That's not the definition I was given. See post #175
Yes it is. Read for understanding, for heaven's sake.
(The definition is not even in that post. It's in a link you quoted in that post.)
The definition gives you a very helpful example. "The sky is green" is logically possible. How can that be l.p. and "there is no tree" not?
Depends on your definition. I've seen it defined as a statement which is necessarily true. If the claim is, "It's noon on 1/29/09 and there is a tree outside my window", that will always be true even though time necessitates a change in tense.
(1) If time necessitates a change in tense then the original statement is NOT always true. Changing the tense makes it a different statement. In any case, even if you phrased it in a way that time doesn't change its truth (I think "there is a tree outside my window at [point in time]" would work, for example)...
(2) ...that statement is
still only true in some possible worlds. There
can be worlds in which there is no tree outside your window at that point in time.
It can be plausibly interpreted that way, but there is no way to test the interpretation.
I think I didn't explain the situation quite correctly. The broader claim being tested (= attempted to falsify) is that macroevolution
can be extrapolated from microevolution. Now, you can't test such a broad claim with a single experiment, so only one very specific part of it is actually tested: that the underlying genetic changes are the same in the specific case of, what was it?, plumage colour in swans.
The null hypothesis is, as usual in science, a hypothesis of "no effect" or "no difference":
H[sub]0[/sub]: the same kind of genetic change underlies within- and between-species colour variation in swans.
The alternative hypothesis is, then:
H[sub]1[/sub]: within- and between-species colour variation in swans is caused by different kinds of genetic changes.
Investigation of the genetic basis of colour gives fairly straightforward support to either
H[sub]0[/sub] or
H[sub]1[/sub]. It doesn't, strictly speaking, confirm the grand hypothesis (that you can extrapolate), but it does tell you that
in this particular case, the two processes are not qualitatively different.
The more such particular cases fail to reject
H[sub]0[/sub], the more confidence you have that the extrapolation can be generalised.
Hypothesis testing and inductive reasoning, that's what science is all about. Never 100% certain, for sure, but often very confident.
That methodology is testable, most obviously against the testimony of such witnesses as the parents.
Because witnesses are
so reliable?
Do you know how easy it is to plant false memories in someone's head? (People do it to themselves, IIRC) To forget or distort memories, especially if they are emotionally loaded? To have an agenda? To simply not
know who the most likely father was? Witnesses are to be taken with a pinch of salt, even if they are perfectly sane, unbiased adults.
I think you have a general problem with abstract reasoning, what with the inability to understand logical possibility and the obsession with witnesses
Then you understand that it was the extrapolations I called absurd and not the evidence, right?
It seemed to me that you think the extrapolation is the whole basis of common descent. Sorry if I misunderstood you.
First I need some reason to think the inference is testable - i.e., that there is some way of determining that the various genetic correlations could not possibly result from anything but common descent.
Being testable does not mean that. Being testable is essentially the same as being falsifiable. It means the logical possibility of being proven false,
not that of being proven right.
"Right" is what remains after you've excluded everything else you could think of.
The case for common descent is so strong because it comes from so many independent lines of evidence, all of which could have proven it wrong, and none of which have (that I know of).