Proofiness - Charles Seife [17]
Figure 4. High Internet usage means a long life.
All of these data sets are correlated with each other. Power consumption, life expectancy, fast-food consumption, garbage production—all of these are large in a rich, industrial society and small in a poorer, agricultural one. At the same time, an industrial society tends to have modern hospitals and medicine, so its citizens tend to live longer and its infants tend to die less frequently. An industrial society also produces more garbage, spends more time on the Internet, eats more fast food, and, yes, needs more power than an agricultural society. So it’s no surprise that these data are all correlated with each other—they all rise or fall together depending on how industrialized a particular society is. But this correlation is just that and nothing more; there’s no “causal” relationship between edible underwear and infant mortality and television ownership. It was silly for Holt to imply that building more power plants raises a nation’s life expectancy, just as it would be silly to suggest that we should be eating more fast food if we want to live to a healthy old age. Holt’s presentation, in fact, was a vehicle for a kind of proofiness that I like to call causuistry.
Casuistry—without the extra “u”—is the art of making a misleading argument through seemingly sound principles. Causuistry is a specialized form of casuistry where the fault in the argument comes from implying that there is a causal relationship between two things when in fact there isn’t any such linkage.
Causuistry is particularly common in health and nutrition research; you might even have altered your diet because of it. Lots of people, for example, don’t eat foods that contain the artificial sweetener NutraSweet for fear of developing brain cancer. This belief comes from a bit of causuistry perpetrated in the mid-1990s by a bunch of psychiatrists led by Washington University’s John Olney.
Figure 5. Brain tumors on the rise.16
These scientists noticed that there was an alarming rise in brain tumor rates about three to four years after NutraSweet was introduced in the market.
Aha! The psychiatrists quickly came to the obvious conclusion:
Figure 6. The higher the budget deficit, the more brain tumors there are.
NutraSweet is causing brain cancer! They published their findings in a peer-reviewed journal, the Journal of Neuropathology and Experimental Neurology, and their paper immediately grabbed headlines around the world.
But a closer look at the data shows how unconvincing the link really is. Sure, NutraSweet consumption was going up at the same time brain tumor rates were, but a lot of other things were on the rise too, such as cable TV, Sony Walkmen, Tom Cruise’s career. When Ronald Reagan took office in 1981, government spending increased just as dramatically as brain tumor rates. If you plot deficit spending—the amount the government spent beyond its budget—against rising brain tumors, yet another correlation is amazingly clear. As one increases, the other does as well. Indeed, the data seem to show that deficit spending has a tighter relationship with brain cancer than NutraSweet consumption does! However, it would be utterly ridiculous to try to publish a scientific paper linking deficit spending to brain cancer.17
The correlation between government overspending and brain cancer is just as solid as the link between NutraSweet and brain cancer. Yet Olney’s causuistry caused headlines across the world, and some people stopped drinking diet sodas out of fear. It’s particularly galling because there was pretty convincing evidence that NutraSweet wasn’t responsible for the increase in brain tumors. Olney almost certainly knew that brain tumor rates had stopped rising (they were even dropping a bit by the early 1990s) even though NutraSweet consumption was steadily increasing. This inconvenient fact didn’t deter him from publishing his paper, nor did it stop anti-NutraSweet campaigners from using it to try to get the sweetener banned. Olney