By Alan Prendergast
By Michael Roberts
By Michael Roberts
By Amber Taufen
By Patricia Calhoun
By William Breathes
By Michael Roberts
By Melanie Asmar
In an effort to streamline what promises to be a lengthy legal process, all fourteen lawsuits filed by families of people injured or killed in the shootings at Columbine High School have been moved to federal court in Denver. But an attorney for two of the families says his clients aren't much closer to getting their most pressing questions answered -- questions regarding what law enforcement knew about killers Eric Harris and Dylan Klebold prior to the massacre, and how police responded to the rapidly developing crisis on April 20, 1999.
Last spring, several plaintiffs went to state court to force the public release of the Jefferson County Sheriff's Office long-delayed "final report" on Columbine, as well as 911 and dispatch tapes, video surveillance footage, ballistics reports and other key documents related to the investigation. The audio and video record, released over a period of months, contradicted the sheriff's account of the five-hour rescue effort on several critical points, highlighting numerous self-serving distortions and omissions in the official report ("The Lost Command," July 13). But now concerns are also being raised about possible omissions in the "edited" evidence that's been made public, too, particularly in the audiotapes of communication between Jefferson County dispatchers, commanders and officers at the scene that day.
"We'd like to know where some people were and what they were doing -- and so far, in the information we've got, that isn't there," says Jim Rouse, an attorney representing the families of slain students Kelly Fleming and Daniel Rohrbough. "There may be a perfectly good explanation why it's not there, but we're not getting as much detailed information as we'd hoped for."
In his order granting the Fleming-Rohrbough public-records request, District Judge Brooke Jackson allowed the sheriff's office to make brief deletions of names, addresses and phone numbers to protect the privacy of various 911 callers, police officers and third parties, including the names of students other than Harris and Klebold who were accused of being somehow involved in the shootings. The judge's order also banned release of a portion of the 911 call from the school library after the killers entered the room because an enhanced version of the tape captures screams and dialogue accompanying the executions. The edited tapes, amounting to more than forty hours of recorded calls, were digitalized and released on two compact discs in July.
Most of Judge Jackson's restrictions refer only to the 911 calls. However, an analysis of the audio files reveals that the radio traffic between dispatchers and various responding officers was significantly edited, too. Police dispatchers are trained to give the time as they acknowledge reports of key events in the field -- "Shots fired, 11:32," for example -- and while the practice is far from consistent on the Columbine tapes, there is sufficient time-stamping on the calls to determine that the actual time elapsed far exceeds the running time of the tapes. In several instances, gaps of five to fifteen minutes in "real" time can be detected between seemingly consecutive calls.
One explanation for the gaps is the elimination of dead air. According to employees of Quality Data Systems of Boulder, the outside firm that transferred the edited tapes to CD for the sheriff's office, there were considerable silences between transmissions on some radio channels, especially as the afternoon wore on and the rescue effort became a cleanup operation. To save the expense of producing several discs with little activity, QDS compressed the dead spots. Thus, the audio file for Channel Two -- which begins with sheriff's deputy Neil Gardner reporting that he's under fire and ends with discussions about securing the crime scene -- squeezes nearly twelve hours of real-time calls into ninety minutes of tape.
But QDS staff could not account for the presence of hefty time lapses on the primary police channel, Channel One, that occur in the first three hours of the siege, when radio activity was at its peak. Neither the compression of dead air nor the deletion of personal data (which takes only a few seconds in other instances) adequately accounts for the phenomenon.
"They were supposed to be editing names, addresses and telephone numbers, basically," Rouse notes. "By the middle of the afternoon, I don't think they'd be worrying about [officers'] home numbers. Everybody and their dog was out there by then."
A detailed examination by Westword of the Channel One calls, which were released in three audio files, shows an increasing disparity between the clock time noted by the dispatchers and the events actually present on the tape. The first fifteen minutes of the attack on the school, for example, account for only eleven minutes of running time. By the end of the first hour, there's more than ten minutes unaccounted for. The first ninety-minute file ends shortly after 1 p.m., around the time a SWAT team enters the west side of the building -- nearly two hours after the shooting started. The second file begins roughly thirty minutes later, at 1:40 p.m. In the first 25 minutes of that file, another hour in real time elapses.
It's possible, of course, that dispatchers consistently misstated the time, or that some mechanical malfunction in Jefferson County's aging communications equipment is responsible for the gaps. But the sheriff's report maintains that the dispatch center encountered no technical difficulties throughout the day, and the agency is so confident of the accuracy of the digital clocks used for the dispatch tapes -- they'd been calibrated with the National Institute of Standards and Technology's atomic clock in Boulder shortly before April 20 -- that the dispatch time was used as the primary source for developing an official timeline of the events of April 20.