“It is hard on our men held in Southern prisons not to exchange them, but it is humanity to those left in the ranks to fight our battles.” – General Ulysses S. Grant, August 18, 1864.
General Grant did not halt the exchange of Union and Confederate soldiers between the summers of 1863 and 1864, although this quotation—chiseled into a monument on the site of Camp Sumter military prison in Andersonville, Georgia—is often cited as evidence that he did.
Once the exchange stopped, prisons got more and more crowded and more and more squalid. Malnutrition and disease were rampant. And, according to a new study, the consequences lasted for generations.
Combing the records
Dora Costa is the chair of the department of economic history at UCLA and a research associate in the National Bureau of Economic Research Programs on the Development of the American Economy and on Aging. She and her colleagues have combed through birth and death records in the National Archives and compared those from US POWs held during that year when no prisoners were exchanged to those from US POWs captured earlier, when conditions were not nearly as bad and prisoners were not held for nearly as long. (Most POWs were exchanged immediately, in part because neither side really had the resources to care for them.)
This was a great way to isolate the paternal effects, because the duration of the trauma was so limited and defined, and the samples came with built-in controls. Non-POW fathers and POWs imprisoned when conditions were relatively better were used as controls for POW fathers held when conditions were at their worst. Children born into the same family before and after the war were also compared. The researchers knew the complete lifespans of the fathers and children and also had detailed information on their socioeconomic status and family structures.
Among those children born after the war who lived until they were forty-five, the sons of no-exchange-period ex-POWs were more likely to die after that mark than sons of non-POWs or exchange-period ex-POWs. There was no difference in the mortality of the sons of exchange period ex-POWs and non-POWs, although these groups of fathers were not indistinguishable; the ex-POWs had generally enlisted younger and tended to be laborers rather than farmers.
Controlling for the fathers’ postwar socioeconomic status like wealth, occupational class, and county population density did not alter the results, nor did the sons’ marital status or occupation or the mothers’ wealth or lifespan. Paternal POW status had no discernible impact on daughters’ lifespans.
The fathers’ POW status seemed to impact only the sons’ longevity, not their socioeconomic outcomes or family structures. And it was the fathers’ POW status that mattered, not their lifespan. The sons’ increased deaths were largely from cerebral hemorrhage and marginally from cancer, although this data comes only from those states that provided cause of death. There were no increases in suicides.
Because the observed effect was gender specific, because it occurred later in life (remember, all of the sons examined lived until they were at least forty-five), and because the analysis looked for but could not find any socioeconomic correlation, the researchers speculate that an epigenetic effect working through the Y chromosome is responsible.
There is some support for this idea. Something called the Överkalix study similarly examined the physiological effects of environment on inheritance as measured through historical records in northeastern Sweden. It found that a paternal grandfather’s food supply was linked to the mortality risk ratio of their sons and grandsons, but not daughters or granddaughters; the same gender specific impact was found for the paternal grandmothers’ food supply and their daughters’ and granddaughters’ mortality risk ratio. So it is not crazy to speculate that there is a nutritional component to this Civil War finding. Studies in mice also indicate that paternal malnutrition can lead to elevated serum glucose and thus hypertension—which can lead to cerebral hemorrhage—in male offspring only.
Whether this correlation between a father’s Civil War experience and his son’s mortality is in fact epigenetic cannot be corroborated, since the National Archives don’t contain DNA samples. Cultural and psychological explanations for these observations thus can’t really be ruled out. Regardless of the mechanism, though, the data does indicate that paternal trauma is transmittable and has real impacts on their sons’ health.