Consider this: there must have been quite a lot of prehistoric humans who died, or otherwise suffered reproductive failure, because they lacked eyebrows.

Breaking major bones, especially in times before modern medicine, was without a doubt a life-threatening event. And considering the prevalence of fractures today, with our largely sedentary lives, it must have happened fairly often

A bit pedantic, but: there doesn't need to be a direct line between a morphological feature, like eyebrows, and death and/or inability to reproduce. Evolution is a statistical process that operates over an entire population of organisms. So, a feature like eyebrows can evolve even if they have a relatively minor impact on organisms with eyebrows, like, say, by making them 3% better at hunting or whatever. Because if you have a population of thousands or millions of organisms, that 3% hunting advantage will make a difference.

1. Eyebrows are a pretty elaborate structure to have evolved purely by chance.

You don't get from not having eyebrows to having eyebrows via a single mutation. It seems reasonably likely that the specific shape of eyebrows (as in, one person has bushy eyebrows and another has thinner eyebrows) might be tied to sexual selection (as in, the variations don't serve a functional purpose other than being attractive to the opposite sex). But that eyebrows exist at all is probably tied to some survival/reproductive benefit.

2. I think you underestimate the impact that something as a small as, say, a 3% improvement in an outcome can have on the evolution of a specific trait.

You can use a spreadsheet to model this yourself. Start with a population where 1% of people possess some trait that allows that subpopulation to have 3% more children per generation that the other 99% of the population. I'm showing that it takes a little over 150 generations for the descendants of that first group to comprise 50% of the population. As a point of comparison, there have been roughly 7,000 generations of humans. So, you can see that natural selection will act even on minute details of anatomy because when you play them out for many generations they matter.

We will probably never have a definitive answer to the question of why we have eyebrows. The best we'll ever do is some decent hypotheses. So if the argument is that eyebrows serve function X, but we find out in the future that eyebrows do not, in fact, serve function X, then a naive observer will think that eyebrows did not evolve to serve any purpose and are, in fact, just a random feature.

IMO, a more thorough argument is to point out that the only way species-wide, complex morphological features like eyebrows can possibly exist is if they serve an important purpose, even if that purpose has a miniscule impact on the outcomes of any given organism. This argument is resistant to any given explanation for the purpose of eyebrows turning out to be wrong.

Plus it gets to a deeper understanding of evolution.

P.S. I'm skeptical about the sweat in the eyes explanation for eyebrows. Anecdotally, my eyebrows have never kept sweat from getting in my eyes. And less anecdotally, if you spend a few minutes looking at somebody's face and imaging sweat rolling down their forehead, you'll see immediately that eyebrows aren't very effective for sweat diversion.

My money is on protecting our eyes, which are our most important sense organs. If you google image search "mma black eye", you'll see what I think is kind of remarkable: despite serious trauma to the surrounding areas, most people's eyes are intact.

That's not an argument. It's a tautology. In any field other than evolution, such a statement would be ridiculed.

There are infinite reasons that could explain eyebrows. Design. Alien intervention. Lamarckisn processes. The Will of the Flying Spayhetti Monster. Out of this myriad of possibilities, if you suggest they arose because and persisted due to conferring a survival/reproductive benefit, the burden is on you to demonstrate such. This can be done in two ways. The first is by storytelling... creating a Kipling-esque 'just so' story involving, perhaps, sweat or sexual attraction or whatever. The second method is by using science. Namely, measuring the degree (if any) that eyebrows effect survival/reproduction. From that number, one can backtrack and determine mathemstically how many generations it would have likely taken for the phenotype to become ubiquitous, and see if it lines up with other assumptions. To summarize, before conjuring up 'just-so' stories, find 1,000 people with no eyebrows, a 1,000 with bushy eyebrows, and show me a measurable difference in their fecundity rates. Or, use chimps. Or, dogs. Doesn't matter. As mentioned, I personally don't think eyebrows have squat to do with survival/reproduction... you can't counter my null-hypothesis by assuming a priori that every phenotype must confer some sort of advantage.

I tend to agree with you that a lot of things that are called "science" are mainly story telling. There is "physics envy" for a reason. Physics focuses on a very small subset of reality that have observations that are true with uncertainties of 5, 6, 9 sigmas. This is the level of truth one needs to build up a models of complex things that match reality almost all the time. Physicists at LIGO or CERN don't publish in journals some observation with P=0.05 and expect others to use it. Hopefully the reproducibility crisis in psychology will get many truth seeking fields to spend more time on understanding what is true at the lowest level. Feynman had a good essay about this issue many years ago about rat studies[1,2].

[1]Feynman's Caltech speech:http://calteches.library.caltech.edu/51/2/CargoCult.htm [2]More recent short essay on the rat study: https://www.realclearscience.com/blog/2014/02/the_rat_experi...

I guess I consider the eye brow just the hair on the surface and maybe the skin/muscles to move it, not the whole bone structure underlying it.

You wrote: But that eyebrows exist at all is probably tied to some survival/reproductive benefit.

I disagree. Not every trait has something to do with conferring a benefit. Many traits are simply random artifacts of an exceedingly complex dynamic system. Your assertion hypothesizes otherwise. That's fine. But, it would be reasonable to expect data to back it up. In my experience, I've seen no correlation between eyebrows and the number of offspring.

As far as your second point, 3% is enormous. But, in any case, the problem is a failure to appreciate that there is a 'noise' threshold below which a probabilistic benefit will have no effect. That is, the length of my pinky may have some miniscule effect on the number of my offspring. But, if the effect is too low, it simply will not have an effect no matter how many millions of generations go by; the randomness of life will swamp it out. To put it in information theory terms, the signal would be indistinguishable from the noise.

Many people pay money to disagree with that assessment, at least according to sales of eyebrow pluckers, waxers, and makeup kits that are specifically designed to make it more attractive.

Either those people are wasting their time (which needs explanation, since over time people who waste their time and money on things that don't aid reproductive success should be selected away). If it does matters to their reproductive success, then it reflects either cultural or a deeper preference.

Personally, I have heard people comment about people who either lack eyebrows (usually through some stupid prank or accident) or have very thick eyebrows.

Not having eyebrows definitely interferes with facial signaling, and we signal quite a bit through eye contact. 'The eyes are the windows to the soul'. The eyebrows frame them.

We've arrived at a circle in this thread, considering the eyebrow example was initially offered to show how even small advantages will, over time, make some phenotype (and genotype) dominant.

This is also mathematically true: given a coin with only the most minimal bias, cumulative results of coin tosses will eventually drift far away from 50%/50%.

Concerning your empirical observation regarding the evolutionary usefulness of eyebrows: How many people have you seen being clearly impacted (in regards to their offspring) by missing toes, being born with only one kidney, having ears shaped sub-optimally for capturing sound etc...?

Because extremely little regarding your body is left to chance. Unless you're one of less than a hundred with a given feature, be assured that it's a variation that, for a significant number of prior generations, has proven to be useful.

This is what troubles me. I understand your point. But, if it takes a trillion years to 'drift away from 50/50' is it meaningful to describe the bias as something other than zero? That is, is there some Plank's Constant kinda-thing for probabilities? The reason this always comes to mind is the absurdly low probabilistic effect on fecundity of, say, a finch having a 4cm beak versus a 4.1cm beak. If the effect is some probability so low that it would take millions of generations to manifest itself, then, in light of the fact that finches were probably single-celled organisms a million generations back, the probability may as well be zero.

In that spirit, here's the next volley:

> I disagree. Not every trait has something to do with conferring a benefit. Many traits are simply random artifacts of an exceedingly complex dynamic system.

As I said previously, given how complicated of a feature eyebrows are (they have a skeletal component, hair, and muscles which contribute in complex ways to facial expressions), there is no plausible way the existence of eyebrows is due to random variation. Eyebrows would require a whole bunch of very specific mutations to occur at once and mutations are rare. The specific form that people's eyebrows take probably has a fair degree of randomness, but that eyebrows exist at all cannot be due to chance.

Further, every single person has eyebrows. If whether or not a person had eyebrows were truly of no consequence to genetic fitness, some people would have them and other wouldn't.

> But, it would be reasonable to expect data to back it up.

This argument goes both ways. As you say, I have no numbers to show that eyebrows are under selection, but you have no numbers showing that they are not. But your position is worse than that, since you are lacking a plausible explanation for how eyebrows can even exist, given how complicated they are anatomically, without having been selected for.

> In my experience, I've seen no correlation between eyebrows and the number of offspring.

That's not entirely true. After all, only people with eyebrows have children. :P I am only party kidding about that, btw. Per my previous point, the fact that everyone has eyebrows means that they contributed somehow to people having children.

But more seriously: surely you can agree that eyebrows have an impact on attractiveness. After all, a great many people, both male and female, pluck, shape, thin, or color their eyebrows. And surely you can agree that attractiveness at one point was linked to reproductive success (though I agree that it probably has a limited correlation in modern life)?

> As far as your second point, 3% is enormous. But, in any case, the problem is a failure to appreciate that there is a 'noise' threshold below which a probabilistic benefit will have no effect. That is, the length of my pinky may have some miniscule effect on the number of my offspring. But, if the effect is too low, it simply will not have an effect no matter how many millions of generations go by; the randomness of life will swamp it out. To put it in information theory terms, the signal would be indistinguishable from the noise.

If you make the change in outcome 0.3%, I'm showing that it takes 1,500 generations for for the offspring of the starting 1% subpopulation to reach 50% of the total population. That is all things being equal, etc., etc., but still not very long in evolutionary terms. The thing about evolution is that it acts on gene frequencies in populations. That allows the effectiveness of very minor changes in genetic fitness to be assayed over time. And the larger the population, the more precise the assay is.

But it also means it might not have appeared for a specific reason in humans. It might very well have just been there and useless (but not detrimental) at one point, then used for something.

They're not that prevalent, and they may not have been more prevalent in the past, sedentary life likely contributes to some form of osteoporosis by providing less continuous stress and it adds "extreme" sports as discrete high-risk events.

And of course just because it's life threatening does not mean it's going to be weeded out, selection is an important part of evolution.

The thing that brought this into full relief for me was the story of Brian Jones, a member of the Starting Strength community. He fell off of a roof and shattered both legs. Doctor told him he would never walk again, etc., etc. But he did. And then he start lifting weights. Of course, his body adapted to the increasing load he was placing on it.

Specifically, his body responded by growing muscle and making his bones more dense. His bones got so dense that the metal pins which were inserted into his leg bones after his accident started getting pushed out into the surrounding flesh and he had to have surgery to remove them (links to his story below).

I would bet good money that the majority of bone breaks (excluding things like car crashes and obvious stupidity), even in children, are a result of our largely sedentary lifestyles interacting with our body's evolved tendency to not engage in metabolically expensive activity like growing muscle and denser bone unless it receives signals indicating that they are actually demanded by our environment.

Brian Jones' story:

Video form: https://startingstrength.com/resources/forum/articles-and-pl...

Written form: https://startingstrength.com/article/barbell_training_as_reh...

About sedentary lives, I'm fairly convinced that nomadism made for stronger everything, bones, tendons, muscles (coordination too) so their lifestyle (granted you're not living near cliffs and have enough food) didn't increase the % of fracture.

Of course human minds would tend to underestimate this. Evolution harnesses the parallelism inherent in reality and atoms. Natural selection operates at all levels of detail at once.