Back when I was an undergrad in Experimental Psychology at the University of Oxford, evolutionary type 'yarn-spinning' was frowned upon by a certain faction of the faculty because it is not falsifiable or experimental. So, this is not to say they were creationists, or that they did not believe in evolution. (They were not, and they did)
So, I agree there is a lot of proof for the theory of evolution. However, it does not as yet provide an answer to 'why things are the way they are' for everything. I trust it will in the future -- is that some kind of faith :) ??
Ok, I'm going to attempt a falsifiable statement thats not reliant on a priori acceptance of evolution.
Evolutionary theory predicts that species that are more like each other in specific ways are more likely to be related than others - or species that are more like each other in DIFFERENT ways.
To clarify - camoflage patterns are not usually considered indicative of close relationship. For instance bees and zebras are both stripy but not closely related; these patterns are both some form of camoflage. Similarity in diet (except in broad generalities), size, sexual system or habitat are also NOT features we expect to see as similar between closely related species. However, similarity in other, specific features is taken as a hint of a close relationship - similar skeletal structure (endo or exo), similar arrangment of internal organs, similar brain structure, similar chemicals used to make these structures.
The null hypothesis from Creationism is that any new species can not be related to existing species in our knowledge base in a logical, coherent way. If God made DNA, there is no reason for him to make it in a branching tree structure, the same way as there is no reason for him to make it follow the predictions I've stated above.
These are the predictions made by evolution & Creationism. Now we test them against real life.
We've been given a brand spanking new species. Now, a priori, there is no reason to classify organism by their skeletal structure INSTEAD of their colouration. (Eg - zebras and bees and ants. If we have no particular framework to work from, we could put bees and ants together (both lacking skeletons) OR bees and zebras together (both stripy.))
Now our brand new species has four limbs with seperate digits built onto an internal skeleton, a uterus stripes, no tail, is exclusively vegetarian, monogamous and colour vision.
A non-evolutionary framework says we may as well call it a new species of bee coz it's stripy and vegetarian.
An EVOLUTIONARY framework says - internal skeleton = vertebrate, four limbs = probably mammal, uterus = definately mammal, seperate digits = Probably either a rodent or a primate, no tail = hmmm, not conclusive.
From OTHER features we say vegetarian = probably a rich environment, monogamous = lots of care from parents required for the young, colour vision = probaly not nocturnal, probably a complex environment, no tail = probably ground living. But this kind of information doesn't tell us about where in the evolutionary tree the new species should be.
If a new species is discovered and the kind of logic I have described results in a classification that is radically different from what the DNA evidence says, then evolution has been falsified. In fact, as there is brand spanking new series in the news at the moment - "Ida" the new "missing link" (bah...) - check out what they say about it.
But the fun of science is in counter-example and falsification, or just the continual attempt to falsify! Not that you are arguing this here, but on a general note, when one decides not believe in (religious) dogma without questioning it, why would you decide to believe in evolution without questioning it?
"Why do Birds of Paradise have long tails" is definately a scientifically unhelpful question. However, based on theory we could guess a few reasons. Basically there are only a few things that matter in life (on an evolutionary scale):
We can totally ask falsifiable questions about any of those. Like: Does a longer tail result in more sex for a bird of paradise. After about sixteen million hours of observation, maybe some genetic tests, we can answer yes or no. Maybe do a few experiments - chop the tails off, stick them onto other birds, put some fake birds in there, see how they respond.
Does that PROVE the tails are "to attract the ladies?" No. But it proves that it has an impact. Proving anything in the past is an ass. Proving a lot of the current is equally an ass. But it doesn't stop us producing falsifiable hypotheses about small subsections of it. And peicemeal we put together a theory to explain why they have long tails.
I know when I was younger having a great tail increased my and everyone else's chance's of getting laid.It could also increase the ability to have a meal bought for you. However a great tail attracted predators, eh, two outta three ain't bad.