Armchair Sleuths Track Anthrax Without a Badge

Investigation Attracts Amateur Detectives Hoping to Crack the Long-Unsolved Case

By ANTONIO REGALADO, Staff Reporter of THE WALL STREET JOURNAL

Richard M. Smith had anthrax on his mind when he drove to Trenton, N.J., and mailed 20 letters one Sunday in July. On the back of each envelope, he carefully printed the time and the location of the mailboxes. On the front, he wrote his own address in Massachusetts.

Mr. Smith, a free-lance computer consultant, was trying to figure out what day the infamous letters sent to Capitol Hill could have been mailed. As a sniper terrorizes the Washington suburbs, Mr. Smith and a handful of other do-it-yourself gumshoes remain preoccupied with another serial killer who struck the nation's capital almost exactly one year ago -- the anthrax mailer.

The attacks, which claimed five lives last fall, remain unsolved. The letters bore no fingerprints or DNA trace, and the Federal Bureau of Investigation is keeping mum about any progress.

"There are a half a dozen of us who are obsessed with it," says the 48-year-old Mr. Smith, who tracks developments and theories about the mailings on his Web site, www.computerbytesman.com. "This morning I was doing it in my sleep."

Such mysteries often attract a following, says Clinton Van Zandt, a former member of the FBI's behavioral-sciences unit in Quantico, Va. Would-be sleuths range from people with "deep-seated psychological problems, to people who are very sincere, have the resources, and it becomes an obsession with them." He adds: "Who would not want to be the person who cracked the anthrax case?"

On Internet chat groups such as "Real Anthrax" and "Anthrax Fans," followers of the case argue their views daily, often vehemently. Some of them offer elaborate conspiracy theories, such as that Iraqi agents were behind the bioterror attack as part of an effort to blackmail the U.S. government. The e-mail dialogue tends to bring in other mysteries, including the Kennedy assassination and the recent sniper shootings.

Joint Reward

The FBI and the U.S. Postal Service are jointly offering a $2 million reward for information leading to a conviction, but Mr. Smith says his motivation is the thrill of playing detective. "It's like a real-life version of 'Law & Order,' " he says.

Mr. Smith has even solved some whodunnits. Along with a few other computer pros, he independently identified the author of the "I Love You" and "Melissa" e-mail viruses.

The best-known of the unofficial anthrax trackers is Barbara Hatch-Rosenberg, a professor of environmental science at the State University of New York at Purchase and chairwoman of the Federation of American Scientists' working group on biological weapons, which seeks to enforce and expand the international treaty that bans their use.

Based on tips that she says she received from defense insiders, Dr. Hatch-Rosenberg this summer publicly pressured the FBI to investigate former government germ scientist Steven J. Hatfill in connection with the case. The FBI searched his Maryland apartment twice this summer, but it hasn't named him as a suspect. Dr. Hatfill has denied any connection with the crime.

Then there's Donald Foster, an English professor at Vassar College, for whom the anthrax case is an irresistible target for his skills as a "forensic linguist." Prof. Foster, a Shakespeare scholar, made his name as a real-world riddle-solver by being the first to identify Newsweek columnist Joe Klein as the author of "Primary Colors," an anonymous novel about Bill Clinton's 1992 campaign. He says he did that by comparing the book's prose style with published articles.

Prof. Foster says he has worked with the FBI on a number of cases, including the early days of the anthrax investigation. A spokesman for the FBI's "Amerithrax" task force didn't return phone calls seeking comment.

Prof. Foster, who works out of a basement office on the Vassar campus, says he compared the anthrax mailings with writings and published statements of scientists. He pressed the FBI to study several past cases of biological hoaxes that he claims bear similarities to the anthrax incidents. In particular, he points to anthrax threats mailed to a Georgia school and television station in 1999, as well as an envelope with a harmless bacteria delivered to the offices of a Jewish charity in Washington in 1997. Prof. Foster says he has told the FBI about two people who have worked at the Army germ lab at Fort Detrick, Md., who "need to be looked at." He wouldn't identify the two individuals.

Other Clues

Mr. Smith has been following other clues. When the FBI released copies of the anthrax letters sent to Sens. Tom Daschle and Patrick Leahy, he noticed that they both bore nine-digit ZIP codes, not the more-usual five-digit codes. "Not a lot of people use them," he says. "So I started looking at this letter and asked, 'How would someone get that kind of address?' "

He tracked down two Web sites that carried nine-digit codes for both senators. Mr. Smith then sent an e-mail explaining his discovery to Juan Cabanela, an assistant professor at Haverford College who maintained one of the sites of congressional addresses. Dr. Cabanela searched his records and found three computer servers that had looked up both addresses. He called the FBI's Philadelphia field office offering to share his findings. Dr. Cabanela says he never heard whether his leads panned out, but he claims an FBI agent tried to recruit him last month. A spokeswoman for the FBI's Philadelphia field office declined to comment.

Mr. Smith later tried the mailbox experiment to determine the earliest date the letters could have been dropped off. Establishing this time frame may prove important to the case. Postal records showed the contaminated Capitol Hill letters were processed at a Trenton, N.J., mail-sorting center on Oct. 9, 2001 -- a Tuesday. In August, the FBI said it had found an anthrax-tainted mailbox in downtown Princeton, and began showing a photo of Dr. Hatfill to local merchants. Dr. Hatfill responded by providing work records that he claimed proved he couldn't have traveled to Princeton on that Tuesday, or even the previous Monday.

When the letters started arriving at Mr. Smith's house a few days after he mailed them on a Sunday this summer, all of them were postmarked Monday. The culprit, he concluded, must have mailed his letters Monday or Tuesday to get a Tuesday postmark.

What Mr. Smith forgot as he performed the July experiment, however, was that Monday Oct. 8 last year was Columbus Day, a public holiday on which mail wasn't collected from letter boxes. That may mean the letters could have been posted as early as the Saturday before.