Is it possible to design a better stop sign?
Check out our Magnum Photos gallery of stop signs.
In an Internet parody called "The Process," a designer is given a corporate gig with a simple brief: to design a new stop sign. "We're seeing reports that people don't know what to do at an intersection," he is told, and from there it descends in an absurd spiral of tweaks and redesigns, with the designer's creative vision cast against the slow strangulation of groupthink. While the video is a hilarious send-up of the corporate design process, its premise—that designing an effective stop sign is actually a simple task—couldn't be farther from the truth.
In reality, the design of the stop sign, however seemingly settled, is not necessarily ideal. In 1998, for example, there were more than 700,000 crashes at intersections marked—or "controlled," as engineers say—by stop signs. More than 3,000 of these were fatal. Laura Bush's new biography, Speaking From the Heart, highlights the stop sign's role in the fatal crash she caused in high school: She drove through an intersection marked by a stop sign, striking the car of a good friend and killing him. She notes, among other factors, that the stop sign was too small (current signs are larger, and mounted higher, among other changes).
We don't know what the fatality numbers would look like if modern stop signs were replaced by something else or taken out altogether, but the fact that the sign is at least indirectly implicated in several thousand deaths and hundreds of thousands of injuries every year suggests that traffic engineers should at least look into improving, or replacing, the device.
Indeed, Gary Lauder, a Silicon Valley entrepreneur, was just the latest to propose a redesign of the stop sign during a recent and much-discussed TED presentation. Using the example of a three-way intersection in which a minor road ended in a "T" at a major road, with stop signs all the way around, Lauder calculated that the stopping led to a collective yearly loss in fuel and time valued at roughly $112,000. Why not just use a yield sign on the minor approach? Well, at certain times of the day a queue backs up there, and cars have trouble making the turn. So Lauder proposed a hybrid "stop-yield" sign, simply labeled "Take Turns," paired with the instruction: "If cars are waiting please stop and alternate."
More on the viability of Lauder's design in a moment. But first it's worth considering how we got the sign we have now. Like many forms of traffic instruction, the stop sign has murky origins. It was adapted from railway controls but without rigorous scientific testing. As Kenneth Todd has pointed out, "the traffic control system developed piecemeal. … [W]hen large numbers of automobiles burst on the scene early in the century, political pressures, guesswork, and panic measures served as substitutes for scientific expertise." Indeed, historian Clay McShane writes that in 1914, "Detroit police sergeant Harry Jackson cut the corners off a square sign to create an easily recognized octagonal shape for first red stop sign or 'boulevard' stop." (The signs were controversial: McShane notes that "Illinois courts briefly ruled stop signs illegal in 1922 as a violation of the rights of individuals to cross streets.") By 1927, a rough standardization of the sign was set in place by the American Association of State Highway Officials. An octagonal shape, with red letters on a yellow background. It wasn't until nearly three decades later that the current design—white letters on a red background—was settled upon, in a 1954 supplement to the Manual of Uniform Traffic Control Devices, the operative rulebook for traffic engineers. Is the current design as good as it could be? There are two ways to think about that problem. We must ask: Do drivers see stop signs? And, more importantly, what do they do when they see them?
For nearly a century, it seems, drivers have been ignoring stop signs. In a 1934 study published in the Journal of Social Psychology, for example, F.H. Allport examined driver behavior at an intersection with a stop sign with approaching cross traffic. A majority (75.5 percent) of drivers came to a full stop—no surprise given the imminent danger. But what about in cases where no cross-traffic was visible? Would people still stop? A 1968 study in Berkeley, Calif., published in Law & Society Review, found that just 14 percent of drivers brought their cars to a full stop "without being forced to do so by cross traffic" (the so-called "California roll" was the norm).
No one has more doggedly pursued the question of stop-sign compliance than John Trinkaus, who conducted an annual stopping survey at the same intersection for nine straight years in the 1970s and '80s, finding a creeping decline. In his culminating 1997 masterwork, "Stop Sign Compliance: A Final Look," Trinkaus revisits his old intersection and finds that the percentage of people making a full stop had dropped from 37 percent in 1979 to a mere 3 percent.
Why did this happen? There are several ways to read the data (and they are not necessarily mutually exclusive). On the one hand, traffic is a social environment, and authors like Robert Putnam in Bowling Alone, or Jean Twenge in Generation Me, have argued that stop sign scofflawism is one minor indicator, among many, of a larger societal shift: a decline of civility and reciprocity, a lesser willingness to follow social rules. The argument is that a society marked by increased self-regard (and hence less regard for others), has neither the inclination nor the situational awareness required to accommodate others, whether by signaling one's intentions, stopping for pedestrians in a crosswalk, or heeding the familiar red octagon. On the other hand, traffic engineers have long known that excessive signage declines in effectiveness. This points to something of a Catch-22. Residents of a neighborhood may complain about drivers speeding down their street and petition the city to install stop signs. But stop signs are not a safety device as such, nor a traffic-calming device: They exist to assign right of way. Faced with more stop signs, some studies have shown, drivers may actually drive faster to make up time lost for stopping at (or really, slowing through) the intersection; the more signs installed, the lower the compliance.
John Staddon, a professor of psychology at Duke University, notes another problem: "The overabundance of stop signs teaches drivers to be less observant of cross traffic and to exercise less judgment when driving—instead, they look for signs and drive according to what the signs tell them to do." He reserves particular opprobrium for the four-way stop. The rules of engagement are somewhat informal—most drivers take it to be "first in, first out." What if two drivers arrive at the same time? Traffic laws state the driver to the right has the right of way (good luck with that). If four cars arrive simultaneously—well, it's best that they do not. "Remind me," asks Staddon, "aside from bewildering the driver, what's the point of stopping traffic in all four directions?" The four way stop, he argues, "weakens the force of all stop signs by muddling the main question drivers need to answer, namely: Which road has priority?" (One thing four-way stops have in their favor, however, is a superior safety record to two-way stops—and to traffic signals, for that matter).
Which brings us back to Lauder's suggested "Take Turns" sign. In the rarefied TED air, where the world is being saved and the old certainties boldly challenged by 15-minute PowerPoints, the idea seems sensible and perhaps even inspired. But the world of traffic is a more complicated place. For one, given the lack of compliance at stop signs, what's to ensure proper behavior in a less clearly demarcated situation? What if three cars approach simultaneously? What if a driver approaching fast on the main road assumes that his speed gives him priority, while the entering driver thinks the fact that he's pulled up to the new sign first gives him priority? (The costs estimated by Lauder in lost wages and fuel are vastly less than the costs of a fatal crash). And what should cars do if pedestrians are present? Lauder is right about the futility of that intersection, but wrong in his solution: If the money saved could indeed buy the adjacent property, then the best solution would be to simply install the safest and most smoothly flowing solution of all: A roundabout.
Of course, there are plenty of intersections in America that are "uncontrolled." In Portland and Seattle, for example, local neighborhoods are filled with any number of four-way intersections without any signs. And somehow drivers continue to negotiate these intersections safely, year after year, in the absence of clear instruction.
If traffic volumes rise above a certain threshold or a crash pattern evolves, other measures are taken—a roundabout may be built, traffic-calming measures deployed, or signage installed. And it is these latter cases, when some form of signage is required, that brings us to the question: Does the stop sign need to be reformed? Actually, the "stopping occasion," as the viral parody put it, has been tinkered with quite a bit. Engineers have sought to remedy visibility and compliance problems by adding any number of add-on accessories to the stop sign, among them "stop lines" painted on the road, rumble strips in advance of the stop, or flashing red lights on top of signs. One study tested a stop sign beneath which were posted a pair of LED eyes, roving back and forth—the idea was a meant as a reminder to look both ways. The study reported a reduction in "right-angle conflicts"—the most dangerous thing about intersections—though this may have merely been a novelty effect.
Interestingly, eyes were more effective than a "look both ways" text add-on. But what if the sign is not visible to begin with—what if it's obscured by a low-hanging branch or a double-parked truck? That's the thinking behind a new stop sign treatment being tested in Nye County, Nevada. Called "Drivers Alert," it attaches a secondary stop sign to the back of the stop sign across the intersection, giving the driver two chances to see the sign. It also graphically advises whether cross traffic is meant to stop or not, a fix that others have suggested. (Some engineers have proposed making stop signs at two-way stops orange, to immediately distinguish them from four-way stops.) Whether this effort is superfluous given existing visibility-reinforcing treatments (e.g., the stop bar) or, per Staddon, simply gives the dumbed-down driver more things to look at apart from what's happening on the road itself is open for debate.
The proposed solution also raises the interesting question of how we see (or do not see) stop signs. In large part, we see stop signs because they tend to exist where we expect them to exist (trouble begins when our "expectancy" is violated). But in crashes, the largest problem is not visibility but driver behavior—drivers either do not come to a full stop or pull out too close to an approaching car (one study found that only 17 percent of crashes at stop-sign controlled intersections involved drivers who "blew" the sign). In this regard trying to improve driver behavior through better signage is as futile as fighting illiteracy with better fonts.
Tom Vanderbilt is author of Traffic: Why We Drive the Way We Do, now available in paperback. He is contributing editor to Artforum, Print, and I.D.; contributing writer to Design Observer; and has written for many publications, including Wired, the Wilson Quarterly, the New York Times Magazine, and the London Review of Books. He blogs at howwedrive.com and lives in Brooklyn, N.Y. You can follow him on Twitter at www.twitter.com/tomvanderbilt.
Illustration by Robert Neubecker.