Explain a bit of how genetics works, then mutation, then heredity, then selection, then show how the combination causes evolution.
This is a lesson from my 10th grade biology class:
Take a bunch of marbles with two different colors. (Lets say blue and white)
Explain that these are the possible alleles in the population, with blue coding for hair and white failing to code for hair.
If a rabbit inherits any blue marbles, it has hair. If it has only white marbles, it isn't born with hair. White marbles represent recessive alleles, but that's not particularly important to evolution.
Now mix all the marbles up in a bag/hat/whatever and draw them two at a time. This produces one rabbit. Write down what that rabbit's traits are. The mixing and drawing represents random mating, saying that all the living rabbits have 2 children, with randomly selected other rabbits. These assumptions aren't accurate, but evolution will still be demonstrated when...
You draw out all the marbles and produce a generation of rabbits. Now say that the winter was really cold and killed all the hairless rabbits. Remove all the pairs of marbles which are all white.
Mix marbles, repeat, take note. Explain that the white alleles are being selected out of the population.
Now just say that surviving rabbits will consume resources to make up for the deaths of the white ones, or not. This basically just demonstrates that a selection pressure will remove deleterious genes from the population.
It will still probably take work to explain that mutations will lead to different traits which are then selected for.
This is a great technique -- you could probably save time/explanatory effort if you used the moths example -- moths going light-skinned or dark-skinned as trees get more/less polluted. The obvious retort to your rabbits, is, ok, that's how mutant freaks get eliminated, great. So species stay the same. You need (at some point) to illustrate how species change through useful mutations, and you may as well do it at the same time as you illustrate natural selection.
I have a creationist friend with no particular rationalist or scientific training. She recently asked me to send her a "list of evidence for evolution that persuaded me." After some prodding, it was revealed that she's getting into an argument with another friend of hers who believes in evolution. I'm assuming that she wants the experience of arguing with someone who's on level footing with her. It seems like a good opportunity to broaden someone's mind in a more general way that'll benefit them in the long term. I don't particularly care whether she believes in evolution (it probably will not impact her or the world in general if she changes her mind about it). But I'd like to phrase my e-mail in a way that's most likely to cause her to re-evaluate her worldview.
Subgoals related to this:
1. Point out that "losing" an argument can allow you to learn things, and if you honestly care about truth you'll try your best to evaluate ideas from other points of view and consider what it would mean if they were true. Do this without sounding condescending.
2. Give her a line of retreat by proposing that evolution is compatible with the Original Sin interpretation of genesis (which is very important to her and I would never attempt to argue against).
3. Give as much background as possible on the scientific method.
4. Still manage to focus the bulk of the e-mail on the most persuasive facts supporting evolution, otherwise I'm obviously not satisfying the criteria she actually gave me. I don't mind taking advantage of her request for my own purposes, but only if I'm actually helping her with her stated goal.
5. Specifically show why macroevolution is not only possibly but likely. (I'm pretty sure she either already believes or could be easily persuaded to believe in microevolution)
6. DON'T focus too much on why creationist arguments are flawed (she hasn't even used any yet, and it sends the wrong message about trying to actually figure out what the truth is)
7. Accomplish everything in approximately 3000 words, without using jargon, designed to be read by someone who's mental architecture isn't particularly adapted to rationalist thinking. (Most people aren't.)
I believe I can do a decent job myself. But it'll be a fair amount of work, and I want to know if anyone had a recommendation for a particularly good essay that I can either link her to or borrow pieces from. I might also include a link to a page of common bad creationist arguments and why they don't make sense.