[skywhopper]

I will just echo the other posts here that the author of this article or his sources severely misunderstand natural selection and the special case of sexual selection. The most important thing to understand about evolution is that natural selection can only work with the existing traits of an organism, the current environment in which the organism exists, and what random changes happen. Furthermore, every change in the phenotype of an organism has a number of sometimes disparate impacts on its likelihood of reproductive success within its own environment. Given all of that, we should _expect_ to see seeming contradictions like this example.

Most birds can fly because their parents could fly. The genes for flying stick around because so many other traits of birds have evolved to benefit or rely on the ability to fly. But that doesn't mean flying itself is some magical end goal. Flying is only useful for natural selection insasmuch as it grants the organism a better likelihood of reproducing.

Why would natural selection necessarily prefer birds that are ideally designed for flying in what we perceive as a graceful manner? In fact, we know that it doesn't. Ostriches, emus, and penguins can't fly, though their ancestors were able to. That is not evidence that they are evolutionary dead ends. The huge variety of penguin species that have evolved since the ancestor of penguins lost the ability to fly proves the opposite, in fact.

This bird clearly flies well enough to continue to survive. If the traits that work against it flying gracefully grant it more reproductive success than flying slightly better would, then natural selection will favor those traits.

Natural selection is often treated far too preciously. Sure, it took a genius in Darwin to identify and clearly describe the phenomenon, but the process itself is tautological. It comes down to, "the things that reproduce better reproduce better". This bird exists, therefore natural selection favored its traits. If we don't understand why, the failure is ours, not natural selection.

[lapsock]

> the author of this article or his sources severely misunderstand natural selection and the special case of sexual selection.

The author of this article is an Evolutionary Biology researcher at Yale. It would be really weird if he severely misunderstood natural selection.

[dkarl]

I think this researcher had his "science journalist" hat on when he wrote the article. It's a formulaic piece of science journalism, complete with the touch of "most biologists are wrong about something that you, the layperson reader, will realize the truth about after five minutes of tenth grade-level reading." "Most biologists" here might as well be replaced with Mos'blogist, the mythical God of Wrongness, whom we invoke when we want to get excited about learning something new. "Mommy, I don't want to read about evolution." "But I have something here that Mos'blogist doesn't knowwwww! Don't you want to be smarter than Mos'blogist?" "Yay, mommy, I love getting the jump on Mos'blogist!"

Yes, Virginia, "most biologists," who get their news from Nature and Cell, are sadly lagging behind laypeople who read the New York Times. Is it any wonder people have so little faith in scientists? I think science journalists have a responsibility to highlight the fallibility of scientists and the flaws of the scientific process, but it is not public-minded criticism to make your readers believe that the majority of scientists working in an area are too dense to understand something that an average person can understand from a newspaper article, or already did understand if they read a single pop science book about evolution in the last several decades. That's just pandering.

Now, we could make the same mistake and assume that we, who do not write science journalism, have noticed something about it that one of its most successful practitioners (published in the New York Times, after all) never has. Or we could assume he understands it better than we do and does this shit on purpose because he likes being published.

[psyc]

It seems the several critics in this thread think that generalizing and flattening fitness into One Principle is a superior perspective. There's nothing wrong with the article, since the whole point of it is to talk about the nuances of different types of selection.

[skywhopper]

Fair enough. I guess I should have said the the text of the article uses an outdated, narrow, and confusing definition of natural selection, which is a great way to sow uncertainty and doubt about how evolution works.

Sexual selection definitely produces some interesting effects, but it's still a form of natural selection. This article portrays it as the opposite. I think it's far easier to understand the outcomes of sexual selection if you realize that reproductive success is the only thing that matters. And the factors that influence reproductive success vary widely. It's important to not get eaten, die of disease, starve to death, or die young. But for organisms that reproduce sexually, it's even more important to be able to find a mate. An organism that lives twice as long but reproduces just once is far less successful than an organism that lives half as long but reproduces twenty times. The whole idea that sexual selection is somehow at odds with natural selection instead of just being one of many selection pressures, is deeply confused.

[mirimir]

> The whole idea that sexual selection is somehow at odds with natural selection instead of just being one of many selection pressures, is deeply confused.

Well, sexual selection is distinguishable, because it's not about the environment or other species, but about interactions between sexes. I mean, consider ducks. How does the penis-vagina arms race have any relevance to overall fitness? Indeed, it seems more like a bug in the sexual reproduction app.

But on the other hand, human consciousness and language are arguably products of sexual selection. And they have increased fitness. So maybe it's most useful to consider sexual selection as just another source of variation, grist for the mill of natural selection overall.

[Analemma_]

It's not a Hacker News thread unless there are a bunch of computer programmers telling people with PhD's how ignorant they are about their own field.

[qb45]

The problem in this case is that TFA presented this phenomenon as some black magic unexplainable by science, despite reasonable explanations having been provided half century ago and popularized even among lay audience by books like The Selfish Gene.

But hell if you are wrong about the general trend ;)

edit:

Really what TFA did is it just falsified some extremely naive and humanly-subjective interpretation of the slogan "survival of the fittest", without bothering to mention that this interpretation had already been known to be defective years ago and there are subtler alternatives which don't have this problem.

[emmelaich]

Well said and funny but it won't stop me piling on :-) ....

> Even the peafowl has a longer tail than she needs.

Surely he means peahen, the female peafowl (of which the male peacock has the extraordinarily long tail.)

[dekhn]

everybody misunderstands natural selection. the field is large enough for a great deal of misunderstanding.

[joe_the_user]

Ostriches, emus, and penguins can't fly, though their ancestors were able to. That is not evidence that they are evolutionary dead ends.

Well, I can't say concerning Ostriches etc but a common pattern in recently emerged islands is for birds to reach the island first. There not being any predators, some of the bird species evolve to being flightless. Once rats or other mammals reach these Islands, these species tend to become extinct. So naturally occurring "evolutionary dead-ends" are certainly possible (European sailors releasing rats and rabbits to the many islands they visited accelerated the process but it is still "inevitable" since mammals will to a given Island sooner or later).

Sure, one can define selection and fitness "internally" and thus whatever survives is "fittest" up to a given point. But it seems reasonable to take an external view and judge that certain evolution paths lead to the end of species. There's no reason abolish this kind of analysis just because one has honed the specific, technical meaning of these terms so that they aren't concerned with the question.

[ccallebs]

> The most important thing to understand about evolution is that natural selection can only work with the existing traits of an organism, the current environment in which the organism exists, and what random changes happen.

I believe most of the time you're right, the changes are random. However, at least one experiment has shown that species can pass adaptive traits to their offspring (called Lamarckian inheritance) [1]. The experiment is intriguing, in that it seems to hint something more is going on than the Darwinian inheritance model.

[1] http://newsroom.cumc.columbia.edu/blog/2014/07/17/effects-of...

[JacksonGariety]

"Clearly the telos of the bird is to fly well, as that of a lyre player is to play the lyre properly!"

It's Aristotelian biology.