A 1907 Experiment Determined the Human Soul Weighs 21 Grams

Photo credit: Ralf Nau - Getty Images
Photo credit: Ralf Nau - Getty Images


"Hearst Magazines and Yahoo may earn commission or revenue on some items through the links below."

I read a ton of nonfiction science books in my spare time, including dives into the most oddly specific areas of study. Mary Roach’s 2005 book, Spook: Science Tackles the Afterlife, is one of those joys. In one chapter, she details a curious experiment that set out to weigh the human soul, and I haven’t been able to stop thinking about it since.

You see, just after the moment of expiration, our corporeal form goes through a number of less-than-glamorous changes. Your muscles relax—all of them, down to your sphincter—which leads to things like free-flowing urine and feces. Your body temperature suddenly spikes as blood stops moving through the circulatory system, causing you to sweat more. Your dying cells release a cadre of enzymes that alert bacteria and fungi, leading to decomposition and the release of noxious gasses.

These things all have mass, meaning it’s perfectly natural for your body to see some humble weight loss when you die. But the religiously devout have argued that something far more precious also escapes the body when the heart is no longer beating: your soul. Unlike your sweat or waste, though, it’s not entirely clear how one measures the soul or if it even has a physical form to measure in the first place.

But that doesn’t mean no one has tried. In 1907, Dr. Duncan MacDougall, a physician from Haverhill, Massachusetts, who published a controversial paper in the journal American Medicine, claimed that he had determined the weight of the human soul to be 21 grams.

In his experiment, MacDougall chose to use six human subjects in nursing homes who were already close to death: four were dying of tuberculosis, one from diabetes, and one from unnamed causes. When their time of death approached, the patients’ beds were transferred to industrial-sized scales that were accurate to two-tenths of an ounce, or 5.6 grams. Accounting for the loss of bodily fluids and feces, MacDougall determined that one of the patients lost 21.3 grams in weight at the time of death. However, he threw out results from the other five subjects. In one case, a patient lost some weight at the moment of expiration, but promptly regained it. Two more patients experienced weight loss when they died, but put on even more weight a few minutes later. The scales were “not finely adjusted” in another example, and one patient died as MacDougall was still calibrating the equipment.

MacDougall then replicated his work with 15 dogs—which he presumably poisoned—to see if they had any weight changes at the moment of death. He found none and took this as evidence that his hypothesis was correct: that humans had souls and dogs did not.

The work is certainly attention-grabbing, and even inspired the 2003 film 21 Grams, but it’s probably not great science, according to Donald Everhart, director of institutional research at AMDA College and Conservatory of the Performing Arts, who has expertise in the sociology and philosophy of science. (In the past, Everhart has also worked as a graduate fellow at the University of California, San Diego.)

For one thing, the sample sizes in MacDougall’s experiments were far too small to be reliable, which is likely an artifact of the time period. “While this is a concern from a present-day standpoint, probability and statistics had a different relationship to scientific experimentation in 1907,” Everhart tells Popular Mechanics via email. “That is not to say that ideas regarding sampling, statistics, and comparison to normal distributions were not already in circulation at the time (for example, statistician Francis Galton’s “Vox Populi” was published a month earlier in March 1907 in Nature). In many cases, logical inference, combined with measurements that were to be as accurate as possible, were more ordinary to most scientific fields at the time than statistical inference.”

Still, MacDougall’s work could have benefited from a more scientific approach. For instance, French sociologist Émile Durkheim used official death statistics in his classic sociological work on suicide in 1897. “MacDougall could have considered the additional epistemological weight that his study might have had if he found more than one case,” Everhart explains.

And then there is the equipment used in the study, essentially a bed balanced on a scale. Everhart isn’t so sure that such an apparatus was ever appropriate for measuring something like the soul in the first place.

Perhaps even more importantly, no one has replicated MacDougall’s experiments to see if they hold up to scrutiny—a hallmark process in the scientific method. MacDougall did gesture at this in his work, stating that he was “aware that a large number of experiments would require to be made before the matter can be proved beyond any possibility of error.” Still, he seemed to believe that he had enough evidence to advance his hypothesis.

The larger question, for Everhart, is why MacDougall was so hellbent on explaining the soul as a physical entity rather than a metaphysical one. “Why, at that time, would MacDougall want to do this? What was going on with science and religion at the time that would motivate someone to conduct and publish these observations and this argument?” Everhart ponders.

If MacDougall ever read the work of French philosopher Auguste Comte, we may have a clue. Comte, who doctrinated the concept of “positivism,” believed that all phenomena could be better explained through the physical sciences, rather than through metaphysical theology, Everhart says.

“It would be quite the success for positivism if the soul, subject of millennia of theological argument, could be demonstrably studied using basic physical methods,” Everhart explains. “Maybe that's what MacDougall was after—an orderly, neat, scientific observation of the complex and ineffable.”

You Might Also Like