Because a human newborn’s brain is less than 30% of its adult size, a smaller fraction of adult brain size than in other primates, and it keeps growing and changing for decades. Fossils suggest that slow brain growth in our family tree goes back at least 3.3 million years. [Plausible] Who looked after babies before writing left almost no trace. One hypothesis says mothers had lots of help [Plausible], and some present-day communities show how widely care can be shared. From the 1950s on, scientists measured how early care affects behavior and, later, the brain. Parts of the answer are still argued over.
Labels: [Documented] the evidence shows it directly. [Strongly supported] strong sources agree. [Plausible] the evidence allows it but doesn’t prove it. [Disputed] qualified experts disagree. [Unknown] the evidence can’t answer it.
Why are human babies born so unfinished?
Newborn chimpanzees arrive with brains about 40% of adult size. Human newborns arrive with brains under 30% of adult size (Dunsworth et al., 2012). [Strongly supported for the pattern; the exact percentages come from one comparative dataset] Biologists call young that are born needing lots of care altricial. The zoologist A. Portmann called the human version “secondary altriciality.” He estimated that a human baby would need 18 to 21 months in the womb, not nine, to be born at a newborn chimp’s stage of brain development. [Documented as his estimate, as reported by Dunsworth and colleagues]
Why? The classic answer, the “obstetrical dilemma,” says a bigger head couldn’t fit through a pelvis built for walking upright. Holly Dunsworth and colleagues argue that energy sets the limit instead. On their account, birth comes when a fetus’s energy needs outrun what the mother’s body can supply. [Disputed]
Analogy: a human newborn is like a house handed over with the frame up and much of the wiring still to finish after move-in day.
How far back does a long childhood go?
A fossil child from Dikika, Ethiopia, catalogued as DIK-1-1, belonged to Australopithecus afarensis, the species of the famous “Lucy” skeleton, about 3.3 million years ago. Growth lines in its teeth put its age at death at about 2 years and 5 months. Philipp Gunz and colleagues compared this child and a second infant from Hadar, Ethiopia, with adult skulls. The brain was organized like an ape’s, they concluded, but took longer than a chimpanzee’s to reach adult size (Gunz et al., 2020). [Plausible: based on two infant fossils] They suggest this “may have been linked to a long dependence on caregivers,” or that slow growth spread a child’s energy cost across more years. [Plausible; the reason is Unknown]
Our own lineage adds a twist. Fossils from Jebel Irhoud, Morocco, about 300,000 years old, are read either as the earliest known members of the Homo sapiens line or as close to its ancestors. Their brains were already in today’s size range, but their braincases weren’t rounded. The rounded, “globular” shape reached today’s range only between about 100,000 and 35,000 years ago, and in people today it forms before birth and just after (Neubauer et al., 2018). [Documented for the shapes measured; Plausible for the timeline, from one team’s analysis of 20 fossils]
Bones record growth, not babysitting. Who did the caring 300,000 years ago can’t be read from a skull. [Unknown]
Who helped raise the babies?
Archaeology has a hard limit here: one caregiver, many caregivers, or occasional help can leave the same traces in the ground. (Our page on care for the injured meets the same limit.)
Sarah Blaffer Hrdy and Judith Burkart argue that apes with our slow life history “could not have evolved” unless allomothers, meaning male and female helpers other than the mother, helped care for and feed the young. They hypothesize this was under way by about 2 million years ago, with Homo erectus, and call their own reconstruction “inevitably speculative” (Hrdy & Burkart, 2020). [Plausible] Marmosets and tamarins, small monkeys, also share infant care widely.
Living communities show what shared care can look like, but they are people living now, not windows into the Stone Age. About 1,000 Agta live in Palanan, in northeastern Luzon, Philippines, the study reports. They fish, hunt, gather, farm a little, trade, and take wage work. After a 2013 census, researchers followed 78 young children there in 2014 (Page et al., 2021). For babies under 2, mothers were the biggest single source of care, at 25.9% of caregiver–child interactions, followed by unrelated people and more distant relatives. Fathers and siblings each accounted for about 7–8%, and grandparents for 3% or less. For children aged 2 to 5, unrelated people (23.8%) outranked mothers (19.0%). [Documented for this study; the authors call their sample small and time-limited]
Hrdy and Burkart also report that Aka infants in Central Africa, in Courtney Meehan’s research, averaged six attachment figures each. [Documented as they report it; we haven’t opened Meehan’s study] Whether human babies form multiple attachments more readily than other apes is, they write, “not known.” [Unknown]
Fatherhood and hormone measurements are also associated in one longitudinal study. Among 624 men in and around Cebu City, Philippines, those who became partnered fathers saw waking testosterone fall by a median 26% over about four and a half years, more than men who stayed single without children. Fathers doing three or more hours of childcare a day had lower levels than fathers doing none (Gettler et al., 2011). [Documented for this study; a link, not proof of cause] Fathering involves biology as well as culture.
How did scientists argue about nature, nurture and care?
In 1874, Francis Galton called “nature and nurture” a “convenient jingle” and argued that when they compete on equal terms, nature “proves the stronger” (Galton, 1874, p. 12). In 1890, Harvard psychologist William James wrote that nervous tissue has “a very extraordinary degree of plasticity,” while also claiming that by age 30 most people’s character “has set like plaster” (James, 1890, ch. IV). [Documented for what each wrote] Nature-first thinking and plasticity ran side by side.
At the World Health Organization’s request, John Bowlby, a psychiatrist at London’s Tavistock Clinic, wrote a report in 1950 on the mental health of homeless children in post-war Europe (Bretherton, 1992). Maternal Care and Mental Health (1951) said a young child needs “a warm, intimate, and continuous relationship with his mother (or permanent mother-substitute).” This is the historical position quoted and discussed by Bretherton (1992), p. 761; it is not advice from this page. Bretherton notes it was translated into 14 languages. [Documented] His formal papers on attachment theory followed in 1958–1960.
Mary Salter Ainsworth (1913–1999), born in Ohio and raised in Toronto (Marvin, 2000), joined Bowlby’s research team in London in 1950. In Uganda in the mid-1950s she ran what Bretherton calls “the first empirical study of attachment,” then, from 1963, a second study in Baltimore. With B. A. Wittig she published the Strange Situation in 1969: a 20-minute, eight-episode lab procedure in which a one-year-old’s mother leaves and returns while a stranger is present (Bretherton, 1992). [Documented] Bowlby and Ainsworth studied behavior and relationships. They didn’t measure brains.
Is attachment the same everywhere? Heidi Keller of Osnabrück University argues it isn’t. The theory, she says, grew out of Western middle-class families, by her estimate about 5% of the world’s population. She reports that most Nso children in Cameroon’s farming villages aren’t afraid when an unfamiliar woman picks them up and carries them away from their mother, and their cortisol, a stress hormone, goes down between first seeing her and being picked up (Keller, 2018). Whether attachment patterns, and one lab test for them, are universal: [Disputed]
What does early experience do to a growing brain?
The Bucharest Early Intervention Project in Romania compared a foster-care program set up for the study with care as usual for children living in institutions. Care as usual did not require a child to remain in an institution. Researchers followed brain electrical activity using EEG (Vanderwert et al., 2016, Methods). [Documented for the study design]
How we know
What evidence exists: CT scans and tooth growth lines from fossils, comparisons with chimpanzees, field observation in living communities, hormone samples, animal experiments, MRI and EEG, and reports such as Bowlby’s 1951 monograph.
Whose voices are missing: no one before writing left words about raising a child. In the 20th-century record, mothers and children mostly appear as research subjects described by experts. The Agta, Aka and Nso appear here through outside researchers. The Agta study reports informed consent and permission from Agta leaders, but our sources include nothing written by Agta, Aka or Nso people themselves. The Bucharest children appear only as measurements and group averages.
How power shaped the record: the largest brain-growth study we used (Bethlehem et al., 2022) says its data were “biased towards European and North American populations.” Keller argues that attachment theory’s picture of normal care came from Western middle-class families. And Bowlby’s mother-centered standard reached readers in 14 languages through a WHO report written for a United Nations programme for homeless children.
Then vs. now
- Then (before writing): babies were born with small brains, grew them slowly, and needed someone to keep them alive. Now: fossils show growth patterns and living communities show possibilities. Neither shows who cared for any particular child long ago. [Unknown]
- Then (1951): Bowlby’s report centered the mother. Now: field studies describe many caregivers, and a study of fathers reports hormone differences associated with caregiving.
- Then (1890): James thought character was mostly set by 30. Now: scans measure brain structure, not character, and they show no single “finished” age. A 2004 study of 13 children scanned between ages 4 and 21 found part of the prefrontal cortex still maturing “at the end of adolescence” (Gogtay et al., 2004); the sample was small, with an average IQ near 126. A 2022 study pooled 123,984 MRI scans from people aged 115 days after conception to 100 years. Grey matter volume peaked at 5.9 years, white matter at 28.7, and the brain kept changing across life (Bethlehem et al., 2022). [Strongly supported that brain structure keeps changing across life]
What we still don’t know
- Why human brain growth slowed: energy, birth, care, or some mix. [Disputed]
- When widely shared baby care began. [Plausible as a hypothesis; timing Unknown]
- How culture changes brain development, as opposed to behavior. [Unknown on the sources opened for this page]
- What any Stone Age parent or child thought or felt. [Unknown]