“Is There Anything Good About Men?” Roy Baumeister asked the American Psychological Association, and he came up with a few suggestions. My post about the speech generated lots of comments and questions — and, I was glad to see, not too many knee-jerk denunciations from readers angry to see anyone suggest that gender differences aren’t due simply to oppression by patriarchal males. Dr. Baumeister, a social psychologist at Florida State University, told me he was glad to see such a stimulating discussion, and he’s written a response to the readers. So has has one of the researchers he was citing, Jason Wilder, a biologist at Williams College.
Let’s start with an assertion in the speech that troubled many readers: we have twice as many female ancestors as male ancestors. Dr. Baumeister called it the “single most underappreciated fact about gender.” Critics responded that it couldn’t be true because every child has a father and a mother. Some readers acknowledged that we might collectively have fewer male ancestors than female ancestors, but they insisted that any individual must have an equal number of males and females in his family tree. And a lot of readers demanded to see the evidence for the assertion.
Rest assured that neither Dr. Baumeister nor I believes in virgin birth. It does indeed take two to tango. But we still have more female ancestors. Before getting to Dr. Baumeister’s explanation, let’s hear from Jason Wilder, a biologist at Williams College who came up with some of the genetic evidence cited by Dr. Baumeister. (You can read a paper by Dr. Wilder here.) And here’s Dr. Wilder’s response to the comments by disbelieving Lab readers:
I’ve run into all sorts of problems when explaining our finding that the breeding sex ratio is skewed in favor of women. (The most common response: “More women have children than men? Duh, of course.”) I’ll explain very briefly the methodology of our study and how we interpret the results.
In a nutshell, we examined the amount of genetic variability on the Y chromosome (which is inherited by males solely from fathers) and mitochondrial DNA (inherited in both sexes solely from the mother). According to population genetic theory, the amount of variation observed among any set of chromosomes surveyed in a population is proportional to two factors, the rate of mutation and the size of the population (in terms of numbers of reproducing individuals). If we factor out differences in the rate of mutation, then any leftover difference in the amount of variation between two samples of chromosomes should be due to differences in the sizes of the populations from which they are sampled. Applying this method, we were able to estimate the relative size of the female and male human populations (from mitochondrial and Y chromosome variation, respectively). We found that the breeding sex ratio is about two females per male.
On average (and over evolutionary time), any given human female has been more likely to reproduce than any given male. Said another way, males have had a higher variance in reproductive success than females. As a consequence, more different females have contributed to the modern gene pool than males. Rather spectacular examples of this phenomenon have been inferred from historical times using genetic data. Asian conquerors (such as Genghis Khan and Giocangga) and their male relatives appear to have made a vastly disproportionate contribution to modern Asian populations. Niall of the Nine Hostages seems to have had a similar effect on the gene pool of the British Isles. These types of events, where one person (or set of related individuals) experiences tremendous reproductive success, can have an effect on the gene pool that lasts for many generations. On the other side of the equation, we have to infer that there are many more males than females who do not successfully reproduce at all.
So what does this mean for the number of males and females in any individual’s family tree? “I would argue,” Dr. Wilder replied, “that it is more likely that every individual has a greater number of unique female than male ancestors. I suspect that the trouble is in convincing people that their family trees do not continually bifurcate back in time. Ultimately the constraints of an historical breeding population of finite size causes reticulations in the tree. These reticulations will more often involve male than female ancestors.”
Now let’s hear from Dr. Baumeister on the questions over our ancestry:
Yes, each baby has one mother and one father, but it is nonetheless possible for combined ancestors to include more females than males. Here is a simple example. Suppose an island contains two men, Bob and James, and two women, Sally and Maria. Bob is rich and charming, while James is poor and uncouth, so both women marry Bob. James remains celibate. Soon, Sally gives birth to Doug, and Maria gives birth to Linda. Count the ancestors so far. Doug’s parents (Bob and Sally) are 50% female. Linda’s parents (Bob and Maria) are also 50% female. But added together, their parents are 67% female (Bob, Sally, and Maria).
Next, suppose Doug marries Linda and they have a baby named Max. Max himself now has more female than male ancestors: Linda, Doug, Bob, Sally, and Maria. Thus, it is possible even for one person to have a family tree that is not 50-50. This is true even though we started with equal numbers of males and females (but poor James was a dead end) and though each child has one mother and one father.
In actual life, incest taboos might have prevented Linda from marrying her half-brother, but if a couple generations had intervened, there would have been no objection. We have more female than male ancestors because of some men having multiple mates (and other men having none) and because of some mating partners having the same male ancestor.
Some of you wrote to ask for sources to look up. There was a fair amount of coverage in the popular media back around September 20, 2004 (e.g., “Ancient man spread the love around”), and you can still find those stories online or elsewhere. They explain the basic findings reasonably well. In contrast, the primary sources are quite technical to read. Look for works by Jason Wilder as first author (Nature Genetics, October 2004; Molecular Biology and Evolution, 2004) and a somewhat more accessible commentary by Mark Shriver in the European Journal of Human Genetics (2005). None of them really treats the psychological implications of the difference, focusing instead on the molecular biology of it and possible implications for demographic spread. But that’s part of what made me label the finding “the most underappreciated fact about gender.”
And as for the 80%-40% numbers, admittedly those are chosen somewhat arbitrarily. It could have been 60%-30% or 70%-35%. The only definite thing was that twice as many previously living women as men have descendants alive today. It depends a bit on how you count, especially because in the past a great many people died before adulthood (so you get higher proportions if you talk about all adults than if you talk about everyone who was born). The crucial implication was that for adult women, the odds of passing on genes were much better than for adult men, and so different strategies were needed.
Some of you wondered whether people really cared that much about having children. Were men taking risks in order to reproduce? This is a point that sometimes confuses people. Sure, there may have been many men and women who didn’t care whether they had children or even who actively wanted to avoid having children. (There still are!) The thing is, they did not leave many descendants. By definition, we are descended from people who did manage to reproduce. Maybe the risk-taking men had no thoughts of having children. But as long as they ended up having more children than the risk-avoiding men, then today’s descendants will have inherited the traits of the risk-seekers, not the risk-avoiders.
The cultural implications of this gender difference also sparked a lot of comments and questions from readers. I’ll present Dr. Baumeister’s response to them in a subsequent post.
Comments are no longer being accepted.