Environment: Delaying Doomsday

Of all the predictions of ecological disaster, none have sounded more persuasive—or alarming—than those put forth last year under the banner of the prestigious Club of Rome. Based on computer projections of the present rate of population and industrial growth, a team of scientists at M.I.T. forecast massive economic collapse and global epidemics by the end of the 21st century. Last month another computer specialist sharply disputed that gloomy outlook. Writing in the British publication Nature, he reported that the computer programs used by the M.I.T. group contain a simple but highly significant “typographical error” that drastically alters their doomsday projections.

The challenge comes from Chemical Engineer Thomas J. Boyle, who did his work at Arizona’s Lowell Observatory. Eager to verify the Club of Rome’s claims—which were widely publicized by Dennis Meadows in his controversial Limits to Growth—Boyle obtained copies of the original computer programs for a study of his own. The programs are, in effect, mathematical instructions or models for the computer. In the case of the M.I.T. study, they were used to show the close relationship between such variable factors as availability of cheap energy and agricultural land, birth control measures and mineral resources on the world’s overall material wellbeing.

As Boyle began translating the M.I.T. programs into the mathematical language required by Lowell’s IBM 1130 computer, one number caught his eye. It was in a sequence concerning the rate at which the M.I.T. group assumed pollution to be generated by industrial output. Boyle was surprised to find that the number was ten times as large as it should have been to be located where it was in the sequence.

Different World. In his book, Meadows, leader of the M.I.T. team, wrote that even when the most optimistic assumptions were made about the availability of mineral resources, agricultural production, cheap energy, effectiveness of population control, etc., the computer yielded the same results. It pointed to a global crisis marked by great famines, depletion of natural resources and disastrous pollution of the seas and atmosphere. Only drastic restrictions on all types of economic growth could avert the calamity, he insisted.

Not so, says Boyle who is now a faculty member at Montreal’s McGill University. Using the same assumptions as those made by the Meadows group—but correcting the pollution factor—he found that his computer runs conjured up a dramatically different world: by the year 2100, they showed, world population would stabilize at 6 billion people. Life expectancy would rise, pollution would be under control and technology would manage to forestall such crises as famine and industrial collapse.

Rejected for publication by the U.S. journal Science before Nature printed it, Boyle’s paper has already stirred up a new scientific row. Meadows, refusing to change his apocalyptic vision, vigorously denies that any typographical errors crept into the published results. He adds: “Every single conclusion that we reported has no relationship to the error purported by Boyle.” But Boyle points out that his computer runs all check out in other respects with the projections in the book—a good indication, he feels, that the program he used was the same as the one on which the book was based. Beyond this, Boyle shares the view of many computer experts that so many factors are involved in mathematical modeling on a global scale that even the best computer programs are still far too crude to make really accurate predictions.