I’ve been interested in the work of UC-Berkeley’s Douglas Cooper and David Ragland on crashes at railway crossings. Looking at incidents that occurred between 2000 and 2004 in the state of California, they found that “of the crashes that occurred, 73 percent occurred at crossings equipped with gates, 59 percent involved vehicles moving over the crossing, and 27 percent involved vehicles that had driven around or through lowered gates. An unbelievable number, 21 percent, involved a vehicle running into a moving train.”

I couldn’t help but think of those findings when I recently came across the following remarkable passage in John Stilgoe’s book Metropolitan Corridor: Railroads and the American Scene, describing the early problem of dealing with vehicular traffic at railroad crossings:

“Adding gates, bells, and electric flashing lights at some crossings at first seemed to help, especially if the gates overlapped each other to prevent motorists from snaking past them onto the tracks. But by 1913, experts knew that numerically as well as comparatively more persons are killed at protected crossings,” at crossings defended by watchmen, gates, bells, lights, and signs. What accounted for “comparatively”? Certainly protected crossings usually passed many more wayfarers than unprotected rural crossings far from towns, but why did proportionately more people collide with trains there? Did carelessness born of some mad scurrying haste account for the deaths, or was it the old “familiarity with the timetable” syndrome? If anything, a sort of early-twentieth-century highway hypnosis might explain the accidents at protected crossings. “How many of you readers heard your clock strike at the most recent hour?” asks Whiting in his 1913 article. People intimately familiar with their route to work, to shopping, to school, simply did not realize the protected crossings. Lost in some sort of waking trance, they walked past the lights or drove directly into and through the gates. “Disgusted railroad men will sometimes tell you that the more you protect a crossing, the worse people behave,” Furnas noted in 1937. “They seem to figure that if the company has taken all that trouble, the drive is absolved of responsibility for himself.” So concerned were California authorities that as early as 1917 they began designing speed bumps into paved highways approaching crossings, hoping that a violent jarring would knock motorists out of their trances and apprise them that they “should cut down speed and be on the lookout for warning signals.” By 1937, after the speed bumps had increased in height to two or three feet, one magazine writer concluded that they did nothing to alert motorists. Drivers simply breezed over them, crashed through gates, and struck trains. When reformers suggested that railroad companies install gates so solid that motorists could not break through them, companies replied that such gates could not be designed. The flimsy gates, they explained, existed to permit motorists to crash through both pairs and escape death, or through the far pair if they entered the crossing as the gates lowered. By the early 1930s, the protected grade crossing displayed the gadgets of mechanical, electrical, and efficiency engineers—and all of the engineers had failed.”

An interesting early example of the challenges of safety engineering in light of human risk compensation, and clearly a longstanding problem that has not been solved.

This entry was posted on Thursday, May 14th, 2009 at 11:30 am and is filed under Risk, Traffic Engineering, Traffic safety, Uncategorized. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.