20 of the Greatest Blunders in Science in the Last 20 Years

In the last two decades, glorious scientific and technical achievements have altered our lives forever. Try, for example, to imagine the world without the existence of those two little words personal and computer. But there have also been — how can this be put delicately?— blunders.

Some were errors in concept: Bad science chasing a bad idea. Some were errors in execution: This would have worked so well if only it hadn't blown up. Others were cases of deliberate fraud, out-and-out hoaxes, or just dopey moments that made us laugh. Perhaps Albert Einstein said it best: "Two things are infinite: the universe and human stupidity, and I'm not sure about the universe." 

Surreal in its beauty, a plume of white smoke ushered in the end of America's romance with space travel after the shuttle Challenger blew up 73 seconds into its scheduled six-day flight on January 28, 1986, at 11:39:13 a.m.

The rocket was traveling at Mach 1.92 at an altitude of 46,000 feet as it incinerated all seven astronauts aboard. According to the presidential commission that investigated the accident, the explosion was caused by the failure of an O-ring seal in the joint between the two lower segments of the right-hand solid-rocket booster.

This failure permitted a jet of white-hot gases to ignite the liquid fuel of the external tank. The O-ring was known to fail in cold temperatures, but the launch had been delayed five times.

Following the "greed is good" mantra of the 1980s, some scientists could not resist shortcuts. "The psychological profile of these people is interesting," says Mario Biagioli, a professor of the history of science at Harvard University. "You usually get B-plus, A-minus scientists who get into hyperproduction mode."

Take, for example, former Harvard researcher John Darsee. In 1981 he was found to be faking data in a heart study. Eventually investigators at the National Institutes of Health discovered that data for most of his 100 published studies had been fabricated.

Or take the case of cardiac-radiology specialist Robert Slutsky, who in 1985 resigned from the University of California at San Diego School of Medicine after colleagues began to wonder how he turned out a new research article every 10 days. University investigators concluded he had altered data and lied about the methods he used. To establish verisimilitude, Slutsky often persuaded scientists more prominent than he to put their names on his articles. 

William McBride, an Australian obstetrician, was hailed as a whistle-blowing visionary in 1961 when he sounded a warning about the dangers of thalidomide, a sedative prescribed for anxiety and morning sickness. In a letter to the journal The Lancet, McBride suggested that the drug was causing infants to be born with severe limb deformities.

Although McBride's hypothesis was based on limited anecdotal observations, subsequent studies proved him right. Thalidomide was removed from the market, and the drug became almost synonymous with pharmaceutical malfeasance. Two decades later, in 1982, McBride published a report about a morning-sickness drug called Debendox that, he claimed, clearly caused birth defects in rabbits. Merrell Dow took the drug off the market amid an avalanche of lawsuits.

But there was a problem. McBride had altered data in research carried out by assistants. The results showed Debendox had no ill effects. After years of investigation, McBride was found guilty of scientific fraud in 1993 by a medical tribunal. 

In 1983, astronomer Carl Sagan coauthored an article in Science that shook the world: "Nuclear Winter: Global Consequences of Multiple Nuclear Explosions" warned that nuclear war could send a giant cloud of dust into the atmosphere that would cover the globe, blocking sunlight and invoking a climatic change similar to that which might have ended the existence of dinosaurs. Skeptical atmospheric scientists argued that Sagan's model ignored a variety of factors, including the fact that the dust would have to reach the highest levels of the atmosphere not to be dissipated by rainfall.

In a 1990 article in Science, Sagan and his original coauthors admitted that their initial temperature estimates were wrong. They concluded that an all-out nuclear war could reduce average temperatures at most by 36 degrees Fahrenheit in northern climes. The chilling effect, in other words, would be more of a nuclear autumn. 

The finding was initially trumpeted as the missing link that proved birds evolved from dinosaurs. In 1999 a fossil smuggled out of China allegedly showing a dinosaur with birdlike plumage was displayed triumphantly at the National Geographic Society and written up in the society's November magazine.

Paleontologists were abuzz. Unfortunately, like the hominid skull with an ape jaw discovered in the Piltdown quarries of England in 1912, the whole thing turned out to be a hoax. The fossil apparently was the flight of fancy of a Chinese farmer who had rigged together bird bits and a meat-eater's tail.

Shocking factoids based on half-baked interpretations of scientific data have been foisted on the public at an alarming rate during the past 20 years. Take the "spinsters beware" theme that gained currency in 1986. Summarizing a study on women and marriage by two Yale sociologists and a Harvard economist, several news agencies reported that single women at 35 had only a 5 percent chance of ever marrying, and unmarried women at 40 were "more likely to be killed by a terrorist."

Never mind the fact that in analyzing data from the 70,000 households the authors of the original study had not looked into what percentage of the over-30 women had made a conscious choice to put off marriage. Indeed, U.S. Census Bureau statistician Jeanne Moorman's follow-up projections indicate that of unmarried women ages 30 to 34, 54 percent will marry; of those ages 35 to 39, 37 percent will marry; and of those ages 40 to 44, 24 percent will marry.

At the University of Utah in 1989, chemists Stanley Pons and Martin Fleischmann announced that the world's energy problems had been solved. They claimed to have created nuclear fusion on a tabletop by electrolyzing deuterium oxide — heavy water — using electrodes made of palladium and platinum.

Deuterium is a naturally occurring stable isotope of hydrogen; its nucleus contains a neutron in addition to the single proton found in the nucleus of ordinary hydrogen. According to the chemists, the deuterium nuclei were squeezed so closely together in the palladium cathode that they fused, releasing energy.

As Robert Park, professor of physics at the University of Maryland and author of Voodoo Science puts it, "Basically if what Fleischmann and Pons said was true, they had duplicated the source of the sun's energy in a test tube."

The problem is, no other scientists have been able to reproduce their results — and not for lack of trying. "There's always some guy willing to say, 'OK, we found something that works, but it only works once in a while,' or 'We're not going to show it to you, because we're worried you'll steal our patent rights,'" says Marc Abrahams, editor of Annals of Improbable Research. 

April 26, 1986, was the day Soviet nuclear experts learned the true meaning of the word oops. During a test of one of Chernobyl's four reactors, they turned off the backup cooling system and used only eight boron-carbide rods to control the rate of fission instead of the 15 rods required as standard operating procedure.

A runaway chain reaction blew the steel and concrete lid off the reactor and created a fireball, releasing 100 times more radiation than did the Hiroshima and Nagasaki bombs combined. Some 4,300 people eventually died as a result, and more than 70,000 were permanently disabled.

The Clinton administration estimates that American taxpayers have paid $25 billion to determine that power lines don't do anything more deadly than deliver power. In 1989, Paul Brodeur published a series of articles in The New Yorker raising the possibility of a link between electromagnetic fields and cancer. Eight years later, after several enormous epidemiological studies in Canada, Britain, and the United States, the danger was completely discounted.

"All known cancer-inducing agents act by breaking chemical bonds in DNA," says Robert Park. "The amount of photon energy it takes is an ultraviolet wavelength. So any wavelength that is longer cannot break chemical bonds. Visible light does not cause cancer. Infrared light is still longer, radio waves longer still. Power-line fields are preposterous. The wavelength is in miles." 

The "better, faster, cheaper" mantra adopted by NASA in 1992 might be reinterpreted today as "you get what you pay for." In September 1999, the $125-million Mars Climate Orbiter plunged to oblivion near Mars. NASA officials were using the metric newton to guide the spacecraft.

That was unfortunate, because Lockheed-Martin engineered the Orbiter to be guided in the English units of poundals. In December, the $185-million Mars Polar Lander went AWOL, and repeated efforts to contact it by space radio antennas failed. Officials now speculate that a signaling problem in the landing legs — caused by one line of missing computer code — doomed the Lander.

In 1996, scientists at NASA declared that a 6.3-ounce rock, broken off from a Mars meteorite discovered in Antarctica in 1984, contained flecks of chemical compounds — polycyclic aromatic hydrocarbons, magnetite, and iron sulfide — that suggested the existence of bacteria on the Red Planet 3.6 billion years ago. "August 7, 1996, could go down as one of the most important dates in human history," intoned one newspaper report. But within two years the theory began to crack.

Traces of amino acids found in the rock, crucial to life, were also found in the surrounding Antarctic ice. More damning, other non-Martian rocks — rocks from the moon, where it is clear life does not exist — showed the same "evidence" of life. By November 1998 an article in Science declared "most researchers agree that the case for life on Mars is shakier than ever."

Sometimes mistakes that were made decades ago take a while to make the force of their foolishness felt. Consider the case of killer bees. In the 1950s, Brazilian geneticists crossbred mild-mannered European honeybees with their more aggressive, territorial cousins from Africa, reasoning that the Africanized bees would be better suited than their European counterparts to warmer South American climes.

They were too right. Before the aggression could be bred out of the resulting cross, the buggers got away, and some immediately headed north. In 1990 the first Africanized honeybees were discovered in Texas. Since that time they've gradually spread to New Mexico, Arizona, California, and in 1999, to Nevada. 

Indeed, antibiotics have been the Mighty Mouse of medicine. Since the discovery of penicillin in 1928, it seemed there were few bacteria that antibiotics couldn't destroy handily. At the turn of the last century, the average life expectancy was 47. Thanks partly to a decline in bacterial diseases like tuberculosis, dysentery, and gonorrhea, life expectancy in the United States has risen to 76 today.

Unfortunately, doctors did not take seriously the consequences of promiscuous antibiotic use. Physicians have long been generous in prescribing antibiotics for minor ailments, even for viral infections like the common cold. Moreover, even when antibiotics were warranted, patients were not sufficiently warned about the dangers of not taking the drugs for the full course of treatment. When the symptoms of their infection abated, patients often threw away their pills, allowing the bacteria that had not been killed off to mutate.

Now there are whole categories of antibiotics that no longer work. And there are some potentially deadly bacterial diseases, including tuberculosis, that can only be beaten by one or two of the strongest, most expensive antibiotics. 

Um, never mind. On March 12, 1998, on the front page of The New York Times, a headline read: "Asteroid Is Expected to Make a Pass Close to Earth in 2028."

Brian G. Marsden, director of the Central Bureau for Astronomical Telegrams at the Smithsonian Astrophysical Observatory, predicted that on October 26, 2028, an asteroid about a mile in diameter would come within 30,000 miles of Earth. That's within spitting distance, spacewise, which evoked comparisons to the asteroid that crashed on the Yucatàn peninsula 65 million years ago, allegedly wiping out all the dinosaurs.

"When you first discover a comet, or any kind of body, you start measuring its position," notes Robert Park. "From that you extract its trajectory. The more measurements you make, the more accurate your trajectory gets."

Marsden issued his warnings based on very early trajectory measurements. Now he anticipates the asteroid will pass Earth at a safe distance of 600,000 miles.

In 1995, it became official: Colorado students would not be tested on evolution, Charles Darwin's theory that, through an endless series of genetic mutations, we all developed from single-celled organisms. "I believe in divine creation," said Clair Orr, Colorado's chairman of the state's board of education.

Colorado is not alone. Kansas removed evolutionary theory from its tests in 1999. Mississippi and Tennessee do not teach the subject at all, and curricula in Florida and South Carolina touch on it only lightly.

Given the trend of treating all theories of how we got here as equal, Marc Abrahams, of Annals of Improbable Research, has a suggestion: Why not teach the theory of Chonosuke Okamura, a Japanese paleontologist who became convinced that patterns of water seepage in rocks were "mini-fossils" and that life was descended from mini-horses, mini-cows, and mini-dragons. "It's kind of like forming an evolutionary theory out of cloud formations," says Abrahams. 

In the early 1990s Michael Weintraub, a researcher at the University of Rochester, concluded that a combination of two nonaddictive drugs that had been around for years — phentermine, a stimulant, and fenfluramine, an appetite suppressant — could be used for the long-term control of obesity. The fen-phen diet craze was born.

Physicians began giving the drug combination off-label to patients who wanted to lose as little as 10 to 15 pounds. In the meantime, an August 1996 report in The New England Journal of Medicine linked the use of fen-phen for more than three months to a 23-fold increased risk of developing primary pulmonary hypertension, a fatal lung disorder.

Subsequent studies revealed that prolonged use of fenfluramine could cause heart-valve defects. By September 1997, the Food and Drug Administration signaled the demise of fen-phen by ordering that fenfluramine be taken off the market. It is estimated that between 1.2 million and 4.7 million Americans were exposed to the drug combination. 

It was intended to solve a pollution problem. Instead, it may be the cause of one of the most serious pollution problems of our time. MTBE, or methyl tertiary butyl ether, is a gasoline additive that came into use in the late 1970s during the phaseout of alkyl lead additives. It helps gasoline burn more efficiently and cuts down on air pollutants.

It also happens to be highly water-soluble and has a nasty tendency to leak from underground storage tanks at gas stations. In California, MTBE contamination has forced water suppliers to shut down wells in many counties.

A recent study by the U.S. Geological Survey found MTBE in 14 percent of all urban drinking water wells it sampled. In March 1999, the Clinton administration announced a ban on the additive. Meanwhile, there appears to be no cost-effective way to remove it from drinking water. 

The award for "Most Expensive Piece of Immediately Obsolete Technology" goes to Iridium, a communications company that 10 years ago promised crystal-clear cellular phone service anywhere on the planet. Sixty-six satellites were launched at a cost of $5 billion.

"The phones were bulky. They cost $3,000. A call cost several dollars per minute, and the system didn't work indoors," says Richard Kadrey, a founder of Dead Media Project, a Web-site collection of failed media and technology. "Most people simply don't need to call Dakar at a moment's notice. In fact, the number of people who do is so small that it is probably dwarfed by the number of people who really need to talk to aliens."

In 1999, Iridium took its place among the 20 largest bankruptcies in history. 

Pamela Anderson had them taken out. So did Jenny Jones. They needn't have bothered, according to an independent panel of medical experts. Never mind that lawsuits over the implants bankrupted Dow Corning, a multibillion-dollar company. The medical panel reported in 1998 that there is no greater incidence of immune-system abnormalities among women with breast implants than there is in the general population.

In the end the science didn't fail us; the lawyers did.

It all got fixed before it could happen, but at a cost of $100 billion. Thanks to purposeful programming, computers were likely to read the year code "00" as 1900 instead of 2000. So we were treated to an entire year of talking heads ranting on about doomsday scenarios, including a world where airplanes would drop out of the sky and banks would register your portfolio value as zero. And some people don't have to buy canned goods for at least a year. All we can say is: Thank you, Bill Gates. 

More Resources: Invasive species bugging you? See a report by the U.S. Geological Survey's Biological Resources Division at biology.usgs.gov/ s+t/ SNT/ noframe/ ns112.htm. For further information on the Chernobyl disaster, visit the Uranium Institute. The official NASA site dedicated to the Challenger disaster is www.hq.nasa.gov/ office/pao/ History/sts51l.html.