October 5 may not be as famous as September 11, but it may prove more historic and seminal. On that day last year the United States suffered its first death ever from a biological warfare attack. Over the following two months, 21 additional people were infected with anthrax, and four of them died. We don’t know who planned the attack or who perpetrated it, where they obtained the anthrax or why it was done. The delivery weapon was the U.S. mail, although even that isn’t certain for all the deaths. The World Trade Center attacks are clear and well-understood, compared to the anthrax mystery.
Although the evidence remains circumstantial, most experts continue to believe that the anthrax terrorist was a disgruntled U.S. citizen, working alone, trying to frighten and kill, or perhaps to probe our biological warfare defenses. Much of this theory is based on handwriting analysis of the anthrax letters, along with reports that the anthrax was the American Ames strain, apparently refined for military use. Barbara Rosenberg of the Federation of American Scientists gives an extensive review of the data (see the FAS website), and argues that the FBI knows the identity of the perpetrator but is keeping it secret, perhaps to protect classified programs.
I disagree with these experts. Judging from such factors as the timing of the anthrax mailings, the delivery method, the quantity of spores used and the information that was publicly available about anthrax’s lethality, I think it likely that the anthrax terrorists were working for Osama bin Laden, and intended to murder thousands of people. In other words, the letters were the main salient of the “second wave” of al Qaeda terrorism. Many political questions cloud the issuefor example, why bin Laden would want to target Senate Democrats. I’d like to set those aside for a moment and explain the scientific reasoning behind my view. My hypothesis may not be widely shared, but unless we consider it seriously, we risk overlooking many productive trails toward a solution of the anthrax mystery.
First, look at the delivery method. A study posted on the Web by the Defence Research Establishment Suffield in Alberta, Canada on September 1well before the anthrax letters were mailed-suggested that envelope-borne anthrax spores could be aerosolized very effectively by the simple act of opening the mail. The report stated that anthrax dispersal from letters was “far more effective than initially suspected”; greater than 99% of the respirable aerosol particles in an envelope were released into the air when test envelopes were opened. (Steven Block of Stanford alerted me to this site.) The report concludes that lethal doses can spread rapidly throughout a room where an anthrax-laden envelope is opened. Any terrorist checking the Web in early September might have found this report and decided to act on it.
Next, consider the amounts of anthrax used. Other data in the public domain suggested that even a few grams of anthrax could, if dispersed with perfect efficiency, kill millions of people. Any terrorist who put this information together with the Canadian study might have concluded that the post was an ideal way to kill a building-full or even a city-full of civilians. If this is true, then the attack was not to be a demonstration; it was not planned to disrupt the mail, or even the U.S. economy. It was intended to commit mass murder, including United States leaders and media personalities.
If mass anthrax deaths were part of the perpetrators’ plan, how did it go so wrong? I suspect that the terrorists were influenced by the misleading technical concept of “lethal dose.” Consider the following paradox: Senator Patrick Leahy, after a briefing on the possible contents of the letter sent to him, announced on Meet the Press that it might contain “100,000 lethal doses.” Yet only five people died from all the letters. Was Leahy exaggerating? No. He was being conservative.
How can we reconcile five with 100,000? Based on primate experiments, the Defense Intelligence Agency estimates that 2,500 to 55,000 spores are enough to trigger fatal pulmonary anthrax infections in half of those exposed (the dose epidemiologists call “LD50”). It is possible that any one spore can trigger the disease, but the probability is low, so many are required, on average. Ninety-four-year-old Ottilie Lundgren, the fifth and last victim, may have been killed by just a few spores. That would explain the absence of detectable anthrax in her home and belongings.
To penetrate into the most sensitive areas of the lungs, the spores or clumps of spores must be small, with diameter not much larger than three microns and a weight of about 10 picograms. Leahy’s letter was reported to contain two grams of finely divided anthrax, 200 billion such particles. If we assume 10,000 particles is a reasonable average for LD50, then the letter contained 20 million lethal doses. So Leahy’s estimate of 100,000 was actually low.
In the worst-case scenario (or the best-case-from the terrorist’s point of view), the anthrax spores sprinkle out of an envelope, disperse like dust, are swept up into a building’s ventilation system and get mixed and uniformly diluted in the recirculating air. A human breathes about a cubic meter of air every hour. With 10,000 particles in each cubic meter, two hundred billion particles from one letter could (in principle) contaminate 20 million cubic meters, almost the volume of the entire New York City subway system.
This worst-case scenario, however, is highly misleading. The primary challenge in the military use of anthrax has always been to find methods to mix the spores thoroughly with the air, and keep them there long enough to be breathed. Most dispersal methods are extremely inefficient. Lethal doses, per se, aren’t meaningful.
I suspect the terrorists didn’t appreciate this. In my scenario, they had managed to obtain a few grams of spores, perhaps stolen from a U.S. research facility. They correctly estimated that they had several hundred million lethal doses. Even at only 1% efficiency (a conservative estimate, they mistakenly thought), they could kill 2 million Americans. Of course, the lethality might be limited to one building, and maybe some surrounding area, so only thousands would dieor only hundreds, if they were very unlucky.
But their initial anthrax attack was a disappointing failure; only one person died, Robert Stevens, a photo editor at the Sun, a tabloid newspaper. The Canadian test, thankfully, may not have been good predictor of anthrax spores’ behavior in the real world. In that study, the anthrax was folded in a contained sheet and was ejected when the sheet was pulled and opened. Perhaps the terrorists just dropped the anthrax into the envelope, where it remained. It is also possible that the anthrax had migrated out of the sheets during the extensive handling by the post office and settled to the bottom. And finally, maybe the anthrax did disperse, but only through the rooms where the letters were opened; the Canadian tests did not include measures of dispersal through ventilation, and this kind of dispersal may not be so efficient. In the tests, the half-life of the anthrax exposure in the chamber was about five minutes, suggesting that it settles quickly. (The air in the test chamber was recirculated, so the loss wasn’t through dilution.) Five minutes is long enough to infect people in the room, but not for spores to migrate far.
At this point, I believe, the U.S.-based al Qaeda agents panicked. They had failed in their mission, and they didn’t know why. They guessed that their anthrax had lost its potency, and in desperation, they mailed out all of the remainder, much of it in pure undiluted form, on October 9.
Anthrax spores were eventually detected not only at the U.S. Senate and the House of Representatives, but also at the White House mail facility, the Supreme Court, the CIA mail facility, the Pentagon, and all over Washington, DC. The prevailing belief is that spores spread to so many sites through cross-contamination in the mail rooms. But I think it worthwhile to consider the possibility that some of the detections were from early, diluted letters. In their first mailings, the terrorists assumed the anthrax could be diluted, and spread in this way to more locations. This may also account for several of the “hoax” letters that Rosenberg describes.
My scenario may seem complex, but real scenarios always are. I don’t claim to have the details correct. No scenario presently explains everything, and to make sense of the complex situation, you must judge your evidence. Which is more credible when the conclusions conflict: a handwriting analyst who says the terrorist was American, or a medical doctor, Dr. Christos Tsonas, who had treated the leg of Ahmed Alhaznawi, one of the September 11 hijackers, and says “the most probable and coherent interpretation of the data available” is that the infection was skin anthrax? If we are searching only for an American who could have mailed the anthrax, we might miss the American who didn’t mail it, but did steal itor maybe just failed to destroy it, when ordered to do so.
My opinion is in the minorityin fact, the tiny minority. According to the October 27 Washington Post, a senior official said “nobody” believes the anthrax attack was the second wave. “There is no intelligence on it and it does not fit any [al Qaeda] pattern.” But whether it fits the al Qaeda pattern depends, in part, on the intended scale of the carnage. It may be a mistake to assume the attack worked out as planned.
If I am right, the terrorists may now be disillusioned with anthrax attacks. But it would be foolish to relax. Bin Laden was building laboratories in Afghanistan that, given time, could have produced not grams but kilograms of spores. Tons of anthrax were grown in Soviet laboratories, and buried on Vozrozhdeniye Island in the Aral Sea (just north of Iran and Afghanistan)possibly along with some smallpox virus. (It was treated with bleach, but tests show much of it is still viable.) The Soviet anthrax was reported to be resistant to most antibiotics. So, despite the limited casualties of this first biological warfare attack on the United States, the prognosis is bleak. Biological terror is likely to prove more accessible and easier to implement than nuclear terror. The “mad scientist” of future fears is more likely to be a biologist than a physicist. Even though I am a physicist, that thought does not give me much comfort.
DeepMind’s cofounder: Generative AI is just a phase. What’s next is interactive AI.
“This is a profound moment in the history of technology,” says Mustafa Suleyman.
What to know about this autumn’s covid vaccines
New variants will pose a challenge, but early signs suggest the shots will still boost antibody responses.
Human-plus-AI solutions mitigate security threats
With the right human oversight, emerging technologies like artificial intelligence can help keep business and customer data secure
Next slide, please: A brief history of the corporate presentation
From million-dollar slide shows to Steve Jobs’s introduction of the iPhone, a bit of show business never hurt plain old business.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.