The Annals of Unsolved Crime

by Edward Jay Epstein

  But the FBI polygraph effort, which went on for another seven years, never produced a culprit. No suspect ever broke. Although the FBI investigation code-named “Amerithrax” consumed “hundreds of thousands of investigator work hours,” making it the most massive inquest in the history of American law enforcement, it produced no further actual evidence identifying the party behind the attack.

  The FBI did not find a single witness to the theft, preparation, or mailing of the anthrax. It did not find any of the equipment used to make the powdered anthrax or any growth media that added silicon, tin, or any other of the trace elements found in the envelopes containing the killer anthrax. It did not find anthrax in the homes, automobiles, personal effects, or even garbage of any of the scientists it was investigating. It did not find the original letter from which the photocopies were made, or the photocopy machine on which it was made (although it examined tens of thousands of such devices). On the four letters and envelopes, or tape sealing them closed, it found no fingerprints, hairs, fibers, or DNA, nor could it identify the handwriting, other than to ascertain that four letters—A, T, T, and A—had been emphasized. Nor did surveillance cameras in the vicinity of the mailbox in Princeton show any of the suspected scientists. The perpetrator had effectively erased any evidence of his or her involvement in the crime.

  The first fatality of the attack was Robert Stevens, a photo editor at American Media in Boca Raton, Florida. There was a prior possible anthrax exposure in Florida. Dr. Christos Tsonas, a doctor at the Holy Cross Hospital in Fort Lauderdale, Florida, informed the FBI after Stevens’ death that two months earlier he had prescribed an antibiotic for an emergency-room patient named Ahmed Alhaznawi, who had a suspicious black lesion on his leg. At the time, Dr. Tsonas was unfamiliar with anthrax. When he was shown pictures of anthrax lesions in October 2001, he then described Alhaznawi’s lesion as “consistent with cutaneous (skin) anthrax.” Alhaznawi was a pilot, one of the four hijackers of United Airlines Flight 93 on September 11. Since he was incinerated in the crash, the FBI could not pursue this lead.

  Without any direct evidence, other than the attack anthrax itself, the FBI tightened the surveillance on American scientists. One such scientist was microbiologist Perry Mikesell, who had access to anthrax in 1999 when he worked at the Battelle Memorial Institute in Columbus, Ohio. The FBI scrutiny of Mikesell became so intense that, according to family members, he began to drink heavily. Although he died from a heart attack in October 2002, family members said that he drank himself to death. The FBI then turned its attention to Dr. Steven J. Hatfill, a virologist who in 1999 worked at the U.S. Army Medical Research Institute of Infectious Diseases (USAMRIID) in Fort Detrick, Maryland, which was experimenting with the Ames strain of anthrax. Hatfill showed stress on a polygraph exam, and he had taken the antibiotic Cipro, which protects against anthrax infection, prior to the attack, and he attracted the attention of a bloodhound trained to react to the anthrax letters. The FBI surveillance agents followed him so closely that one FBI tail car actually ran over his foot. The FBI searched his home, property, girlfriend’s apartment, and workplaces, after the media was alerted. As a result, Hatfill lost his job, consulting contracts, and contact with many associates. Even isolated, Hatfill did not break under the pressure. Instead, he sued the government. A federal judge expressed outrage that the FBI had pursued him for five years without a “scintilla of evidence,” leading the Department of Justice to exonerate him and pay him $5.82 million in August 2008.

  Meanwhile, the FBI had contracted with The Institute for Genomic Research (TIGR) for an analysis of the DNA of the anthrax that could more closely pinpoint its origins. It used the DNA of minute changes, or “morphs,” in the spores as “genetic fingerprints.” Of all the samples the FBI had collected from different labs using the Ames strain, only one matched the “fingerprints” of the killer anthrax in the letters. It was from flask RMR-1029 in Dr. Bruce Ivins’ lab. This anthrax had been created in 1997 at the Dugway Proving Ground and sent to Ivins at USAMRIID for vaccine tests. If the match was correct, the FBI finally had the “murder weapon.” The custodian for this flask had been Ivins, a microbiologist researcher working on anthrax vaccines. It was part of his job to dispense anthrax to other scientists in the world doing approved research.

  Even though Ivins had passed his polygraph exam, he now replaced Hatfill as the prime suspect. The FBI then worked to eliminate all the other scientists to whom Ivins had given access to RMR-1029, for one of three reasons: a scientist had not worked solo in a lab and therefore lacked the privacy needed to process the wet spores into dry powder; a scientist lacked the skill set to do the job; a scientist was located too far away from Princeton to mail the letters. Through this Sherlock Holmes process of elimination, the FBI narrowed the list to one man: Ivins. He worked alone in his lab, he had the skill set, and he could have made the nine-hour round-trip drive to Princeton.

  The FBI now needed to break Ivins or find evidence that would hold up in court. It spared no effort. It even attempted to trace back to him the 6 ¾-inch “Federal Eagle” pre-franked 34-cent envelopes in which the anthrax was mailed, even though nearly 70 million of these virtually identical envelopes had been manufactured for the U.S. Postal Service. To narrow this Augean task, an FBI task force attempted an exotic technique called “microscopic print defect analysis.” The idea is that each time a manufacturer prints a new order of envelopes for an area of the country, at some point in the printing process, errors occur. The strategy used was to compare the microscopic errors in the anthrax envelopes to a sample of envelopes collected from different areas of the county. From this analysis, the task force determined that those envelopes were likely received by the Dulles Stamp Distribution Office, which distributed to post offices inside the Capital Beltway and southern Maryland. This was undoubtedly great forensic work. The problem was that this area included the employees of a large number of bio-technology facilities that dealt with anthrax. It also covered areas where many employees of intelligence services, foreign embassies, and defense contractors reside. To tie the envelopes, the FBI sought to find them in the search of Ivins’ home, office, and past mailings, but came up with no matches. The envelopes could not even be definitely traced to his local post office in Frederick, Maryland, because the office had returned all its stamped envelopes when postal rates changed.

  In late 2007, Ivins was harshly interrogated. By this time almost everything he had, including his home, computer, car, and personal effects were searched. Yet no trace of anthrax was found on any of his personal effects. His twin children were then offered a $2.5 million reward to provide evidence against their father, but they did not provide any. Meanwhile, his security clearance was withdrawn, barring him from his work. Even though he had passed his polygraph exam, the intensive searches produced no evidence that he had ever handled anthrax outside his lab or had driven to the Princeton-Trenton area in September 2001, and the FBI came up with no evidence linking him to the killer anthrax. However, the FBI found creepy behavior in Ivins’ personal life. He collected pornography, had web aliases, and admitted that he had become obsessed with a sorority girl. As the pressure continued, and his legal fees threatened to bankrupt him, he drank heavily, took prescription mind-altering drugs, acted erratically, and had a psychiatric breakdown. Finally, his life a shambles, on July 29, 2008, he killed himself with a Tylenol overdose. Though he was never charged or brought before a grand jury, the FBI identified him as the anthrax killer.

  Less than a week after Ivins’ apparent suicide, the FBI declared him to have been the sole perpetrator of the 2001 anthrax attacks and released bizarre emails that he had written which, though unrelated to anthrax, suggested that he was a disturbed individual. Even though Ivins had not been indicted, and none of the case against him had ever been tested in a legal process, the FBI justified its extraordinary finding of guilt on the basis of the scientific evidence, which, it suggested before Congress, proved that Ivins, and Ivins alone, was behi
nd the anthrax attack.

  I had been investigating this case for seven years at this point, and I had interviewed scientists with whom Ivins had worked or who were otherwise familiar with his lab work. All of them told me that they did not believe that Ivins had the means to grow and then convert the liquid lab anthrax into the powdered attack anthrax without others in the lab observing this process. Jeffrey Adamovicz, one of the lab scientists I interviewed, wrote me: “Even if Bruce (or anyone else) wanted to use the fermentor in our labs, [and it] was non-operational, this person would have been observed even if it were operational.” I also found out that the FBI had never reverse-engineered or otherwise replicated the process by which the anthrax was produced. Indeed, almost all the attempts failed to reproduce the silicon content found in the anthrax in the letter sent to the New York Post. After I reported this FBI failure in a commentary in the Wall Street Journal on January 25, 2010, D. Christian Hassell, assistant director of the FBI Laboratory, replied to the Wall Street Journal, “The FBI is confident in the scientific findings that were reached in this investigation. We utilized established biological and chemical analysis techniques and applied them in an innovative manner to reach these findings.” But had the science actually validated the FBI’s conclusions?

  To validate the science that the FBI had used to narrow the search to Ivins, the FBI contracted with the prestigious National Academy of Sciences (NAS) to conduct an independent assessment of its methods. The NAS took until February to complete its assignment. The result was not what the FBI expected. The NAS report concluded that the FBI’s key assertion that its genetic fingerprinting showed that the killer anthrax could only have come from the flask in Ivins’ custody was flawed. “The scientific data alone do not support the strength of the government’s repeated assertions that RMR-1029 was conclusively identified as the parent material to the anthrax powder used in the mailings,” it stated. “It is not possible to reach a definitive conclusion about the origins of the B. anthracis in the mailings based on the available scientific evidence alone.” Without a valid scientific basis for tracing the killer anthrax to Ivins’ lab, the FBI case against him is based on inference from his suspicious behavior.

  The NAS report also raised questions as to where the anthrax was stolen from and where it was processed into powder. It revealed that the killer anthrax in the New York Post letter had a silicon content of “10 percent by mass.” Ivins’ lab anthrax had no silicon. The FBI had suggested that the silicon could have come naturally from the mixture of water and nutrients in which the anthrax was grown. But when the FBI attempted to test this theory by having the Lawrence Livermore National Lab grow anthrax in broth laced with silicon, it failed to match the silicon signature. Out of fifty-six tries, most results were a whole order of magnitude lower, with some as low as .001 percent. The FBI had other labs attempts to reverse-engineer the killer anthrax, but the closest it came to the silicon signature was in some results from Dugway Proving Ground.

  Ivins could still have committed the crimes, but he would have needed a collaborator at Dugway or elsewhere who had access to the powdered anthrax or the process to make it. This complication led Senator Leahy to tell FBI Director Robert Mueller, “I do not believe in any way, shape, or manner that [Ivins] is the only person involved in this attack on Congress.” Although the FBI did not agree, the case remains unsolved a decade later.

  There remain a number of viable theories of the case. The FBI’s theory is that Ivins, and Dr. Ivins alone, was behind the attack. Senator Leahy’s theory is that Dr. Ivins may have been guilty, but that he could not have acted alone. Finally, there is the theory that a terrorist or foreign intelligence service was behind the attack. The NAS report found that, though the results were inconsistent, some samples collected abroad from an unnamed location possibly matched the attack anthrax.

  My assessment is that the FBI failed to find the perpetrator of the anthrax attack. Indeed, it failed twice. Its first wrong man was Dr. Steven J. Hatfill, whose career was ruined by its 24/7 investigation of him; the second wrong man was Dr. Bruce Ivins, who committed suicide under its relentless pressure. In a case such as the anthrax attack, in which the weapon leaves microscopic traces everywhere, the equivalent of Sherlock Holmes’ dog-that-did-not-bark is the conspicuous absence of evidence. The FBI investigation was possibly the most massive in history in terms of expended man-hours. Consider then what evidence the FBI did not find that would have implicated their final suspect, Dr. Ivins.

  First, no traces of anthrax were found in Ivins’ home, car, clothing, garbage, or personal effects, so there is no evidence that Ivins ever removed anthrax from his lab.

  Second, none of the dozens of scientists, technicians, and security guards that worked or visited Ivins’ lab, and were relentlessly interviewed by the FBI, ever witnessed him with the apparatus or growing vats necessary to process liquid anthrax (which was stored in a lab flask) into the powdered anthrax used in the attack. Yet to grow and dry this attack anthrax required gallons of media, as well as drying equipment such as a fermentor, which, according to other scientists who worked with him (and whom I interviewed), were not available to Ivins.

  Third, the FBI was not able to find tollbooth records, gas purchases, or surveillance-camera evidence that Ivins drove his car to Princeton, or was in Princeton, during the period when the anthrax letters were mailed. If he was there, no one saw him, and he did not leave a trace of his presence.

  Finally, the FBI could not trace the copies of the anthrax letters to any photocopying machine available to Ivins (or any other scientist at Fort Detrick). Nor could it find any pre-paid envelopes in Ivins’ home that matched those in which the anthrax letters were mailed. In short, there is no physical evidence actually linking Ivins to the anthrax attack.

  Since the science by which the FBI attempted to eliminate all other suspects also turned out to be flawed, according to the exhaustive review by the National Academy of Science panel, it is not certain that the attack anthrax originated at the Fort Detrick lab. It could just as likely have come from the Battelle Memorial Lab in Ohio or the Army’s Dugway Proving Ground in Utah. In fact, the silicon signature in the anthrax points directly to Dugway. The Dugway lab had been making dry anthrax there, using the same strain of anthrax, since the late 1990s to test anthrax-detection devices. The concern here was that Iraq would use anthrax against U.S. forces, so Dugway scientists employed a multi-stage process to mill test anthrax using a silica-based flow enhancer that was not available at Ivins’ facility at Fort Detrick. As there had been a number of security breaches at Dugway during this period, it is plausible that someone at Dugway stole a minute ampule of anthrax anytime after 1997 and, like any classic espionage operation, delivered it to another party, either foreign or domestic, who used it in the September 2001 attacks.

  The lesson to be taken away from this unresolved crime is the vulnerability of even the most massive investigations to a preconception embedded in a behavioral profile. Just weeks after the attack, the FBI had arrived at a “behavioral assessment” pointing to a lone American scientist. It described, as David Willman reports in his 2011 book Mirage Man, a well-educated “single person” with “legitimate access to select biological agents,” who is “stand-offish,” and who prefers not to work in a “group team setting.” It was this assessment that steered the FBI first to Hatfield and then to Ivins. When the FBI’s behavioral science unit was created in 1972, its primary mission was profiling unknown serial killers by using historical data. But in the case of the anthrax attack, there was only one prior case of bioterrorism in the United States, the deliberate contamination with salmonella of salad bars in ten restaurants in Oregon. But this biological attack, which resulted in the poisoning of 751 individuals, was not perpetrated by a lone American. It was the act of a conspiracy in which the main perpetrators were foreign-born followers of the Indian spiritualist Bhagwan Shree Rajneesh. Nor did the text of the anthrax letters themselves point to a single American s
cientist, since it used the plural “we,” contained the misspelling of the word “penicillin,” and also had jihadist formulations, such as “Death to America.” If, as the FBI assumed, the text was a deliberate deception, it left little material to support its profile of a lone American scientist. Yet, by clinging to this profile, it neglected investigative paths that may have resolved this crime.

  CHAPTER 6

  THE POPE’S ASSASSIN

  On May 13, 1981, Mehmet Ali Agca, a twenty-three-year-old Turkish citizen, who was amidst a throng of people in the piazza in front of St. Peter’s Basilica in Vatican City, suddenly raised a Browning semi-automatic pistol, and shot Pope John Paul II no fewer than four times. The fusillade of bullets gravely wounded the pope, who survived the attack. Agca was captured gun in hand at the scene by Vatican security and Italian police. He had arrived in Rome by train only three days earlier. Under interrogation, Agca freely admitted firing the shots and said that he acted alone. Two years earlier, Agca had been sent to prison in Turkey for assassinating Abdi Ipekci, the editor of the newspaper Milliyet, a crime he confessed to on Turkish television. He had also threatened in a letter to kill Pope John Paul II, whom he called “the Commander of the Crusades” against Islam. He escaped from prison in November 1979. Based on his confession, the Italian investigation concluded that he acted as a lone fanatic when he shot the pope. In July 1981, he was tried in Rome, convicted, and sentenced to life imprisonment. Even with the conviction, it was not a closed case.