The best theory we have for the evolution of aging says that bearing children should shorten your life span. The best theory is wrong.
Current Evolutionary Theory
Here’s the surprise from genetic research in the 1990s that changed the way evolutionary scientists think about aging: Aging takes place under control of an intricate regulatory system, ultimately governed by genes. What is more, some of these genes have been conserved over a huge range of life forms, dating back to the dawn of multi-cellular life. No one had anticipated this, but once it was established, there could be no more talk of aging as a passive process, the body “wearing out” like an old shoe or a rusty car.
Then the question arose, what keeps these genes in the genome? Why have they not been eliminated by natural selection long ago? Darwinian evolution is supposed to select only what’s good for you, and aging is bad for you.
So evolutionary scientists hypothesized that aging is caused by “tradeoffs”. In order to maximize fitness, natural selection has made a sort of devil’s bargain, accepting deterioration and death in return for enhanced fitness. In the most prominent theory of our day, enhanced fitness comes in the form of fertility.
This theory – called “Antagonistic Pleiotropy” – predicts that fertility and longevity are genetically and metabolically linked on a see-saw, so that more of one means less of the other. The most direct and important consequence of the theory is that having children ought to accelerate aging.
Testing predictions of the theory
Scientists have looked and looked for this effect, to no avail. They’ve studied zoo animals and lab animals, modern medical records, and historical records going back hundreds of years. All these studies have come up negative. Everywhere they look, zoologists and demographers have all found there is a small positive correlation between fertility and longevity.
There is one exception. For a woman, giving birth after the age of 40 shows a large positive correlation. Bearing children after age 40 adds 3 years to a woman’s life expectancy.
From the lab:
In the UC Irvine laboratory of Michael Rose, fruitflies were artificially selected for increased longevity. Year after year, their life span increased from 2 week to 3, then 4 and beyond. Theory said that as life span increased, fertility would have to go south. But in fact, Rose found that fertility was actually rising with longevity. (He has explained the result as an artifact of inadvertent selection for fertility. I say that no matter how the experiment was conducted, you’re not supposed to be able to simultaneously increase fertility and longevity. His results contradict the Antagonistic Pleiotropy theory.)
From animals in captivity:
Unlike animals in the wild, zoo animals are protected from disease, famine and starvation, so they typically live long enough to die of old age. Robert Ricklefs solicited zoo records from around the world and analyzed 18 species of mammals and 12 birds. He sliced and diced the data 68 different ways, and the only significant correlations he found were positive. He concludes that there is no accelerated aging from laying eggs or bearing young.
From historic studies of humans:
In a historic study of French Canadian women in the 17th and 18th centuries, a small positive association was found between the number of children they had and their age at death. A historic study of Amish women found the same thing. A 200-year Finnish study found a positive correlation and here is another historic study from a German database.
Natalia Gavrilova and her husband Leonid Gavrilov have been collecting and analyzing data on aging at University of Chicago for many years. Their book chapter summarizes these and other historic studies. To be fair, a lot of women died in childbirth before the 20th century, and all these studies have had to separate the immediate risk of fatal complications from any long-term effect on the rate of aging.
The remarkable thing I find in reviewing these papers is how consistent they are, and yet each one reports the results as if it were a unique surprise. It seems that no matter how often countervailing data are confirmed, expectations are still controlled by the theory.
Contemporary studies of humans:
This recent study of Norwegian women found a positive relationship between fertility and longevity, and reported it as though it were a departure from expectation.
Just last month, a new study of 16,000 Swedish twins reports that childless men and women lived longer than their twins who had children. The three authors from Univ of Southern California mince no words in announcing that their results contradict the pleiotropy theories.
And most amusing is the “dirty old man” study, published with a stiff upper lip in the British Medical Journal a few years ago. Sexual activity for men was found to be positively associated with longevity.
The way science works, and the way science is supposed to work
The British scientist who originated one of the tradeoff theories himself did a historic study of British aristocracy, with records going back to the 11th century. This was the only study to report a confirmation of the theory. For a small subset of the women, using an inappropriate statistical test, he was able to force the data to corroborate his expectation. The study has been severely criticized by others besides me. But the funny thing is that this positive study was reported prominently in Nature, and it has been cited more times than all the negative studies I’ve mentioned combined.
In a healthy scientific community, theories that make the wrong prediction time after time are discredited. But this can take a generation or more. In the meanwhile, the “best theory we have” about the evolution of aging is not much use to us. Still, literature in the field continues to interpret new results in the context of Antagonistic Pleiotropy. It was Max Planck who warned us, “science progresses funeral by funeral.”
One result that stands out
Thomas Perls of Harvard Medical School recruits centenarians and has compiled a database of their genetics, their relatives, and their life styles. The earliest and most striking thing he found was that women who live to be 100 are 4 times more likely to have given birth to a child after age 40 than other women born in that time period. Because of the theory, he was reluctant to argue for a causal effect. He tried to explain the correlation qualitatively by saying that women who are still fertile at 40 are more likely to have the longevity genes that enable them to live to 100. I have argued that this explanation doesn’t work quantitatively, because there are too many women who are fertile at 40 and too few centenarians. My interpretation also agrees with the French Canadian study mentioned above. I believe that having a child when you are past 40 actually increases your life span, and I calculate that the benefit is more than 3 years.