I didn't expect such a big response. Thanks. Y'all absolutely confirmed my suspicion that this is the kind of question that a lot of SF fans might have given thought to. Unfortunately, you are also supporting at least, if not confirming, my suspicion that there's no evidence.
To touch on a few points made:
Yes, a reasonable person would assume birth ratios would be reasonably close to 1, in the absence of evidence to the contrary. That, after all, derives mostly from the biology of spermatogenesis. Good of you to make that point explicitly though. I was taking it as axiomatic, but one can imagine environmental factors that might differently effect prenatal survival of males and females. Different survival curves by gender could start pretty early though - gender-selective infanticide could be a huge influence for example.
Survival curves for both genders would almost certainly be of the Gompertz type reflecting a daily roll of the dice - that's almost always the case with any animals in a state of nature. Certainly it means the age distribution would be far more weighted toward the young, but that's as far as I can go on the basis of theory. As y'all pointed out, that still leaves quite a range of possibilities.
Sure it would be affected by circumstance. But would it differ
much? Did guys really go out hunting dangerous game
often? I kind of doubt it personally. Any environment with big dangerous animals to hunt probably had plenty of smaller game that was a lot easier to get. Why borrow trouble? I'd think N-great grandpa probably brought home a lot more rabbits and turtles and deer, than lions and tigers and bears.
If risky behavior benefited the kinfolks, sure it would be the men that would do most of it, but I wonder if there was much of it to do before people got crowded enough to start engaging in warfare. Enough to offset men's lower risk of dying in child birth? But that's just guessing. I was hoping somebody might have run across actual evidence or convincing theoretical constraints.
Olive, you've obviously given it some thought, wondered at some of the same sociobiological implications I have, and formed a few similar suspicions (I'm particularly doubtful about extreme assumptions of male scarcity being the general rule). But
Please correct anything if wrong.
Since you asked, I'll point to one questionable item:
Men on average, die earlier than women. They are larger and larger species tend to die younger overall.
Actually, while I don't have a reference to hand, I'm pretty sure size is very strongly correlated with
longer life span across species, not shorter. Think whales, elephants, horses, dogs, mice and may flies. (It's also
independently correlated with brain size.) It's only
within certain species, or maybe some genera, and then only some of them and under some circumstances that the opposite prevails. The best example is dogs.
I think you're more on track here:
Men also take bigger risks.
And "risks" needs to include the subtle, frequent and small, not just the dramatic, rare, and large. Almost every time a new study is done on male & female longevity that corrects for additional behavioral factors not previously considered, the reported gap narrows. Women are more likely to eat better, and make choices (not just in food) based on their health merits and that alone may account for most of the differences that people think are purely biological. Of the bio factors you mentioned, while several
may be influence directly by gender, the only ones that
clearly are, are the risks of child bearing. All the rest are influenced by diet for example, and different dietary choices could explain the difference.
To expand a little on Star-child's response to Parson: Yes, talking about mean life span and particularly ignoring the huge effect of the incidence of childhood mortality on it, creates a great deal of confusion about historical changes in longevity. Age at death has probably always had a bi-modal distribution with the first peak very young and the second peak is what most of us intuitively think of as "typical" life span. That really hasn't changed very much through historic time and probably not since we became recognizably the species we are. That peak is a lot higher now because we don't have as many deaths from infected wounds, dysentery, childbirth, highwaymen, or being eaten by lions and tigers and bears. So a lot more people
get to that station that used to drop off the train prematurely. But the actual location of the peak has only moved to the right a fairly small amount. Witness, as mentioned the 3 score and 10.