Sarah Blaffer Hrdy has done some of her most interesting work promoting the "grandmother hypothesis." There is a puzzle why women have menopause long before life ends. Even in prehistoric societies, there would have been lots of grandmothers. The ingenious hypothesis is that women stop bearing kids early so that they can help raise their grandchildren. Since human babies take so much longer to mature than other animals, alloparenting (nurture by other than mothers and fathers) by grandparents would be a huge help in the survival and lives of children. And of all grandparents, the mother's mother is, other things equal, most likely to invest deeply in helping her daughter with the grandchildren.
The puzzle that Hrdy addresses in Mothers and Others comes from the widely accepted finding by seminal anthropologist George Peter Murdock that most societies were patrilocal. Even if a child's mother's mother was alive and ready to help, if mother and father moved in with his family in another village, her willingness to help would be to no avail. However, Hrdy reports, when Helen Alvarez re-examined Murdock's data, she found that the situation was not so cut-and-dried. Even if a society was normally patrilocal, often the new parents would stay with her mother at first when the first grandchild was born, to learn the ropes. And in other societies, (including our own) it was common for an expectant mother to go back to her mother's house to have the baby, then come back to her husband's house some months later. Citing other studies, Hrdy also reported that in polygamous societies, if a man married sisters, their mother was likely to move near them.
It does take many helpers and many hands besides mother's and father's to raise a child. And the most useful other hands, the best alloparents, are grandmothers.