[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Re: Origins (was: Re: Sharovipteryx)

Honorable Person Ralph Chapman said:
First of all - The mixing of the two hypotheses in questions, origins of
taxa versus origins of flight, is logically awful - they are totally
independent questions...
 Make all the arguments you want about bottom-up or top down, but you have
said nothing that affects the discussion of whether birds are from dinos,
are only distant cousins of dinos, or dinos are from birds. Nada. Nil.

>From following the discussion, taxa have been considered related because of
features related to function.  For example, the observation has been made
here that feathers evolved only once.  Therefore, any animal with feathers
must be related to any other animal with feathers, and both animals must be
related to (or be) birds.
Further, I've seen the assumption that if the feathers are effective for
flight then the animal must have flown (or leapt or swooped...).
So, feathers can define the relationship of taxa and can indicate the origin
of flight.
(Of course, if feathers could have evolved more than once, including in a
line parallel to birds and their immediate ancestors, then this logical
construct disappears.)

So, what is a just-so story?  I think of it as one which fits the available
information and explains some existing situation in a way which has not been
confirmed or refuted, directly or indirectly.  Without indirect
confirmation, a historical science would inevitably fit this definition.  Of
course, so would plate techtonics or macro-evolution or the other theories
mentioned in other postings.
A theory is a just-so story with such substantial indirect proof that it is
no longer treated as just-so.  The question being debated so strongly is
what constitutes sufficient indirect proof.  All the discussion of coding
characters is based on the assumption that the means of analysis used can
produce sufficient indirect proof.
However, just as a whole construct rests on the assumptions that feathers
evolved once, so character coding is based on the assumption that it is poss
ible to devise an algorhythm capable of weighing relative degrees of
likelihood in any situation.  The problem is confirming that the assumption
is correct, and this can be done only indirectly.  (An indirect proof of an
abstract conclusion!  You'd have to be careful that assumptions without
direct confirmation were not shared by the algorhythm and by the means of
testing it.)

This type conclusion is different from conclusions coming directly from
observation.  Given many similarities between the observed effects of
herbivory on modern herbivores and the presence of all these markers on a
member of a group of dinosaurs, it would seem reasonable that the dinosaurs
were herbivores.  Until someone comes up with other markers of omnivory.   I
think I read someone recently did just that on a group of dinosaurs.

 The certainty which is a contrast to a just-so story is not that of
metaphysics, but of some parts of chemistry, for example.  My daughters do
experiments which produce the expected results.  (They also do some which
produce foul smelling dark substances, but these are the result of their lab
partners' intervention.)
Honored Person Chris Brochu said:
We know that people with the flu tend to be infested with one or another
strain of the influenza virus; we know from
laboratory work how viruses (including influenza) work within a host cell;
we even know some of the specific mechanisms by which influenza viruses
cause some of the symptoms of the flu.  But since we have never actually
seen influenza viruses doing this in real flu sufferers, we don't actually
*know* that the influenza virus is the ultimate cause of the flu.
Well, we do know one type of virus from another, and we can, if we were
mean, infect someone with the virus and find exactly the result predicted.
This is direct proof, without considering what we can and cannot see.  I
wonder, in fact, if we haven't been able to see so much directly that the
'germ theory' hasn't ceased to be a theory?

For contrast, the Very Honored, Patient, and Tolerant Mickey Rowe commented:
If you find that, for example, parakeet, anole and goldfish acquired one of
the principal components of color vision from their most recent common
ancestor then you can infer that dinosaurs should have inherited that
component as well.
We can find:
-the most recent common ancestor of parakeets, anoles, and goldfish
-what a principal (essential?) component of color vision might be
-a way of determining that this component was inherited straight through to
the modern species
-the connection to dinosaurs
-the dominance of this component so well proven that we can infer that dinos
must have inherited it.
I know you were joking, but it's a wonderful example of rampant assumptions.

So, in brief (too late), may I suggest to you George that we find a
substitute for the term 'just-so story'?  To avoid treading on
sensibilities, may I suggest the term 'testable' or 'falsifiable'
'hypothesis'?  Has a good scientific ring to it, no?  (Kidding, kidding!)