I think this is what they mean. Not offspring dying per se, but still an evolutionary pressure
There is no indication of the rate of drowning, but it would be a constant risk across all females, so no real evolutionary pressure would be evident (aside from a risk to the species as a whole). That is to say: the rate of drowning for females would have to be very low, otherwise the species would fail. If the rate was too high, you’d expect to see a counter pressure to have a higher rate of females born versus males.
The point here is that as reproductive structures evolved, the female developed equally complex structures. She might not a choice in if mating occurs, but she does have a choice in whether a particular male fertilizes her eggs. That is the evolutionary pressure… not in response to death from forced mating.
> Whatever the females were doing, they were succeeding. In ducks, only 2 to 5 percent of offspring are the result of forced encounters.
So regardless of if the mating was forced or not, female ducks are quite successful in choosing who fertilizes her eggs.
There would be quite a lot of pressure to not die from forced mating. Can't reproduce when you are dead after all.
A male that kills his mate has fewer offspring than a mate that manages to keep the female alive, doesn't he?
The risk of drowning during mating is the same for both groups. So, neither group gets a natural selection advantage from being less likely to drown, because it is a constant risk for both groups.
The key here is that the males don’t know if the successfully mated with the female (passed on their genes) or not. Only the female knows.
Seems to be the strongest evolutionary signal would be where the female has lots of offspring, whether desired or not.
So you get multidimensional optimization of a single quantity - become too aggressive and your sperm won't pass, but if you're not aggressive enough you might not eat.