Support Provided ByLearn More

The Evolution of Motherhood

Sometime after about two million years ago, somewhere in Africa, mothers in a certain group of apes—our ancestors—began rearing their young in a new way. Unlike chimpanzees and gorillas today, these ape mothers allowed others to help in the rearing of their infants. This seemingly modest behavioral change, argues anthropologist Sarah Blaffer Hrdy in Mothers and Others: The Evolutionary Origins of Mutual Understanding (Harvard, 2009), had a major impact on the evolution of modern humans. In this conversation with Graham Townsley, producer of "Becoming Human" and a trained anthropologist himself, Hrdy explains why she believes this change led to the evolution of some of the signature attributes of humans, including empathy, cooperation, even our large brains.

Sarah Hrdy
Sarah Hrdy, whom E. O. Wilson has called "the leading scientific authority on motherhood," says that the roots of our ability to cooperate with one another lay not in war-making, as sociobiologists have long argued, but in the unique way that early modern humans reared their children.

Receive emails about upcoming NOVA programs and related content, as well as featured reporting about current events through a science lens.

Photo credit: Courtesy Sarah Hrdy


Support Provided ByLearn More

NOVA: What's the really important thing we should know about the transition from earlier ancestors to Homo, our own genus, about two million years ago?

Sarah Hrdy: Well, no other mammal depended on so many others so much for so long as human youngsters growing up. It takes on the order of 13 million calories to rear human youngsters from birth to nutritional independence when they can produce what they need for themselves without a lot of subsidies. This is so much more than a mother by herself could provide, especially because she's likely to give birth again to an even more dependent child. So like human mothers today, Homo erectus mothers needed an awful lot of help.

Which, as traditionally argued, came from her mate.

A lot of our hypothesizing about early hominins started with some version of the Man-the-Hunter/Sex Contract model. So when the

Australopithecus afarensis "Lucy" was discovered [she lived about three million years ago], it was simply assumed that, oh yes, she needed help, so it must have been her mate who provided it. In a very influential article on the "Origin of Man," the paleontologist Owen Lovejoy argued that men became bipedal so that they could carry prey back to a mother and her young waiting back at camp.

The difficulty of keeping offspring alive cannot be overestimated.

Well, there are a lot of reasons why this just doesn't work well. First of all, mothers weren't waiting at camp. They were foraging and gathering just as chimps do and as all hunter-gatherer mothers do. But also we have to think more about fathers and what it means to be a hominin father. [Hominin refers to all species that are human or human ancestors, including early humans such as Homo erectus as well as earlier ancestors such as Australopithecus. See Who's Who in Human Evolution.]

Our best proxy organism there is human fathers, especially those still living as hunter-gatherers. And while we know that human fathers can be terribly committed to their offspring and can be critical for their survival, we also know that some fathers die, defect, or simply prove inadequate in terms of how much food they are able to provide. After all, hunting is a dicey occupation. So to depend on a father alone to provide regular meals for offspring is just not practical.

So how could mothers be producing these incredibly needy offspring? I mean, Mother Nature—my metaphor for Darwinian natural selection—does not ordinarily set reproducers up to fail. This is not how natural selection operates. Mothers must have been producing these costly offspring with the expectation that should help from fathers not be enough, there might be recourse to help from others as well.

old woman holding baby
In modern hunter-gatherer groups such as the Bushmen of the Kalahari, grandmothers and other relatives help mothers in child-rearing.
Photo credit: © Anthony Bannister; Gallo Images/Corbis


Hence the title of your book—Mothers and Others.

Yes. Just in the last 10 years, as we've started to actually look at what it takes in a hunter-gatherer context to keep children alive, what we're learning highlights the importance not just of parents but also of alloparents—group members other than the genetic mother and father. Grandmothers, aunts, uncles, cousins, and older siblings.

Others were needed because survival, for mothers and their infants, was just a lot harder two million years ago.

Yes. We have to keep in mind that we're not talking about a modern context. When someone like me gives birth, I have good reason to expect that my children are going to survive. Ninety-eight percent or more of children born in a western industrialized nation survive to maturity. This was not true for our ancestors. Nor is it true for other primates under natural conditions. Average mortality rates of 40, 50, 60 percent of youngsters dying before they mature are common. The difficulty of keeping offspring alive cannot be overestimated. It was just really rough.

One reason, I think, that hominin populations remained so low for so long, so steady throughout so much of the Pleistocene [the epoch dated to roughly 1.8 million-12,000 years ago], was that these individuals had trouble replacing themselves. There's no indication until maybe around 20,000 years ago that these populations were starting to grow. Once they start to grow, though, then in record time we circled the globe and hit six billion. But before that it was really touch-and-go for a long time. So I think we need to pay much more attention to how these people managed to replace themselves.

Are humans the only primates that have this alloparenting?

Well, about 40 to 50 percent of all primate species have some form of shared care. Only maybe 20 percent have both shared care and shared provisioning, and, besides us, the only one with a great deal of alloparental provisioning are in the subfamily Callitrichidae, composed of the marmosets and tamarins.

gorilla mom & infant
In gorillas and all other great apes except us, the mother is the exclusive caregiver of her infant.
Photo credit: © Jason Nichols/WGBH Educational Foundation

But we're the only great ape that has shared care of any kind. In chimpanzees, orangutans, and gorillas, the mother cares exclusively for her infant. For example, a mother orangutan will not allow any other individual to take her infant. She will be in constant skin-to-skin contact with that baby for at least the first six months of life, not a moment out of contact, and that baby's going to nurse for as long as seven years. This is a very single-minded, dedicated, exclusive kind of care. Unfortunately, poachers know this and are very aware that the only way to capture, say, an infant chimpanzee to sell in the market is to first kill the mother.

Human mothers are very different, in that in every traditional society for which we have information, mothers do allow others to hold and help them carry and care for their infants shortly after birth. This is a really major difference, although until now not much was made of it.


This difference between us and chimps might come as a surprise to many people, considering we're so closely related, right?

Yes, it does. People assume that because in our closest relatives, like chimpanzees and bonobos, the mother is the exclusive caretaker of her infant, that this must have been true of our ancestors as well. Indeed, the most influential models of infant care, such as Attachment Theory [which holds that a child needs a secure relationship with a single adult caregiver for normal social and emotional development to occur], came out of this assumption that the environment of evolutionary adaptiveness for humans was mother-only care. If you go to the American Museum of Natural History and you look at those wonderful panoramas—panoramas that I love—you'll see the mother hominin holding her infant, and her mate will be there, maybe with his arm draped around her shoulder. But she's the one doing all the caretaking.

Babies learned to read the emotions and attentions of others from an early, early age.

But if you study hunter-gatherers still living that way today, or if you look at ethnographies from the middle of the 20th century when people were still living by hunting and gathering—perhaps not exactly the same as our ancestors two million years ago but nevertheless our best proxies for how people in those environments kept their children alive—if you look at the ethnographic evidence, care is provided by a range of group members, not just the mother.

But let me be clear. I don't think anyone would question that the mother is a central figure or that an infant becomes powerfully attached to the nurturing mother who suckles him. But the point is there are these allomothers in addition to the mother who, from a very early age, even before the infant is weaned, are kiss-feeding, providing sweetened saliva and little pre-masticated treats, responding to the infant, holding him and so forth so the mother is free to forage and perform other vital activities.

orangutan mom & infant
An orangutan mother cannot rely on help from other orangutans in rearing her young, which is one reason the interval between one baby and the next is so long.
Photo credit: © Eric Gevaert/istockphoto

How important was this to the evolution of Homo, to our evolution?

I believe it was critically important. Human infants are born larger and even more helpless than any of the other apes, and yet when you compare the birth intervals in modern human hunter-gatherers with the birth intervals in any of the other great apes, you find that babies are weaned sooner and their mothers are breeding at a much faster pace than great apes. At the extreme end, the interbirth interval in an orangutan is seven to eight years, maybe five or six years in a chimpanzee. By contrast, the interbirth interval for human hunter-gatherers is closer to two or three years, suggesting that help from others, especially the supplementary feeding that allomothers provide, is allowing mothers to breed at a faster pace.

With their children buffered from starvation this way, populations could also persist even in the face of very challenging environments, and also migrate into novel habitats, and, of course, eventually out of Africa and around the globe.


So how would you sum up the human family two million years ago?

If I had to sum up the human family in the Pleistocene, here's what I would stress: Family composition was flexible, and people were strategic. They moved away from adversity and towards opportunity, and the opportunities they moved towards were not just water resources, better food resources, more game, but also the alloparental resources. They counted on these others to help them to care for and provision their young and look out for them, even in the face of climate fluctuation and tremendously challenging environmental conditions. These families were flexible and dynamic, and I think this flexibility was the hallmark of the human family.

Can you talk about the evolution of the mother/infant bond? You've described it in a chimpanzee world. What is it that you imagine starting to happen with these early kinds of Homo?

One of the most striking discoveries about chimpanzees in recent years has been how alert newborn babies are. They are able to scan the faces of others and even imitate them, just as we know human babies can do. But the difference is that in the context of chimpanzee social life, that baby is going to be primarily focused on its mother and can be relatively confident that as long as its mother is alive, she's going to care for it to the extent of her ability.

Human infants, human apes, were born in a totally different developmental context. Their mother's commitment was more contingent, as is also the case in other cooperatively breeding primates. Babies were off their mother more, needing to constantly monitor their mother's whereabouts as well as her emotional state and commitment, and they also needed to monitor others and their level of commitment and intentions. Who was likely to be worth soliciting something from, and who not? By having multiple attachment figures, even with the mother the central figure, they learned to read the emotions and attentions of others from an early, early age, creating a very different kind of ape phenotype. [Editor's note: Phenotype refers to an organism's visible characteristics and traits produced by the interaction of its genes with its environment.]

So I'm envisioning for Homo erectus a two-stage process. The development of the infant in a context with multiple caretakers producing a different phenotype, a little offspring conditioned to pay more attention to and monitor the intentions of others more—that's the first step. And then in the second step, natural selection favors those individuals who are a little bit better at doing this, because they would be the little infants best cared for, best fed, and most likely to survive. So a different developmental process producing a novel ape phenotype, and then a novel selection pressure actually favoring the little apes better at things like theorizing about the minds of others.

chimpanzee mom & infant
Human babies, Hrdy argues, have to more aware of what other individuals besides their mothers are doing and even thinking than do chimp babies, who need only focus on their mothers.
Photo credit: © David Parsons/istockphoto


You mean "theory of mind." What exactly is theory of mind, and why is it so important in our evolution?

Theory of mind is being able to attribute a mental state to someone else. It's very interesting that the first studies of theory of mind were actually done in great apes. People trying to decide if one ape knew what another knew about where some food was hidden. As psychiatrists have developed this concept, they focused more on what I would call "intersubjectivity."

In a way this is a more useful concept for thinking about early humans, because we want to know when people not only attributed intentions and mental states to others, but also began to care about what they thought and care about what that individual thought about them—pride and shame. These are very human emotions, closely tied into empathy, so it's very hard to come up with separate definitions for theory of mind, intersubjectivity, empathy. These are all related, and all very much a part of what we think it means to be human and what I think of as emotionally modern.

When did humans develop this gift for attributing mental states and feelings to others and for caring about what others thought about them? [These are] absolutely critical underpinnings for later developments like the spectacular capacity of humans to cooperate. This is a really important sociocognitive difference between humans and other apes. It's because we can read the intentions of others and we know what someone else is trying to do that we can work together with them to cooperate.

Big brains need caring much more than caring requires big brains.

This challenges the leading theory of how cooperation developed, doesn't it?

At present, kind of the reigning theory—indeed, it's just about the only theory you hear about for how humans develop this capacity to cooperate—is the idea that there were these competing groups, and the only way they could survive would be if they cooperated to, for example, kill off their neighbors.

Well, maybe. It could be right. But it also could be wrong. This is why it strikes me as such a bad idea to have just one theory to explain something as important as the underpinnings of cooperation. So that is one reason why I have spent so much time exploring alternative hypotheses for the origins of human intersubjectivity and of this spectacular capacity humans have to cooperate with others, even strangers.


What is the scenario in which this makes evolutionary good sense? A huge amount of effort goes into looking after these things that just never grow up. Why is that a good idea evolutionarily?

Well, taking a long time growing up has real drawbacks, because you could die before you get a chance to breed. There are plenty of chances to starve in there. On the other hand, taking a long time to grow up means that if food is short, you could just slow down for a while and grow a little less, take your time, and have a catch-up period when food becomes more abundant. You can build a stronger system, stronger body, stronger immune system.

And then, I think, there was a special bonus from taking so long. A number of anthropologists argue that the reason human childhoods are so long is because we need that period to grow our big brains in order to fill it with the useful cultural and linguistic information that humans need to succeed in life.

human & chimp brains
Did a bigger brain or a longer childhood come first in early modern humans? Hrdy holds that a longer childhood enabled our big brains to evolve. Here, human and chimpanzee brains.
Photo credit: © Anastasia Cronin/WGBH Educational Foundation

Maybe so, but I have wondered if what really happened wasn't that because human childhoods were so long—and we know, by the way, that among cooperatively breeding animals generally, you have longer periods of post-weaning dependency, or what in birds is post-fledging dependency, and in humans what we call childhood—that maybe having a longer childhood to start with meant that humans could have the luxury of evolving a bigger brain.

One of the drawbacks about big brains is that they are so costly. And if you had to take longer to grow up so that you could grow a big brain, I'm not sure that they would ever pay off, because some just slightly faster-growing but stupider lout could breed successfully before you did. But if childhoods were already long to begin with, then you could have selection favoring larger brains.

So you think a longer childhood came before the big brain?

I do. I think longer childhoods came as a by-product of being a cooperative breeder and that because we had longer childhoods, there was a wonderful opportunity for big brains to evolve—even though they only had a slight advantage initially, because the thing about brains is, growth is incremental, and the benefits might not show up from having a little bit bigger brain, especially when that other lout was going to outbreed you.

In evolutionary terms, offspring survival is where the rubber hits the road.

So this type of emotional sophistication does not need a big brain?

In my view, big brains need caring much more than caring requires big brains.

That is a beautiful phrase.

When you look across cooperative breeders, marmosets and tamarins, for example, these guys have tiny brains, brains no bigger than a walnut. And yet they are very caring and sensitive to the needs of others and, in that sense, pre-adapted for social learning. You probably know that when one macaque or chimpanzee stares into the eyes of another, this is a sign of aggression, right? Marmosets are the only primate besides humans where you can stare into somebody else's eyes and it's not taken as aggression. It's a way of communicating.

So what I'm arguing in terms of, say, humans and marmosets is that if you take an ape with the cognitive sophistication that we know that chimpanzees and orangutans have—a very good sense of causality, quantitative skills that rival [those of] a two-and-a-half-year-old human child, the ability to make tools, rudimentary theory of mind, rudimentary empathy—and then you take that ape and you introduce a cooperative-breeding child-rearing system such as you have in marmosets that encourages increased emphasis on sharing, sharing not just food but also information, you have a novel combination of the cognitive expertise of an ape with the giving impulses of a cooperative breeder—that is what I think was happening to humans at the beginning of the Pleistocene.

baby marmoset
"Marmosets are the only primate besides humans where you can stare into somebody else's eyes and it's not taken as aggression," says Hrdy. Here, a young pygmy marmoset.
Photo credit: © Eric Isselee/istockphoto

So it's as if the ever-growing brain was an accidental by-product.

It's an opportunistic windfall, but one our species subsequently made the most of. Clearly our entire 1,300-centimeter brain didn't evolve just as a bonus from cooperative breeding. Lots else was going on that favored intelligent individuals. But what I'm saying is that long childhoods, this period when youngsters were buffered from starvation by alloparental as well as parental provisioning, provided a novel opportunity for brains to get bigger.

There was just so much emphasis on hunting and warfare that we overlooked how important child-rearing was. After all, in evolutionary terms, offspring survival is where the rubber hits the road.

So we've kind of missed the point?

So much of the literature on human evolution stresses hunting and intergroup fighting and mate choice and competition for mates that we tend to forget that unless the offspring conceived survive, all these activities are just so much sound and fury signifying nothing.

National corporate funding for NOVA is provided by Draper. Major funding for NOVA is provided by the David H. Koch Fund for Science, the NOVA Science Trust, the Corporation for Public Broadcasting, and PBS viewers.