In response to:

A Reader's Guide to the Century from the July 15, 1999 issue

To the Editors:

Garry Wills, in his otherwise enlightening review of books about the twentieth century [NYR, July 15], makes two egregious errors in his discussion of demography. First, he says that “the normal life span” in industrial countries has grown from forty-five to seventy-five, and in nonindustrial countries from twenty-five to sixty-five. But these figures do not indicate life span, or the age at which a human body more or less wears out and we “die of old age.” Life span has always been and still is somewhere between eighty and ninety-five for most people. The figures Wills gives are those for life expectancy, which is not the normal span of life, but the average number of years a newborn person can expect to live. In 1900, this figure was approximately forty-five in the United States not because most people died at forty-five (hardly anybody did), but because the large number of deaths in infancy and early childhood balanced the deaths at older ages. Similarly, in nonindustrial countries, almost nobody died at twenty-five, but about half the people died before five, meaning many others lived into their fifties or even sixties and seventies. What has happened in the twentieth century is that infant and child deaths have been reduced drastically, so that in Japan, for example, about 97 percent of the people can now expect to live at least to age forty-five, and the average length of life has begun to approach the life span.

Second, Wills presents the astonishing figure that in 1850, Europe, with a population of 400 million, outnumbered all other regions of the world by two to one. But the population of China alone in 1850 is generally agreed to have been about 430 to 450 million….

Stevan Harrell
Center for Studies in Demography and Ecology
University of Washington
Seattle, Washington

This Issue

December 16, 1999