The most famous instance of the FBI's disturbing propensity to take action built around a predetermined (and factually flawed) scenario is the Richard Jewell case, when agents decided that the former security guard had perpetrated the pipe bombing of the Atlanta Olympics in 1996.

Not only was Jewell's name dragged through the mire, but the trail of the real killer grew cold as agents focused their energy on a man later proven to be innocent. It was only when the likely bomber, right-wing terrorist Eric Rudolph, set off more bombs at gay bars and abortion clinics around the South that the FBI finally picked the trail back up. By then, of course, it was too late for his subsequent victims -- and Rudolph to this day remains at large.

Combine that with fiascoes at Ruby Ridge, Idaho, and Waco, Texas, where federal agents decided to bring the full force of their armaments into play against religious fanatics, and a pattern has emerged of an agency unwilling to shift from an original theory or plan of attack.

That stubbornness has frequently appeared in play in the government's Oklahoma City bombing investigation. A month before McVeigh's 1997 trial, prosecutors were prepared to argue that he had used a fertilizer-diesel oil mix for the bomb, even though there was clear evidence in lab analyses of the explosion's force that the bomb was not of that type. But the FBI had obtained receipts showing Nichols had purchased two tanks of diesel for his pickup truck, and had even gone so far as to enlist as witnesses entomologists who had done autopsies on dead insects found in puddles of diesel at Geary Lake, Kansas, where McVeigh and Nichols had put the bomb together. It was evidence FBI agents had found -- even if they knew it didn't prove anything.

But then, on the eve of the trial, Playboy magazine ran an article on the case that included information from documents leaked by McVeigh's defense team. In those documents, McVeigh had said that he used nitromethane, a high-combustion jet fuel, in the bomb and had obtained it at a racetrack south of Dallas.

"Of course, Lori Fortier had been telling them this all along, that McVeigh said he had nitromethane," says Kevin Flynn, a veteran reporter for the Rocky Mountain News in Denver who covered the trials. "But since they couldn't prove it, and they couldn't locate where he might have gotten it, they couldn't go to the jury with it, so they went with a false story -- a story they had reason to believe was false, that it was diesel fuel, only because they could show that Terry Nichols bought two tanks of diesel fuel that weekend, and that he had a siphon. So they were going to go with a false story, because they couldn't prove the one they thought was true."

According to Flynn's reporting, though, after the Playboy story prosecutors realized they needed more. "They went down to Dallas, they found the racetrack, they looked at the calendar, they got the race dates there in October and they tracked down all the nitromethane sales people. And one guy said, 'Yeah, a guy came in and paid cash for three barrels, who kind of looked like McVeigh.' He couldn't do a positive ID two and a half years later, but he was a credible enough witness that they changed their whole story on the eve of the trial."

There are other indications that the FBI chose to simplify the case at the expense of potential leads. According to testimony by the FBI's own fingerprint witness at the McVeigh trial, there were thousands of fingerprints recovered in the investigation. The witness testified on cross-examination that there were roughly about 1,100 fingerprints that they did not try to identify. Instead of cross-referencing them in a broad search, they simply checked them against a handful of potential John Doe No. 2 suspects. The agency has never explained why it chose such a limited check.

A more traditional problem with the FBI's modus operandi cropped up in the investigation, too -- namely, its high-handed treatment of local law enforcement officials, which often takes the form of ignoring important information they possess. This was especially striking to police in the rural Kansas precincts where McVeigh and Nichols constructed the truck bomb. A number of them offered leads that still appear promising, but their attempts fell on deaf ears.

"I can tell you that the frustration level around here was just enormous," says Suzanne James, a former deputy prosecutor in Topeka who had information on some of the militiamen with whom Morris Wilson associated. "When the FBI came in here, they just plain wouldn't listen to anything anybody local told them.

"If nothing else, I think the FBI owes the public an explanation, if these people were investigated, why they were eliminated as possibilities," says James. "Otherwise, you have one of these endless things like the Kennedy assassination."

Even analysts whose work is often devoted to debunking conspiracy theories are troubled by the lingering riddles in the Oklahoma City bombing.

"I think it's not a closed case," says Mark Potok, director of the Southern Poverty Law Center's intelligence-gathering arm. "I think that certainly there's the possibility that there are two or three or perhaps more people out there still. I absolutely don't think that's certain. That said, I think there's no question there are unanswered questions."

Now, the best prospect for settling the mysteries of Oklahoma City no longer lies with the investigators at the FBI, or whatever secrets may emerge among the thousands of recently disclosed documents. And it appears it may very well not happen before McVeigh is executed. However, not all of McVeigh's secrets will die with him. Nichols will remain very much alive, pending the outcome of his state trial. And in that setting, there is at least a reasonable chance -- particularly if the sentencing judge replicates the offer Judge Matsch made to Nichols -- that the identity of John Doe No. 2, or whoever it was that helped him bomb Oklahoma City, could finally come to light.

Recent Stories