So she quietly continued her study, working behind the scenes with other researchers while she directed her own studies towards less controversial animals, such as insects. Eventually she discovered something that appalled even her with its simplicity. Not only do mothers sometimes kill their own children, they are almost never insane when they do so. On the contrary, for a mother to murder her own child is an evolutionary adaptation without which our species would not have survived some of the environmental and social disasters of the past. What's more, the actual reasoning behind this is so simple that a straightforward simple equation in four variables is sufficient to provide a reliable estimate of the probability that any particular mother will murder any particular infant: the age of the mother, whether or not this child is the gender that the mother wanted (which, itself, turns out to be easily and universally predicted based on only two variables, the mother's social status and the predicted reliability of the food supply), the child's birth weight (and to a lesser extent other indicators of long-term viability), and her estimate of whether or not attempting to nurture this particular child will only get both her and the child killed. When she took her early estimates for this equation to the 1990 conference, she discovered that epidemiologists studying SIDS, primatologists studying infanticide (following her 1976 tip), historians digging through old records to try to quantify infanticide throughout the ages, criminologists and social psychologists trying to come up with statistical models to predict mother-on-child infanticide, and anthropologists trying to statistically analyze what variables are most consistent with cultures that have high versus low rates of infanticide, had all independently discovered the same equation. And from her viewpoint as an evolutionary biologist, Hrdy demonstrates that any sane, healthy, normal, intelligent mothers who weren't capable of coldly murdering their own infant children almost certainly had no surviving descendants at all to be our ancestors during some of the species-wide threats that have been demonstrated to have happened from the fossil record and from studies of rates of genetic drift.
I mention SIDS. One of the researchers, she says, was an epidemiologist who, in the process of trying to quantify his hunch, initiated a study in which social workers and police very, very intensively interviewed and background checked a long string of crib deaths that had been explained away as unexplained random respiratory failure. It turns out that his equation was able to predict, with high (but not absolute) reliability, which infants had actually been the victims of homicide or malign neglect. If the infant was a boy when the mother wanted a girl or vice versa, if the infant was born weighing less than 8 pounds, or if the mother was in any kind of economic or physical danger if this child survived, then the baby was doomed. His final estimate, from that initial study, was that seventy five percent of all SIDS cases are actually homicides. But, he admitted, just acknowledging this possibility puts us in an awful dilemma. To catch the 3 out of 4 women whose babies suddenly die that were actually murderers, we have to treat all SIDS cases as potential homicides, therefore piling yet more heartbreak and tragedy on the 1 out of 4 who just randomly went through the worst tragedy any family can know, the sudden and unexpected death of a beloved child. Even using the predictive equation to narrow the field of homicide investigations, we'd still be casting a very scarily public accusation of homicide on an uncomfortably large number of grieving mothers.
