About the SA Blog Network

The Primate Diaries

The Primate Diaries

Notes on science, politics, and history from a primate in the human zoo.
The Primate Diaries Home

A Grimm Tale of Reproductive Conflict

The views expressed are those of the author and are not necessarily those of Scientific American.

Email   PrintPrint

A new study argues that in-law competition drove the evolution of menopause. But is the story too good to be true?

"Evil Queen" by Nathaniel Gold

"Evil Queen" by Nathaniel Gold

In the classic Scandinavian folktale “The Twelve Wild Ducks,” also known as “The Twelve Brothers” in Grimms’ Fairy Tales, a wicked old Queen is jealous of her daughter-in-law’s beauty. In a rage of envy she kidnaps her own grandchildren during the night and throws them into a snake pit. She then tells her son that his young wife is actually a witch who has eaten their children and demands that she be burned at the stake. In the end, the daughter-in-law gains the upper hand and the wicked Queen is bound between twelve wild horses who rip her body into pieces.

Such stories were widespread in Scandinavia and northern Germany during the 18th and 19th centuries. A common theme in many of them–especially the more popular ones such as “Snow White,” “Cinderella,” or “Sleeping Beauty”–is of a beautiful damsel who is persecuted by an older mother-in-law or stepmother who mourns her own lost youth. Invariably her plot fails and the evil matriarch is sent to a grisly death. Similar tales can be found in Russia, India, and Japan. For an 18th century peasant the moral could not have been more clear: intergenerational conflict leads to disaster. But could there be something deeper behind the “collective unconscious” found in so many of these tall tales?

“Evolutionary biological insights yield a powerful set of instruments with which to understand literature and, in the process, ourselves,” write David and Nanelle Barash in their book Madame Bovary’s Ovaries: A Darwinian Look at Literature. According to Barash and Barash, these intergenerational conflicts between women may be so common in our folk traditions because they are part of a common biological heritage.

Now, in a new study published last week in the journal Ecology Letters, biologist Mirkka Lahdenpera of the University of Turku in Finland and colleagues argue that this conflict between female in-laws has a very real impact on child survival. What is more, they conclude that this intergenerational reproductive competition is a leading factor behind the evolution of menopause.

By utilizing Finnish church records from between 1702 and 1908 the researchers were able to assemble a remarkable data set that detailed every birth, death, marriage, and migration in five separate communities across three generations. Their analysis revealed that when women and their mothers-in-law had children within two years of one another infant survivorship was reduced by 50% for older women and 66% for daughters-in-law. In contrast, there was no decline when women and their own mothers reproduced within the same time frame.

“Our results confirm that intergenerational conflict between in-laws can be intense,” the authors conclude, “with substantial reductions in the survival of offspring from mothers of both generations.” They explain this by citing William Hamilton’s theory of inclusive fitness, that individuals will tend to cooperate more with those whom they share more genes.

As the result of meiotic cell division in sexual reproduction a woman’s mother-in-law would be expected to share 25% of their genes with her grandchild (assuming that her son is actually the father) but would share 50% with her own children. In an environment where there are limited resources and two infant mouths to feed in the same household the struggle for existence between in-laws could become fierce. Mothers may have then found themselves playing out the same roles as those in the folktales they grew up listening to around the fireplace.

However, while this first interpretation has strong empirical support, where the story goes next becomes somewhat more fanciful. Because of the significant decline in infant survivorship that occured when in-laws had children at the same time, Lahdenpera and colleagues argue that it would therefore be adaptive for women to stop ovulating before their daughters-in-law began having children themselves. The limited reproductive overlap seen between generations in our recent past is but an echo of the stabilizing selection pressure that occurred long ago over evolutionary timescales. Menopause at around 50 years old was the ultimate reproductive compromise.

In this way Lahdenpera’s Reproductive Conflict Hypothesis of menopause is the sinister mirror image of what is currently the leading interpretation for why women have such lengthy post-reproductive lives. The Grandmother Hypothesis, developed by University of Utah anthropologist Kristen Hawkes, argues that long post-reproductive lifespans were selected for during our evolution because it allowed grandmothers to help their own children to successfully rear the next generation. Previous studies have made a strong case for the Grandmother Hypothesis, including one by Lahdenpera that appeared in Nature in 2004.

This earlier study found that the presence of either a mother or a mother-in-law resulted in significantly more grandchildren surviving to adulthood. It was, Lahdenpera wrote, “the equivalent of post-reproductive women gaining two extra grandchildren for every ten years that they survive beyond age 50.” However, eight years later, it is unclear how the same data could generate seemingly contradictory findings. For evolutionary researchers like Hawkes it mars what is an otherwise “strong and provocative result.”

“I admire the work of this research group and the ways they’ve used this Finnish data set to explore questions about human life history,” Hawkes told me. “That said, in their 2004 paper they found positive grandmother effects through both sons and daughters. What’s going on here? They say these Finns were very monogamous and that both sons and daughters either would not or could not disperse.”

This issue of dispersal and residence patterns could be a serious problem for the Reproductive Conflict Hypothesis. A sedentary farming village in 18th century Finland is far removed from the migratory lifestyle of our hunter-gatherer ancestors. While Lahdenpera and colleagues acknowledge this in the introduction, they do not seem to take the implications fully into account.

Modern hunter-gatherers, and presumably our distant ancestors as well, are often what anthropologists refer to as multilocal in which individuals migrate between their husband’s or wife’s kin depending on where they can receive the most help. It has even been shown, for example among the Hadza in Tanzania, that grandmothers will travel to where they can be most useful. This dynamic would seem to undermine support for intergenerational reproductive conflict. If conditions were too onerous with the mother-in-law young families could simply “vote with their feet.” At the very least it complicates the assumptions that Lahdenpera and colleagues included in their model.

“This is exactly why assumptions we make about behaviors that are or are not adaptive can be so problematic,” University of Illinois anthropologist and Scientific American blogger Kate Clancy told me. Settled agricultural populations might simply be subject to more conflict between the generations whereas foragers would have much less. Furthermore, unlike these Finnish communities, only 14.5% of modern preindustrial societies can be classified as monogamous (though they would still include some monogamous individuals). The idea that a young woman could be beholden to her wicked mother-in-law would certainly not be as common among our hunter-gatherer forebears as it was for the contemporaries of Jacob and Wilhelm Grimm.

“Rather than this study demonstrating adaptive value for post-reproductive life,” Clancy said, “it seems to provide more evidence that moms and their mothers-in-law had legitimately different reproductive strategies that can sometimes be in conflict.” If the grim tales of murder and deceit between generations of women do have a basis in biology it will probably be this more limited interpretation. For now, at least, it would appear to be the Grandmother Hypothesis that lives happily ever after.

Eric Michael Johnson About the Author: Eric Michael Johnson has a Master's degree in Evolutionary Anthropology focusing on great ape behavioral ecology. He is currently a doctoral student in the history of science at University of British Columbia looking at the interplay between evolutionary biology and politics.

Follow his work on Facebook and Google+. Follow on Twitter @primatediaries.

The views expressed are those of the author and are not necessarily those of Scientific American.

Rights & Permissions

Comments 10 Comments

Add Comment
  1. 1. greg_t_laden 1:18 pm 09/4/2012

    OK, but… the archaeological record of Europe tends to show rather large long term settlements from the “Mesolithic” (which was really the Early Neolithic in some areas) onwards. The classic forager dynamic of camps with mostly patrilocal but very flexible residence and conflict (like we would have with close-age reproductive in-laws) resolved by residential shifts may be more likely in both arid and very wet environments in the tropics and subtropics.

    So, we have a situation with multiple (perhaps reasonable) explanations coming from very different economic and social contexts for the same seemingly universal phenomenon.

    The question I had in reading this is this: If menopause is universal human adaptation to avoid in-law conflict among reproducing females, why were those Finnish in-laws producing this data indicating high levels of competition? Second question: Is there an infanticide explanation that would work here, and is this an outcome of residence pattern with low dispersal having more to do with lineal competition. In other words, I want to be talked out of the possibility that this is male-induced infanticide before I seriously consider the evolution of female menopause as the outcome of this kind of competition.

    And finally, were there really “snake pits” back then and why?

    Link to this
  2. 2. EricMJohnson 2:25 pm 09/4/2012

    I had the same thought when reading the paper. If menopause was an adaptation to limit conflict than why was there still so much conflict? Of course when I looked at the figures I realized that the impact might have been large but it was a relatively infrequent event. Out of 206 years of church registers there were only 31 mother-in-law/daughter-in-law overlaps and 19 infant deaths as a result (or about one every fifty years per community). So the argument was that, because of the significant impact on childhood survival when this overlap did occur, the conflicts could have been even more profound among our evolutionary relatives. However, my main problem with the argument is that an evolutionary interpretation needs to take the whole range of human variability into account. From my perspective the Grandmother Hypothesis does this while the Reproductive Conflict Hypothesis does not.

    I’m afraid there’s nothing in the paper about the possibility that these deaths were from male infanticide. The authors don’t indicate the cause of death at all, they merely interpret the heightened mortality rate during reproductive overlap as female in-law conflict. Again, I think that’s a reasonable interpretation but it doesn’t account for different residence patterns.

    The snake pit is a bit of a mystery to me as well. In the German variant of this folktale the old Queen is “put into a barrel filled with boiling oil and venomous snakes, and died an evil death.” These are deeply Christian communities, so perhaps the snakes are metaphorical.

    Link to this
  3. 3. Derick in TO 2:30 pm 09/4/2012

    This is fascinating, but it seems to miss 1 obvious point – were women even living long enough to reach menopause during our hunter-gatherer days?

    Everything I’ve read and heard (nothing scholarly – all stuff aimed at the layman) suggests that stone age people lived, on average, to about age 30. If this is true, even accounting for women typically living longer than men, surviving to menopause around age 50 would be really rare. To me this suggests that menopause itself could probably not impart much of a survival advantage.

    On the other hand, women are born with their whole compliment of eggs – they don’t generate more throughout their lives. This means that eventually, they’ll run out, and when the eggs run out the biological mechanisms involved in the menstrual cycle will have to deal with that fact. Voila: menopause.

    There’s probably a lot that goes into making those eggs in the first place – it’s likely very energy- and resource-intensive. An adaptation that makes enough eggs to last almost twice the average lifespan of a typical person minimizes the chance that a woman will run out of eggs before she dies while avoiding wasting resources on extra eggs that will almost certainly never be used (in a stone-age context, anyway).

    So maybe the adaptation in question is just the limiting of the number of eggs women are born with, with the benefit being additional energy and resources available for the development of other useful traits in a growing embryo, like a larger brain. Perhaps there is no social component at all to this adaptation.

    Link to this
  4. 4. greg_t_laden 2:56 pm 09/4/2012

    This is fascinating, but it seems to miss 1 obvious point – were women even living long enough to reach menopause during our hunter-gatherer days?

    Fortunately, hunter-gatherer days are still with us so we know this (plus from archaeology. The answer is: Yes, often enough for there to be selection for menopause.


    Today, at birth, females have a life expectancy of about 80 years (they will live on average to the age of 80). The average modern age of menopause today may be something like 52. At 52 modern women have an average live expectancy of about 30 years, or almost the same (they would live, on average, to 82)

    There is no good estimate of “forager” life expectancy. To get good actuarial data, you would need to have thousands of lifespans and there are no such data. Archaeological data suck for many reasons. But, life expectancy at birth for various non-modern populations was anywhere from 25-40, with most or perhaps all of those estimates made by people who presume it to have been short. In contrast, the average life expectancy at older ags (15-21) in studie that include this is about 39-45 or so (i.,e. with an average age of death close to modern menopause).

    If the average age of death of reproductive females (the latter in the just mentioned subset of samples, after childhood mortality is factored out) = menopause then about half (depending on how the average is counted, but assume it is the median) of women live into menopause, and perhaps conservatively estimated 35% or so live past one childbirth. This is all not considering the original grandmother hypothesis data, but looking at it independently (in my own head on a comment on a blog).

    This is a good paper for the grandmother research: Hawkes, K. (2003). “Grandmothers and the evolution of human longevity”. American Journal of Human Biology 15 (3): 380–400.

    Link to this
  5. 5. greg_t_laden 3:00 pm 09/4/2012

    On the egg thing: I think this is entirely a phylogenetic effect (referring here to Tinbergen’s four ways of looking at biological phenomena). Our very distant ancestors, the crown mammal LCA, were small fast living mousy things that were usually scarfed up by snakes or stepped on by dinosaurs long before something like menopause, or running out of eggs, could be considered. Mammals are traditionally not long-lived. Primates evolved much later but were stuck with that feature and some primates as well as elephants and some whales seem to live beyond the limit of eggs.

    The question here, then, is not why have too few eggs to keep reproducing, but rather, why not expend all the energy one has reproducing and stuff with the end-game being to die at about 80 or 90% eggs used up (to be safe) rather than going on for a decade or two (or three) beyond the last egg… why the extension of live beyond the phylogenetically constrained end of reproductive lifespan for humans.

    Link to this
  6. 6. CherryBombSim 9:36 pm 09/5/2012

    So, the Grandmother Hypothesis is now supported by evidence that they KILL their grandchildren? The irony is priceless.

    Seriously, this is pretty good data. Should be able to get something out of it. Like, you could try to figure out what the selective advantage would be if fertility lasted beyond menopause. Let’s get quantitative! Second thought is that if wives went to live with their husband’s family, their infant deaths might be correlated even if everybody helped each other out.

    Link to this
  7. 7. spocka1p 2:00 pm 09/7/2012

    I don’t see how these findings don’t contradict the same team’s 2004 paper, or the grandmother hypothesis for that matter. showing that babies suffer when inlaws breed together doesn’t invalidate the finding that babies do better when the household includes a postreproductive grandmother. The paper tested an explicit theoretical prediction (made in 2008) using a rare dataset – in other words, good science. Agreed that no-one knows ancestral dispersal patterns for sure, but handwaving about this about this doesn’t undermine the main conclusions.

    Link to this
  8. 8. EricMJohnson 11:05 pm 09/7/2012

    It’s not handwaving to critique the assumptions that form the basis of a mathematical model. The first part of this paper, the empirical evidence for in-law conflict, is excellent and an innovative use of historical data to test an evolutionary question. I also consider their 2004 paper in Nature to be a brilliant example of this. However, in my view and that of the evolutionary anthropologists I’ve spoken to about this, the mathematical model they construct is not a convincing argument for the evolution of menopause.

    Link to this
  9. 9. Tuvalu 2:36 pm 09/8/2012

    You do realize that Murdock’s classification of a small number of pre-industrial societies as monogamous is because the vast majority had some degree of polygyny, ie, women were/are monogamous, men were/are polygynous. As the women are nearly always monogamously paired across all these societies then the only difference compared to ‘monogamous’ societies (where both sexes are monogamous)is that the husband’s mother is going to have far more daughters-in-law than sons in some instances, and no daughters-in-law in others (if some sons can’t get a mate due to polygyny).

    Link to this
  10. 10. spocka1p 3:34 am 09/9/2012

    In this case the critique is about the assumption of female-biased dispersal as an ancestral pattern, on the grounds that

    Modern hunter-gatherers, and presumably our distant ancestors as well, are often what anthropologists refer to as multilocal..

    Hunter gatherers are an imperfect window into the past, so it is important to approach this question from as many angles as possible. Female-biased dispersal is more common than male biased dispersal in foragers;

    Marlowe, F. W. 2004 Marital residence among foragers.
    Curr. Anthropol. 45, 277–284. (doi:10.1086/382256

    Genetic evidence comparing variation in X and Y chromosomes suggests a history of female biased transfer;

    Sielstad, M. T., Minch, E. & Cavalli-Sforza, L. L. 1998
    Genetic evidence for a higher female migration rate in
    humans. Nat. Genet. 20, 278–280.

    Comparative analyses suggest that the LCA of chimps and humans showed strong local structuring and female-biased dispersal

    Link to this

Add a Comment
You must sign in or register as a member to submit a comment.

More from Scientific American

Email this Article