that
would have prevailed in earlier times.
Quoted message said:
Is selection on one multi level unrestricted? Are even the numbers of different levels within one
multi level restricted , even within just over simplified models?
I don't know what a multilevel is, so I cannot begin to answer these questions.
Cheers,
Guy

Report
Name And Addres
Archived post

30 December 2003, 12:24 UTC #5

[email hidden] (Jim McGinn) wrote in message news:<[email hidden]>...>
Quoted message said:
Quoted message said:
GH:- I see D.S. Wilson as a different story altogether. Rather than being locked into a very
narrow view of reality,I think Wilson has one of the more open minds every among theoretical
evolutionary biologists.
He's taken a few steps in the right direction. Not enough steps, however, to break away from the
intellectually stultifying grasp of the current paradigm. If I had to estimate, I'd say he's,
maybe, twenty to fifty years behind myself in this respect.
Then why aren't you publishing your insights in recognised journals, Jim? Why confine your
brilliance to brief, confusing, badly referenced posts to a newsgroup?

Report
Malcolm
Archived post

30 December 2003, 12:24 UTC #6

Quoted message said:
I don't understand how you can overlook the fact that every test of rb=c has failed. In fact it's
impossible to test rb=c because it maintains assumptions which themselves are impossible to
confirm in reality.

The problem is that the only "currency" is surviving offspring. The best way of measuring is
counting grandchildren, however this is very difficult to do in a field situation. In a lab it is
relatively easy to manipulate a situation so that something is worth 100 calories to a test animal
and 200 to his sibling. However such experiments don't represent situations that the animal would
experience in its normal life. An exception might be food sharing by vampire bats, when quantity of
food transferred and value, in terms of hours flying time before death, can be measured. However
generally altruism by animals is very hard to measure. rb
Quoted message said:
c seems to be soundly based theoretically, even if empirical evidence is

thin on the ground.

Report

Active in the last 60 minutes

Active in this thread

0 users · 0 guests ·0 bots ·0 total

No signed-in users are active right now.

No known search crawlers active right now.

Jump through the discussion

What we know about this thread

Continue through CyclingForums

Active in this thread