Cause of death best predicts how long chimpanzee mothers carry their dead infants

The sample is the published record. Reece Hammond and Thomas A. Puschel, both at Oxford's School of Anthropology and Museum Ethnography, did not follow one community for years; they pulled cases that other researchers had already written up ^[18][19]. Eighty-three cases spread over 15 sites is about five and a half per site ^[2][22]. Hammond said the spread is the point: "By bringing together a large multisite sample of ICC cases from 15 distinct sites, we can draw generalizable conclusions about the behavior at the species level. This is something that cannot be done by using cases from just one site or subspecies" ^[16]. Every entry in the pool is a bout of carrying. The data can order predictors of how long a bout lasts; the number of infant deaths where no carrying followed is not in the dataset ^[2].

The durations are lopsided. The shortest documented bout was about half an hour and the longest 126 days, a ratio of roughly 6,000 to 1 ^[3][23]. The mean of about 11 days therefore sits near the bottom of that range, and the longest case runs about eleven times it ^[4][24]. The phys.org account lists the four credible predictors and skips the effect sizes and model comparison statistics ^[26]. That leaves the direction of each association, not its magnitude.

Of the four, cause of death held up most consistently ^[11]. The explanation attached to it is about signal quality: a death from disease may be more ambiguous to a mother than a death by infanticide, so she carries longer ^[12]. That explanation assumes chimpanzees can recognize death, and the study's own summary says the capacity has not been scientifically demonstrated ^[13]. It also has to cover the mothers who carried infants well beyond mummification, a condition that probably provides a clear and persistent signal of death ^[14].

The design never measures bond strength directly. Infant age at death stands in for it, on the reasoning that older infants are more closely bonded to their mothers ^[7]. The second maternal variable is demographic: at sites with longer average intervals between births, bouts ran longer ^[8]. Nine hypotheses went in, among them the mother's age and experience, the infant's sex, and whether mothers learn the behavior by watching others; five drew no credible support ^[5][27].

The comparison works through relatedness: chimpanzees and bonobos are our closest living relatives, and most experts place the last common ancestor between 6 and 8 million years ago ^[20]. On death recognition, one of the authors put the limit plainly: "Our findings show that we don't need to assume that chimpanzee and bonobo mothers understand death as irreversible to explain how long they carry their dead infants" ^[17].