Re: How did a monkey give birth to a human ?

Friar Broccoli wrote:
On Apr 5, 9:21 am, John Harshman <jharshman.diespam...@xxxxxxxxxxx>
wrote:
Friar Broccoli wrote:
On Apr 4, 8:06 pm, John Harshman <jharshman.diespam...@xxxxxxxxxxx>
wrote:
Friar Broccoli wrote:
On Apr 4, 5:56 pm, John Harshman <jharshman.diespam...@xxxxxxxxxxx>
wrote:
Friar Broccoli wrote:
On Apr 4, 3:34 pm, John Harshman <jharshman.diespam...@xxxxxxxxxxx>
wrote:
[snip]
Natural selection is differential reproduction
resulting from differences in genotype. That's brief.
.
Shouldn't that be:
Natural selection is differential reproduction resulting from
differences in phenotype?
.
I don't think so. Of course differences in genotype that don't
affect phenotype (very broadly defined) are invisible to
selection. But differences in phenotype that don't reflect
differences in genotype may result in differential
reproduction, but are not inherited and so are not selection.
You could substitute "heritable characters" for "genotype" if
you like, but it comes down to the same thing.
Hmm, yes, I hadn't noticed that phenotype also has problems.
In my opinion "heritable characters" is better but still doesn't
quite hit the mark.
In my opinion "genotype" hits the mark exactly. Some genetic
differences are neutral, but so what? Some phenotypic
differences are neutral. It's the genetic differences that
make selection, even though they're filtered through
phenotypes.
Although it pushes the boundaries of the definition of
phenotype it certainly seems possible to have a heritable
characteristic that has no effect on phenotype, for example a
change in a protein's folding that has no effect on its biological
properties.
Of course there are. Aren't alleles heritable characters?
Yes, thanks.
But I don't have a better concise alternative to heritable
characteristics.
.
You could explain what's wrong with "genotype", for a start.
I thought you expressed it very well when you said:
"differences in genotype that don't affect phenotype (very
broadly defined) are invisible to selection.", so unless we
have a better alternative expression one is unnecessarily
including an unwanted class.
[For the benefit of backspace who may not find our wording
clear: genes/genotype are sometimes different in ways that
do not affect characteristics/phenotype]
Yes, but there are also neutral heritable characters and neutral
phenotypic differences. The only class of characters you could use in
your definition that would fit your expressed requirement would be a
circular one: "Selection is differential reproduction based on those
heritable characters that are relevant to selection."

Why can't we just change "are relevant to selection" to
something like "change operational phenotype" to avoid
circularity here? (You may perhaps want to show that
"operational" reduces to "relevant to selection", but this will
be hard because it is easy to imagine an operational
characteristic [say method of achieving coloration] that is not
actively selected on, although it could be in other circumstances.)

If there are some "operational phenotypes" that aren't selected, how would that be better than "genotype", since your objection is that some genotypic differences are not selected? Now in fact I have no idea what "operational" means here.

Also, just a point of personal philosophy, I don't get too worked
up about whether my words exactly reflect reality, because I don't
believe that my internal representations of reality (including
words) reflect (or can reflect) external reality. I am
perfectly aware that my model of reality breaks down at the edges.

That's nice, though I don't know what this is in reference to. Let's note that you're the person trying to take issue with my terminology.

On the other hand knowing this has made me intellectually lazy,
and I often miss better fits between model and reality.


I think your suggestion "heritable characters" is better for
three (and a half) reasons:
1) It refers directly to what is selected for (characteristics).

Ah, but not all characteristics are selected for; some are
neutral. And alleles are characteristics. This doesn't address
your objection, and in fact your objection can't be addressed.
You just have to accept that selection doesn't apply to all
characters at all times, whatever characters you're looking
at. Then again, the definition doesn't claim that it does. It
doesn't say "depending on any differences in genotype
whatsoever", just "differences in genotype".

I completely agree that not ALL of my objection can be addressed.

My point is simply that "heritable characteristics" gets rid of one
class of unwanted neutral characteristics, namely those that are
expressed in the genotype without any consequences at all
for phenotype.

No it doesn't, unless you think that genotypic characters with no phenotypic consequences are not heritable. I assure you that they are.

2) It specifies that the characteristics selected for must be
transmissible.
Thus from 1 and 2: heritable characteristics is more
complete than either phenotype or genotype by themselves.

But what characters other than genotypic ones are transmissible? Is this
a distinction without a difference?

Obviously yes, since we are just using "heritable" here as a synonym
for "differences in genotype".

So why do you want to make it?

3) The words are common ones and thus more easily understood by
general readers without this specialized vocabulary.
That at least may be true.

Well, one base hit for the home team !!

Though I'm not sure that a general reader would understand
"differential reproduction" either. I wasn't intending that
definition for a general audience that didn't know the meaning
of "genotype".

You're efforts frequently cause me to use the "define:" prefix in
Google searches which I see as good.


So your first two objections are invalid.

I presented no objections. I was arguing FOR the more inclusive
terminology you suggested.

But it isn't more inclusive. It may be less technical, which is something else.

Your second could be valid or invalid depending on intended audience.

On UseNet, if any audience is reached, it is often not the
intended one.

.

Relevant Pages

  • Re: How did a monkey give birth to a human ?
    ... resulting from differences in genotype. ... Natural selection is differential reproduction resulting from ... affect phenotype are invisible to ... In my opinion "heritable characters" is better but still doesn't ...
    (talk.origins)
  • Re: How did a monkey give birth to a human ?
    ... resulting from differences in genotype. ... Natural selection is differential reproduction resulting from ... affect phenotype are invisible to ... In my opinion "heritable characters" is better but still doesn't ...
    (talk.origins)
  • Re: How did a monkey give birth to a human ?
    ... resulting from differences in genotype. ... Natural selection is differential reproduction resulting from ... affect phenotype are invisible to ... In my opinion "heritable characters" is better but still doesn't ...
    (talk.origins)
  • Re: Howard Hershey: "Lewontin is wrong"
    ... Pitmansay that selection happens at the "genotypic level"? ... Selection happens at the phenotypic level, ... phenotype is due to genotype will there be a selective evolutionary ...
    (talk.origins)
  • Re: How did a monkey give birth to a human ?
    ... resulting from differences in genotype. ... affect phenotype are invisible to ... but are not inherited and so are not selection. ... In my opinion "heritable characters" is better but still doesn't ...
    (talk.origins)