> John Edser  wrote in message news:
> l03130301ba05cbc0218c{at}[10.0.0.3]>...

> >All you have done, is by definition, allowed
> >Hamiltonian fitness to represent ALL
> >fitness within evolutionary theory

Hence "inclusive" fitness.  As its name suggests, it is
all-encompassing.

> > using however,
> >just an oversimplified model  of fitness.
> >
> >Like Hamilton, you did not define and simply
> >omitted Darwinian fitness from the simplified
> >model. Darwinian fitness is the only refutable
> >theory of fitness we have.

Define Darwinian fitness, and justify this last remark.

> > Hamilton's model, "toy"
> >fitness is just an oversimplification of Darwinian fitness.
> >Until you include Darwinian fitness as a valid constant

A valid constant?  What?

> >within your model, you are, by simple default, allowing
> >Hamiltonian fitness to contest Darwinian fitness
> >and win, by omission. This is logically invalid when
> >Hamiltonian fitness is just an over simplification of
> >Darwinian fitness, anyway.

I was not aware that there was any serious competition.  Hamiltonian
fitness, as you call it, is simply fitness.

> >> >  JE:-
> >> > Obviously  c is _not_ just a cost to rb it is also a competitive
> >> > form of fitness in its own right which can validly be expressed
> >> > positively. Within the rule rb can also be regarded as
> >> > a cost to c depending of course, which form of fitness is
> >> > actually operating  within _nature_.
>  
> >> NAS:-
> >>  there is some predefined tradeoff between the two.  this isn't
> >> necessarily the case.
> >
> >JE:-
> >It does not have to be the case if they are
> >not competing for the same, finite, reproductive
> >resources available to any selectable form. Clearly,
> >they ALWAYS ARE competing for such resources

What?  c and b are not competitors competing for resources?  They are,
respectively, the cost of an act to an actor and the benefit of the
act to a recipient.  They are not competing entities.

> >unless Hamilton just assumed that these resources were
> >infinite per unit of selection. In this case Hamilton's
> >rule is not even a simplified model, it is just
> >a hypothetical that need have no relationship
> >at all to reality.

I believe you have grossly misunderstood the concept of inclusive
fitness.

> >Nobody can deny that Darwinian fitness
> >is the only fitness that can be tested to be
> >operating within _nature_ so at all times, rb must be
> >a cost to c 

Why must rb be a cost to c?  What does this mean?  That an action
which benefits a related organism *must* be detrimental to the actor? 
This is certainly not the case.  How the action benefits/costs the
recipient and the actor depends on the action.

> >and not the reverse within any valid
> >_science_ of evolution. Hamilton's clever reversal
> >of this situation, using however, just an oversimplified
> >model  invalidly reversed cause
> >and effect within all subsequent fitness calculations.

Please expand.  What was reversed?  

> >This led to the erroneous view that selfish geneism
> >can validly explain Darwinian fitness altruism
> >(Hamilton, Dawkins, Wilson, Trivers etc etc) producing
> >a multi million dollar industry based on just an invalid
> >assumption of nature.
> >

Multimillion dollar industry?  

> >> NAS:-
> >>  it is easy to think of scenarios in which an
> >> act can simultaneously increase both direct fitness and the direct
> >> fitnesses of relatives.
> >
> >JE:-
> >Yes this is fitness mutualism. However _unequal_
> >fitness mutualisation can superficially look
> >like fitness altruism. 

Yes it can.  It does not mean that it is.  And neither does it mean
that altruism  ("-/+" interaction) cannot occur.  It is true that b
and c and r can be hard to estimate, but this does not invalidate
hamilton's rule.

> >At all times, Darwinian
> >fitness mutualisation  is the only relationship that
> >can produce an ESS (evolutionary stable strategy).
> >All model instances of fitness altruism are logically
> >invalid and are not needed, anyway.

What about the widespread use of Hamilton's rule in sex allocation
theory, perhaps the best empirically-supported branch of evolutionary
biology?  Hamilton's rule clearly seems to work there.  As i've
mentioned before, it is so reliable that Hamilton's rule is now used
to obtain very exact estimates of inbreeding in parasite populations
by looking at their sex ratios (sex ratios are often very much more
easily measured than inbreeding rates).

> >
> >> > JE:-
> >> > In any instance where Darwinian fitness absolutely increases no
> >> > fitness altruism can exist, no matter how high  inclusive fitness
> >> > is, even when inclusive fitness is relatively >
Darwinian fitness.
> >> > Inclusive fitness (rb) and Darwinian fitness (c)
> >> > do not have to compete against each other for finite organism
> >> > reproductive resources as Hamilton suggested, they can compliment
> >> > each other such that rb only represents an investment  cost for an
> >> > absolute increase in Darwinian fitness (c), even when
the Darwinian
> >> > gain, in relative terms, is less than the inclusive
fitness gain. As
>  long
> >> > as Darwinian fitness absolutely increases,  inclusive fitness must
> >> > be absolutely < Darwinian fitness.
>  
> >> NAS:-
> >> again, i believe this to be a misunderstanding of what inclusive
> >> fitness actually is.  it is the total fitness, the maximand under
> >> optimization.  hamilton tried to convey this, hence the use of the
> >> word 'inclusive'.
> >
> >JE:-
> >Hamilton's rule is entirely misused as
> >an explanation for all fitness within
> >evolutionary theory because a "toy"
> >simplified model cannot ever
> >_conclusively_ describe all fitness
> >within evolutionary theory.

Well, yes and no.  To truly be able to describe fitness we would
require a deterministic universe (as in, if we rewind the tape, and
replay, we would have the same events over again), because
stochasticity plays a vital role in evolution.  But since we often
don't like to equate lucky individuals with fit individuals, our
definition of fitness is more in keeping with some sort of potential
that an entity has for levering itself through generations, rather
than its actual, realised success.  We are then interested in asking
how an individual maximises its 'potential'.  This is akin to
partitioning allele frequency change into that due to natural
selection and that due to all the other processes, as Fisher did in
1930.  Insofar as we are interested in this natural selection
component, Hamilton's rule is a tautology, a necessary truism.  Yes,
it is not a complete picture, but the whole point of science is that
we are trying to pick the patterns out of the noise.  If i see an
animal walk to a pool, stoop to drink, and then get hit by some debris
that has fallen out of the sky from space, i do not consider this
exciting suicidal behavour on the part of the animal.  I just consider
the animal unlucky.  If i want to know why the animal walked to the
pool, i understand that the space debris was unforseen, and that the
walk to the pull was a fitness-enhancing activity after all (even if
it didn't turn out that way).

> >Only a testable theory that uses
> >a defined fitness, i.e. Darwinism
> >can do this even if Darwinian fitness
> >was implicitly defined.
> >
> >
> >> > > > JE:-
> >> > > > so no altruism has been proved, just assumed.
> >> > > > In reality rb< c where helping rb _increased_ c
> >> > > > (Darwinian fitness) more than it cost.
>  
> >> > > NAS:-
> >> > > increasing c is increasing the cost!
>  
> >> > JE:-
> >> > Not necessarily!
>  
> >> NAS:-
> >> increasing c is increasing the cost.  regardless of what happens to
> >> the benefit, the costs are still increased.
> >
> >JE:-
> >"Increasing c is increasing the cost'" is
> >only valid if you assume c to always be a cost to
> >rb, irrespective.

Again, you are clearly missing the point.  c is a fitness component,
rb is a fitness component.  they are not entities competing for
resources.  it may be true in some cases that there is a tradeoff
between the two (although it would not be much of a tradeoff, as
surely selection will favour those who invest in rb rather than c!),
but as you know it is quite possible that animals can engage in
mutually beneficial interactions.

> > Clearly, this is not the case since
> >you  admitted that -c is not a cost. If is not
> >a cost in this instance. Here rb must be a cost
> >to c or no costs now exist within the rule and
> >everything has been done for nothing.

rb is the overall benefit to the recipient, weighted by relatedness. 
If you want, you may subpartition them.  If i lift a cup to my mouth
and drink, i pay an energetic cost and gain a nutritional benefit. 
the c in this case is the net fitness effect to me.  if the net effect
is that the action is beneficial, then c will be negative, as it has
been defined as a cost.

> >Hamilton's rule now turns into a perpetual
> >motion machine for fitness.
> >

I don't believe that you have read any of Hamilton's papers at all.  

> >The variable c can be regarded as a cost to rb
> >when rb> c.  It is not logically excluded within
> >his model for rb< c. In this case rb is better
> >regarded as a cost to c than the reverse.
> >

?????

> >
> >> > JE:-
> >> > If both rb and c _absolutely_ increase, then normal
> >> > Darwinian fitness is alone, operating.
> >> > Now rb is logically a cost to c
> >> > (the exact opposite to what Hamilton supposed).
>  
> >> NAS:-
> >> You have not demonstrated this at all.
> >
> >JE:-
> >It is valid to suppose that rb and c can
> >both increase, by simply suggesting an
> >absolute  increase in both. Fitness
> >mutualism, which you indicated you
> >agreed with, alone, produces any
> >mutual fitness increase.
> >

so now you admit that rb and c can both be increased!?!?

> >If rb is an investment cost to D, Darwinian
> >fitness (the opposite of Hamilton's supposition)
> >then both rb and c can mutually increase. However,
> >the relative difference between this mutual, absolute
> >increase may indicate that rb had increased more than c.
> >In this  instance, any relative measure of the
> >difference  between  these two fitnesses, that compete
> >for the  same, finite, reproductive  resources will appear
> >to satisfy Hamilton's rule so that rb> c when in fact, in
> >the absolute sense, the rule has not been satisfied,
> >i.e. here rb was actually < c, _absolutely_ but rb just
> >appeared >c, relatively.

This is clearly going to take more work.  But at the moment i have to
run.  I suggest you read my comments, get hold of Hamilton's papers,
and then have a rethink.
---
þ RIMEGate(tm)/RGXPost V1.14 at BBSWORLD * Info{at}bbsworld.com

---
 * RIMEGate(tm)V10.2áÿ* RelayNet(tm) NNTP Gateway * MoonDog BBS
 * RgateImp.MoonDog.BBS at 11/26/02 6:12:17 PM