Scott wrote:But when species G evolves from F which evolved from E which evolved from D which evolved from C which evolved from B which evolved from A, the difference between A and G may be much greater than the difference between A and B or between B and C, etc. This is part of why the moon analogy fails--because the question isn't so much whether the difference between G and A is categorically different than the difference between B and A but rather the question is whether the difference between G and F is categorically different than B and A.
Meleagar wrote:So far, it hasn't "failed". You've only asserted the contrary.
No, I didn't only assert the contrary, I explained it in the quote above. The jump from A to G is categorically different than the jump from A to B. But the jump from F to G is not. The jump from a creature who cannot move very fast to one who can fly well is categorically different than the small leaps in evolution by natural selection that have been demonstrated. The jump from a creature who has absolutely no sense of light waves to one who can see and then to one who can see with stereoscopic eyesight is categorically different than the small leaps in evolution by natural selection that have been demonstrated. And indeed the jump from creatures who have no legs or arms to erect, bipedal creatures is categorically different than the small leaps in evolution by natural selection that have been demonstrated. They are all categorically different from A to B in the way that A to G is categorically different than A to B. But they can be broken down into small steps in which case each of those smaller steps is not categorically different in any way that would make it unreasonable to, after getting from point A to point G, conclude we did it by doing the same thing done with point A to B and to point B and to C to go from point C to D and from D to E and from E to F and from F to G.
I don't know how I could convince you that it is reasonable to assume that if you can walk across any one town that is 5 miles long that you could walk across the whole state that is 500 miles long, if you have gotten across the state and there is no evidence of any other form of travel. If you want me to show you that each 5-mile-consecutive-set in that 500 is not categorically different than all of the others, I don't know how to do that if I don't have the time or specific knowledge about every single 5-mile-consecutive-set. But if every 5-mile-consecutive-set at which we do look in-depth, we see you can and did get across by walking it then it is reasonable not a compositional fallacy to assume you can and did get across all the 100 of the 5-mile sets particularly if you have already gotten across the full 500-miles and there is no evidence of any other explanation. Your moon example and your War and Peace example are each a false analogy and they may, being false analogies, be compositional fallacies unlike a reasonable induction. Indeed, ironically, it does seem fair to conclude that you can't walk to the moon because you can't walk to the clouds. And if you could walk to the clouds between where you were on Earth and the moon, then I think that would make it more reasonable to conclude you can walk to the moon, particularly if walking was the only known form of travel and you have traveled to the moon.
A required aspect of any scientific theory is falsifiability. The reasonably achieved theory that I can walk across the 500 mile state would be disproved by the discovery of a magical force-field in the middle to which nobody can walk-through but would require some kind of special spaceship to cross. The theory of gravity would be disproven if we woke up one morning all floating away. The theory of evolution by natural selection would be disproven if there was a single small leap that couldn't have occurred or big leap that couldn't be broken down into smaller leaps. So if your point is that the theory is fallible or falsifiable, then I agree. It is as at least as falsifiable as the theory of gravity. Every logically valid induction is--technically--fallible. But it is incorrect to say these theories commit a so-called composition fallacy such as the alleged fallacy in your moon example.
Together we personally can try to research the known data to see what evidence there is that a certain small leap could have or did occur by evolution by natural selection or not if that's what you really want to do. But if we don't want to spend years and get a doctorate doing it, then we need to break it down into one or two small leaps, such as the jump from winged creatures that could glide long distances to ones that can actually fly or the jump from creatures who can't walk on two legs to ones who can but not for long (like my old cat), or we could do ones that usually don't walk on two legs to those who usually do. We personally don't have the resources to do them all, just as we may not have the resources to research in-depth each mile on a 500-mile-consecutive set of land. But if we look at some different sets of 5-consecutive-miles within the full 500-miles and find each of those are walkable or were walked, it is reasonable in the absence of any evidence otherwise and especially under the fact that we have travelled the 500-miles to conclude we did it by walking; right?