Evolution is not random.

I'm reeeeeeally curious what scientific journals all you smartypantses who disagreed are reading. This isn't an opinion, you know. :)

The beauty of Jyte is that there's no need for someone to actually know what the hell they're talking about to pipe up. If you don't believe me, look at me and Rachel. One of us is clearly wrong a lot, so one or the other of us is clueless.

I changed my answer because it is right, but it's driven by mutation which is a random process, which is what I was thinking of when I disagreed.

Even I make mistakes.

Maybe you should clarify what you mean, Ryan. I thought that there was a notion of random mutations being the basis for most changes from one generation to the next, but with more "rational" natural laws governing which are kept and which aren't. So in one sense, the whole universe is random, whereas from a wider perspective it's orderly.

I'm curious about what you mean by "not random". It's true that not all offspring are equally likely to survive, so there is a force that encourages changes that promote fitness. However, random events are a part of evolution. Something can die out due to a completely random accident. A better mutation can be lost for a completely random reason.

Yep, random mutations are one of the primary mechanisms of evolutionary change, but those mutations are meaningless without natural selection, which is as un-random as you can get.

Mutation alone cannot drive evolution; only mutation plus natural selection.

At its core, I suppose it's really a semantic issue: "evolution" is growth or development arising from change, not just change itself, therefore by its very definition it must be a non-random process.

Rachel, you're correct that random events can have serious effects on the evolution of a species, but just because the cause is random doesn't mean the effect is also random.

If a seemingly random weather event causes a particularly cold winter which in turn causes the extinction of the Tri-winged Foo Bird of Timbuktu, the bird went extinct because it could not adapt to the cold. Other species survived because they could adapt (or had already adapted). That's a non-random selection event caused by a random catalyst.

Yes, but if the highly appropriate for its environment weasel with new and useful mutations just happens to be crushed by a rock while its less well-adapted mate is to the left, then that will have an affect that is not at all based on fitness.

Overall, the more fit will survive more often, and that will cause non-random development. But very good mutations can be lost do to completely random events, and poor ones can survive for the same reason.

If the weasel gets crushed by a rock, he hasn't evolved, he has been squished. Unless he reproduced before he met Mr. Rock, his genetic information has been removed from the pool and evolution has not occurred. Therefore there was no random evolution.

If he reproduced before getting squished, then perhaps his weasel-babies saw him get squished and learned a valuable lesson about not standing under rocks. That lesson, then, would be a non-random result of a random event (although it probably wouldn't quite count as evolution).

My point is about the weasel who was right next to him who has inferior traits (say worse eyesight), but didn't get squished, goes on to reproduce and have 10 weasely babies, while the new mutation of better eyesight got squished under a random rock. That could certainly happen. Evolution is not going to save every positive mutation or always have the genetically better reproduce, because sometimes random events will alter things.

On average, the fitter will survive. But sometimes a random worse mutation will instead.

Also, if someone has two mutations or rare genes, one quite good and the other somewhat bad, just randomly having both, then the somewhat bad one might propagate because of the random fact of being had by someone with a different difference that is a major advantage.

I understand the point you're trying to make, but it's not valid. In your example, evolution has failed to happen. It hasn't happened randomly.

You're correct that evolution does not always result in something that's better than what came before, but this isn't the result of random evolution; this is the result of changing selective pressures. While the causes behind selection may be rooted in randomness, selection itself is not (and is inherently incapable of being) random. That's why it's called selection.

Ryan, I see what you are saying. It does boil down to semantics, but I personally feel they are important, and you are correct. In fact, I hadn't realized that my argument on the claim that inspired this is—that evolution is random—is exactly that which is used by creationists to refute evolution.

Thanks for the insight and the inspiration to do some research.

Prunella: My pleasure. :)

Roland: There is no book describing Rachel's weasel-squishing rock, because as cool as weasel-squishing rocks are, they don't bring about evolution (at least, not in the example Rachel gave). But even if they did, it still wouldn't be random evolution. ;)

Evolution may not be random. But I certainly am. No Digg.

You've hit the nail on the head, Roland.

I think the point of the weasel example was missed.
Traits that adapt the creature to its environment are preserved and passed on. If falling rocks are statistically the major cause of failed reproduction then you can't assume the squished weasel was carrying a beneficial mutation if it was crashed by a rock. Also, the probability of a single event is meaningless. It's frequesncies that count.

This is really confusing.

Random weasel-squishing rocks are not generally covered in texts on evolution, but random mutations are. There may be a great little mutation waiting to happen, but if the random bit-flipping that produces mutation somehow never makes it happen, then natural selection will never get the chance to operate on it. Rocks aren't key to evolution, but randomness is.

We don't generally consider the impact of weasel-squishing rocks on evolution because it's a low-probability event or, if it's a high-probability event (because your proto-weasels live in unstable terrain), it won't preferentially target one weasel rather than another. While a random rock-squishing event might happen to head off a promising genetic development for a while by crushing the weasel that expresses it, the odds are that if the development occurred once, it can occur again. One random event can change the course of history (Gavrilo Princip going to Schiller's for a sandwich) but random events tend to get 'smoothed out' over time by sheer weight of numbers.