Why Ants Don’t Get into Traffic Jams (and Humans Do): On Cooperation as a Cure for Congestion - Traffic: Why We Drive the Way We Do - Tom Vanderbilt

Traffic: Why We Drive the Way We Do (and What It Says About Us) - Tom Vanderbilt (2008)

Chapter 4. Why Ants Don’t Get into Traffic Jams (and Humans Do): On Cooperation as a Cure for Congestion

Meet the World’s Best Commuter:
What We Can Learn from Ants, Locusts, and Crickets

When insects can follow rules for laneing, why couldn’t we the humans?

—road sign in Bangalore, India

You may feel you have the worst commute in the world: the grinding monotony of sitting in congestion, alternately pressing your brake and accelerator like a bored lab monkey angling for a biscuit; the drivers who stymie you with their incompetence; the slow deadening of your psyche caused by the ritual of leaving home forty-five minutes sooner than you would like so you can arrive at work ten minutes later than your boss would like.

And yet, in spite of all this mental and physical anguish, there’s at least small consolation awaiting you at the end of your daily slog: Your fellow commuters did not try to eat you.

Consider for a moment the short, brutish life of Anabrus simplex, or the Mormon cricket, so named for the species’ devastating attack on Mormon settlers in Utah in the legendary 1848 “cricket war.” Huge, miles-long migratory bands of flightless crickets, described as a “black carpet unrolling across the desert,” are still a dreaded sight in the American West. They travel many dozens of miles, munching crops and carrion. They heedlessly spill across roads, causing death for themselves and headaches for another traveling species, Homo sapiens, whose cars may slip on the dense mat of pulsating crickets. “Crickets on Highway” signs have been posted in Idaho. It turns out the insects are actually katydids, but the point is well taken.

Viewed as a scurrying mass, the Mormon cricket band seems a well-organized, cooperatively driven collective search for food—a perfect swarm designed to ensure its own survival. But when a group of researchers took a closer look at a mass of Mormon crickets on the move in Idaho in the spring of 2005, they learned that something more complicated was going on. “It looks like this big cooperative behavior,” says Iain Couzin, a research fellow at the Collective Animal Behaviour Laboratory in Oxford University’s zoology department and a member of the Idaho team. “You can almost imagine it like a group of army ants, sweeping out to find food. But in actual fact we found out it’s driven by cannibalism.” What looks like cooperation turns out to be extreme competition.

Crickets choose food carefully based on their nutritional needs at the moment, and they often find themselves wanting in the protein and salt departments. One of a cricket’s best sources for protein and salt, it turns out, is its neighbor. “They’re getting hungry and they’re trying to eat each other,” says Couzin, an affable Scotsman wearing a faded “Death to the Pixies” T-shirt, in his small office. “If you’re getting eaten, the best thing for you to do is to try and move away. But if you’re also hungry and trying to eat, the best thing to do is move away from others that are trying to eat you, but also to move toward others to try and eat them.” For crickets in the back of the pack, crossing over ground that has already been stripped of food by those in the front, another cricket may be the only meal in sight.

This seems a recipe for anarchy, not well-coordinated movement. What is actually happening is an example of the phenomenon known as “emergent behavior,” or the formation of complex systems, like cricket bands, that “emerge,” often unexpectedly and unpredictably, from the simple interactions of the individuals. Looking at the swarm as a whole, one might not easily see what is driving the movement. Nor could one necessarily predict by studying the local set of rules guiding each cricket’s behavior—eat thy neighbor and avoid being eaten by thy neighbor—that this would all end up as a tight swarm.

For complex systems to work the way they do, they need all, or at least a good number, of their component parts to play by the rules. Think of the “wave” at football stadiums, which begins, studies have shown, on the strength of a few dozen people; nobody knows, however, how many waves simply died for lack of participation, or because they tried to go in the “wrong” direction. What if some crickets got tired of avoiding their neighbors’ ravenous jaws and decided to leave the swarm? Some of Couzin’s colleagues hooked up small radio transmitters to a number of individual crickets, which were then separated from the larger band. Roughly half of those separated were killed by predators within days. Among the radio-tagged crickets kept within the band, none died. So whatever the risk of being eaten by one’s neighbors, no matter how stressful and unpleasant the experience, it’s still a better option than going solo.

What’s remarkable about the formation of these systems is how quickly the rules—and the form of the group—can change. Another insect Couzin has studied, both in the Oxford lab and in the wild in Mauritania, is the desert locust (Schistocerca gregaria). These locusts have two personalities. In their “solitarius” phase, they’re harmless. They live rather quietly, in small, scattered groups. “They’re shy, cryptic green grasshoppers,” Couzin says. But under certain conditions, such as after a drought, these Dr. Jekylls of the insect world, driven into closer contact by the search for food, will turn into a vast brown horde of marauding, “gregarious” Mr. Hydes. The impact is massive: Swarming locusts may invade up to 20 percent of Earth’s land surface at a time, Couzin says, affecting the livelihood of countless people. Knowing why and how these swarms form might help scientists predict where and when they will form. And so the team assembled a large group of Oxford-raised locusts, put them in an enclosed space, and used custom tracking software to follow what was going on.

When there are few locusts, they keep to themselves, marching in different directions, “like particles in a gas,” says Couzin. But when forced to come together, whether in a lab or because food has become scarce in the wild, interesting things start to happen. “The smell and sight of other individuals, or the touch on the back leg, causes them to change behavior,” Couzin says. “Instead of avoiding one another, they’ll start being attracted to each other, and this can cause a sort of cascade.” Suddenly, once the locusts reach a “critical density,” they will spontaneously start to march in the same direction.

Now what does all this have to do with traffic? you may be asking. The most obvious answer is that what the insects are doing looks a lot like traffic and that what we are doing on the road looks a lot like collective animal behavior. In both cases, simple rules govern the flow of the society, and the cost for violating those rules can be high. (Picture the highway police car or crashes in the role of predator.) Insects, like humans, are compelled to go on the move because they need to survive. Similarly, if we did not need to provide for ourselves, many of us would probably not choose to drive at the very same time everyone else is. Like insects, we have decided that moving in groups—even if most of us are alone in our own cars—makes the most sense. Virtually since traffic congestion began, plans have been put forward to stagger work schedules so that everyone is not on the roads at the same time, but even today, with telecommuting and flextime, traffic congestion persists because having a shared window of time during which we can easily interact with one another still seems the best way to conduct business.

In both insect and human vehicular traffic, large patterns contain all kinds of hidden interactions. A subtle change in these interactions can dramatically affect the whole system. To go back to the comparison between the Late and the Early Merge, if each driver simply adheres to one rule instead of another—merge only at the last moment instead of merge at your earliest opportunity—the merging system changes significantly. Like the pattern of locusts’ movement, human traffic movement often tends to change at a point of critical density. In a reversal of the way that locusts go from disorder to order with the addition of a few locusts, with the addition of just a few cars, smoothly flowing traffic can change into a congested mess.

The locust or cricket commuter, by staying within a potentially cannibalistic traffic flow, is, as Couzin suggests, clearly making the best of a bad situation. And in many ways, we act like locusts. Our seeming cooperativeness can shift to extreme competition in the blink of a taillight. Sometimes, we may be those harmless Dr. Jekylls, minding our own business, keeping a safe distance from the car in front. But at a certain point the circumstances change, and our character changes. We become Mr. Hyde, furiously riding up to the bumper of the person in front of us (i.e., trying to eat them), angry at being tailgated (i.e., trying to avoid being eaten), wishing we could leave the main flow but knowing it is still probably the best way home. One study, taken from highways in California, showed a regular and predictable increase in the number of calls to a road-rage hotline during evening rush hours. Another study showed that on the same stretch of highway, drivers honked less on the weekend than during the week (even after the researchers adjusted for the difference in the number of cars).

Another creature does things differently, taking the high road in traffic. This is the New World army ant, or Eciton burchellii, and these insects may just be the world’s best commuters. Army ant colonies are like mobile cities, boasting populations that can number over a million. Each dawn, the ants set out to earn their trade. The morning rush hour begins a bit groggily, but it quickly takes shape. “In the morning you have this living ball of ants, up to five feet high, perhaps living in the crevice of a tree,” says Couzin, who has studied the ants in Panama. “And then the ants just start swarming out of the nest. Initially, it’s like a big amoeboid, just seething bodies of ants. Then after a period of time they seem to start pushing out in one direction. It’s unclear how they choose that direction.”

As the morning commuters spread out, the earliest ones begin to acquire bits of food, which they immediately bring back to the nest. As other ants continue pushing into the forest, they create a complex series of trails, all leading back to the nest like branches to a tree trunk. Since the ants are virtually blind, they dot the trails with pheromones, chemicals that function like road signs and white stripes. These trails, which can be quite wide and long, become like superhighways, filled with dense streams of fast-moving commuters. There’s just one problem: This is two-way traffic, and the ants returning to the nest are laden down with food. They often move more slowly, and often take up more space, than the outbound traffic. How do they figure out which stream will go where, who has right-of-way, on “roads” they have only just built?

Interested in the idea that ants may have evolved “rules to optimize the flow of traffic,” Couzin, along with a colleague, made a detailed video recording of a section of army ant trail in Panama. The video shows that the ants have quite clearly created a three-lane highway, with a well-defined set of rules: Ants leaving the nest use the outer two lanes, while ants returning get sole possession of the center lane. It is not simply, says Couzin, that the ants are magically sticking to their own chemical-covered separate trails (after all, other types of ants do not form three lanes). Ants are attracted to the highest concentration of chemicals, which is where the highest density of ants tends to be, which happens to be the center lane.

A constant game of chicken ensues, with the outbound ants holding their ground against the returning ants until the last possible moment, then swiftly turning away from the oncoming traffic. There is the occasional collision, but Couzin says the three-lane structure helps minimize the subsequent delay. And ants are loath to waste time. Once finished with the evening commute, home by dusk, the entire colony moves, in the safety of darkness, to a new site, and the next morning the ants repeat the cycle. “These species have evolved for thousands of years under these highly dense traffic circumstances,” says Couzin. “They really are the pinnacle of traffic organization in the actual world.”

The secret to the ridiculous efficiency of army ant traffic is that, unlike traveling locusts—and humans—the ants are truly cooperative. “They really want to do what’s best for the entire colony,” says Couzin. As worker ants are not able to reproduce, they all labor for the queen. “The colony in a sense is the reproductive unit,” Couzin explains. “To take a loose analogy, it’s like the cells in your body, all working together for the benefit of you, to propagate your genes.” The progress of each ant is integral to the health of the colony, which is why ant traffic works so well. No one is trying to eat anyone else on the trail, no one’s time is more valuable than anyone else’s, no one is preventing anyone else from passing, and no one is making anyone else wait. When bringing back a piece of food that needs multiple carriers, ants will join in until the group hits what seems to be the right speed. Ants will even use their own bodies to create bridges, making the structure bigger or smaller as traffic flow passing over it requires.

What about merging? I ask Couzin later, in the dining room at Balliol College. How are the ants at this difficult task? “There’s definitely merging going on,” he says with a laugh. “There seems to be something interesting going on at junctions. It’s something we’d like to investigate.”

Playing God in Los Angeles

Doesn’t matter what time it is. It’s either bad traffic, peak traffic, or slit-your-wrists traffic.

—The Italian Job (2003)

“Sorry, the traffic was horrible.” These five words rival “How are you?” as the most popular way to begin a conversation in Los Angeles. At times it seems like half the city is waiting for the other half to arrive.

But there is one night when being late simply will not do, when the world—or at least several hundred million inhabitants of it—wants everyone to get to the same place at the same time. This would be Oscar night, when eight hundred or so limousines, ferrying the stars, arrive in a procession at the corner of Hollywood and Highland, depositing their celebrity carriage at the Kodak Theater. On the red carpet, the media volley questions: “How are you feeling?” “Who are you wearing?” But on Oscar night no one ever asks a larger question: How did eight hundred cars get to the same party in a punctual fashion in Los Angeles?

The answer is found in the labyrinthine basement of City Hall in downtown L.A. There, in a dark, climate-controlled room with a wall-sized bank of glowing monitors, each showing strategic shots of intersections across the city, sits the brains of the Los Angeles Department of Transportation’s Automated Traffic Surveillance and Control (ATSAC). Traffic centers like this one are essential in many modern cities, and one sees similar setups from Toronto to London (in Mexico City the engineers delightedly showed me footage of speeding drivers giving the finger to automatic speed-limit cameras).

The ATSAC room in Los Angeles would normally be empty on a Sunday, with only the quietly humming computers running the city’s traffic lights—ATSAC will even call a human repairperson if a signal breaks down. But since it’s Oscar night, an engineer named Kartik Patel has been in the “bunker” since nine a.m., working on the DOT’s special Oscar package. Another man lurks at a desk and does not say much. Teams of engineers have also been deployed in the field at strategic intersections. On a desk sits a little statue of Dilbert at a computer, to which someone has attached a label: “ATSAC Operator.”

Since the city cannot shut down the entire street network for the Oscars, the limos must be woven through the grid of Los Angeles in a complex orchestration of supply and demand. Normally, this is done by the system’s powerful computers, which use a real-time feedback loop to calculate demand. The system knows how many cars are waiting at any major intersection, thanks to the metal-detecting “induction loops” buried in the street (these are revealed by the thin black circles of tar in the asphalt). If at three-thirty p.m. there are suddenly as many cars as there normally would be in the peak period, the computers fire the “peak-period plan.” These area-wide plans can change in as little as five minutes. (For a quicker response, they could change with each light cycle, but this might produce overreactions that would mess up the system.) As ATSAC changes the lights at one intersection, it is also plotting future moves, like a traffic version of IBM’s chess-playing computer Big Blue. “It’s calculating a demand,” says Patel. “But it needs to think ahead and say, ‘How much time do I need for the next signal?’”

Over time, ATSAC amasses a profile of how a certain intersection behaves during a given time on a given day. Patel points to a computer screen, which seems to be running a crude version of the game SimCity, with computer renderings of traffic lights and streets but no people. An alert is flashing at one intersection. “This loop at three-thirty on a Sunday has a certain historical value, for a year’s period of time,” Patel explains. “Today it’s abnormal, because it’s not usually that heavy. So it’ll flag that as out of the norm and post it up there as a possible incident.” It will try to resolve the problem, says Patel, within the “confines of the cycling.”

But on this occasion, the engineers want certain traffic flows—those conveying the stars’ limos—to perform better than ATSAC would normally permit, without throwing the whole system into disarray. In the late afternoon, with the ceremony drawing near, it becomes apparent just how difficult this is. Harried requests are beginning to come in from field engineers, who are literally standing at intersections. “ATSAC, can you favor Wilcox at Hollywood?” asks a voice, crackling from Patel’s walkie-talkie. Patel, on his cell phone, barks: “Man, did you happen to copy Highland and Sunset? There’s quite a queue going northbound.” At times Patel will have his cell phone in one hand, the walkie-talkie in another, and then the landline phone will ring. “The limos are starting to back up, almost at Santa Monica,” someone cries through the static.

As Patel furiously taps on his keyboard, lengthening cycle times here, canceling a left-turn phase there, it becomes hard to resist the idea that being a traffic engineer is a little like playing God. One man pushing one button affects not just one group of people but literally the whole city, as the impact ripples through the system. It is chaos theory, L.A. style: A long red light in Santa Monica triggers a backup in Watts.

This is when it begins to look as if something odd is going on here this afternoon. Patel seems particularly concerned with the intersection of La Brea Avenue and Sunset Boulevard. “Yeah, Petey, what’s up?” he shouts into his phone. “How many people are there? That’s good.” Patel then admits that his unit has a “labor problem.” Some three hundred municipal engineers, on a sick-out, are picketing on the same streets on which the limos are trying to get to the Oscars. What better way to draw attention, and who better to know the streets on which to demonstrate? Some of the calls Patel receives are from engineers wondering why the limos have been held up, and some of the calls are from picketing engineers seeking updates about which intersections they should cross on foot. “Tell them to walk more slowly, they’re going too damn fast,” Patel says into his phone. Reports coming in say that police are hustling the picketers across the intersections, so as to not block traffic. “Oh my God, how can they kick you out? You have a legal right to cross. Any unmarked crosswalk, you can cross it…just keep on crossing there, moving slowly.”

Patel is both trying to get the limos to their destination and coaching the picketers on how to best interrupt that progress. Does that mean he can give the sign-toting pedestrians more time, which would further their cause? A strange smile crosses Patel’s face, but he says nothing. He later excuses himself and goes to an office in the back, where he takes phone calls. Is he a coconspirator? Or does his traffic-engineer side override his labor solidarity side? One cannot say for sure, but interestingly enough, Patel and another engineer were later charged with tampering with traffic lights at four key intersections as part of the ongoing labor dispute, and the case, which attracted the attention of the Department of Homeland Security, was in criminal court as of this writing, with the defendants facing several years in prison if convicted.

Despite the picketers, the limos arrive on time. The winning picture, ironically, is Crash, a film about Los Angeles traffic on literal and metaphorical levels. Then the limos leave the Kodak Theater, rejoining the city’s traffic, and head for the postevent parties.


That Oscar afternoon was a small but perfect illustration of how complicated human traffic is when compared to ant traffic. Ants have evolved over countless centuries to move with a seamless synchronicity that will benefit the entire colony. Humans, on the other hand, propel themselves around artificially, something they have done for only a few generations. They do not all move en masse with the same goal but instead travel with their own agendas (e.g., getting to the Oscars, staging a demonstration). Ants all move at roughly the same speed, while humans like to set their own speeds, ones that may or may not reflect the speed limit. And, crucially, ants move as ants. They can always feel their neighbors’ presence. Humans separate themselves not only across space but into drivers and pedestrians, and tend to act as if they are no longer the same species.

Los Angeles, like all cities, is essentially a noncooperative network. Its traffic system is filled with streams of people who desire to move how they want, and where they want, when they want, regardless of what everyone else is doing. What traffic engineers do is to try to simulate, through technology and signs and laws, a cooperative system. They try to make us less like locusts and more like ants.

Take traffic signals. It’s common to hear drivers in Los Angeles, as elsewhere, lament, “Why can’t they time the signals so they’re all green?” The obvious problem with so-called synchronized signals is that there is a driver moving in a different direction asking the same thing. Two people are competing for the same resource. The intersection, the fundamental problem of the traffic world, is an arena for clashing human desire. John Fisher, the head of the city’s DOT, uses the analogy of an elevator in a tall building. “You get on the elevator, and it stops at every floor because someone presses the button. They want to get off or on. Now, it stops at every floor—is it synchronized or not synchronized? The reality is if there are many stops, it’s going to take a while to get there. It’s the same with signals.”

Engineers can use sophisticated models to squeeze as much “signal progression” as possible out of a network, to give the driver the “green wave.” Fisher says that when he came to the DOT in the 1970s, “we tried to hold the line and keep the signals at a quarter-mile spacing.” By doing that, and setting the cycle time (or the time it takes to cycle through green, yellow, and red on the traffic light) at sixty seconds, vehicles traveling at 30 miles per hour could reasonably “expect to find a green.”

But over time, as the city has grown more dense, so too has the pressure to add more traffic lights. In certain places there is now a light at every block, which means there is a potential demand to cross at every block. Engineers have been forced to expand the length of the cycle to ninety seconds—typically the maximum in cities. “Let’s say you go to a ninety-second cycle,” Fisher says. “Even if you have quarter-mile spacing it means your progressive speed is not thirty miles per hour anymore, but something like twenty miles per hour. If you complicate that further, and the signal spacing is every block or sixteenth of a mile, there’s just no way you could progress from one end to the other. The best you could do is go a couple signals and stop, a couple signals and stop, in all directions.” The green wave works well on major streets where the demand from side streets is small. But in Los Angeles, Fisher explains, “we have traffic going in all directions, and generally the same quantities.” Some intersections receive so much competing demand that they are “oversaturated,” as Fisher says, beyond the help of even ATSAC’s computers.

To further complicate matters, there are, even in Los Angeles, pedestrians. Despite the hilarious scene in L.A. Story that showed Steve Martin driving to his next-door neighbor’s house for dinner, people do walk, and not just to and from their parked car. As a profession, traffic engineering has historically tended to treat pedestrians like little bits of irritating sand gumming up the works of their smoothly humming traffic machines. With a touch of condescending pity, pedestrians are referred to as “vulnerable road users” (even though in the United States many more people die in cars each year, which leads one to wonder who exactly is more vulnerable). Engineers speak of things like “pedestrian impedance” and “pedestrian interference,” which sound like nasty acts but really just refer to the fact that people sometimes have the gall to cross the street on foot, thus doing things like disrupting the “saturation flow rate” of cars turning at an intersection.

As a testament to the inherent bias of the profession, no engineer has ever written a paper about how “vehicular interference” disrupts the saturation flow rates of people trying to cross the street. In cities like New York, despite the fact that pedestrians vastly outnumber cars on a street like Fifth Avenue, traffic signals are timed to help move the fewer cars, not the many pedestrians—has anyone ever had an uninterrupted stroll up Fifth Avenue, a green wave for walking? Unlike in pedestrian-thronged New York City, where most push buttons to cross the street no longer work (even though they still tempt the impatient New Yorker), in Los Angeles the relative rarity of pedestrians means the buttons do work. The walker humbly asks the city’s traffic gods for permission to cross the street, and, after a time, their prayers are answered. If you do not press the button, you will stand there until you’re eventually ticketed for vagrancy.

Sometimes the traffic deities encounter even higher authorities. A curious fact of Los Angeles traffic life is that, at roughly seventy-five signals, in places ranging from Century City to Hancock Park, the button does not always have to be pressed to cross. These intersections run instead on what is known as Sabbath timing. As Sabbath-observant members of the Jewish faith are not supposed to operate machines or electrical devices from sundown on Friday to sundown on Saturday, or during a number of holidays, the act of pressing a button to cross the street is viewed as a violation of this tenet. With the only alternative rampant jaywalking, the city installed automated “Walk” signs at certain intersections (causing what Fisher jokingly calls “sacrificial interruptions” to traffic flow even when no pedestrians are present). “We have the Hebrew calendar programmed into our controller,” Fisher told me.

When the DOT suggested installing “smart” devices that would sense the presence of a pedestrian at a crosswalk and activate a flashing signal, it was gently rebuffed by the Rabbinical Council of California, which opined that activating the light via a signal, even if it was done passively, violated the Sabbath restrictions. If pedestrians were unaware that their presence was triggering the device, the council noted, the smart device would be acceptable, but “people would quickly realize its presence and avoid using the crosswalk on the Sabbath.”

These nuances pale before the overwhelming fact that Los Angeles is handling more traffic now than was ever thought possible. “A lot of major streets, like La Cienega and La Brea, carry sixty thousand vehicles a day,” says Fisher. “Those streets were designed to carry thirty thousand vehicles a day.” Years ago, engineers used capacity-expanding tricks like reversible lanes on Wilshire Boulevard and other major thoroughfares, changing the normal direction of one lane to help carry traffic in from the freeway in the morning and send it back out in the evening. That is no longer possible. “When you’re getting a split like sixty-five percent of traffic one way, thirty-five percent the other way, reversible lanes work very well,” Fisher says. “Today we rarely have that type of peaking anywhere in the city.” The highways are no different. The San Diego Freeway, or I-405, was projected to carry 160,000 vehicles when it was completed at the end of the 1960s. It now carries almost 400,000 per day, and the junction where it connects to the Santa Monica Freeway is the most congested in the United States. The Santa Monica used to be a traditional sort of urban highway, with a heavier morning peak toward downtown and the reverse in the afternoon. “You try to go outbound in the morning and that often seems heavier than the inbound is,” Fisher says.

“We used to have typical days where we would give volumes,” notes Dawn Helou, an engineer with Caltrans, the sprawling and omnipresent agency in charge of California’s highways. “A typical day is Tuesday, Wednesday, Thursday in a month preferably without holidays, in a week preferably without holidays. No rain, no holidays, no summer vacation, no incidents. We’re running out of those typical days.”

The thing that keeps the whole system from breaking down is precisely that advantage that humans have over ants: the ability to see, and direct, the whole traffic system at once. By making all these decisions for the drivers, by coordinating the complex ballet of wants and needs, supply and demand at intersections, engineers have been able to improve the city’s traffic flow. A study a few years ago by the DOT showed that the area containing real-time traffic signals reduced travel times by nearly 13 percent, increased travel speeds by 12 percent, reduced delay by 21 percent, and cut the number of stops by 31 percent. Just by quickly alerting the DOT that signals have malfunctioned, the system squeezes out more efficiency. What the traffic engineers have done is added “virtual” capacity to a city that cannot add any more lanes to its streets.

The flow of information is crucial to maintaining the flow of traffic. With no spare capacity, irregularities in the system need to be diagnosed and addressed as soon as possible. Engineers at Caltrans say that as a rule of thumb, for every one minute a highway lane is blocked, an additional four to five minutes of delay are generated. The inductor loops buried in the highway can and do detect changes in the traffic patterns. But the highway loops are not in real time. There can be a gap of anywhere from a few minutes to a quarter of an hour before the information they’re recording is processed. Often, visual confirmation by camera is needed to verify that there is a problem. In that time, a huge jam could develop. Or sometimes the loops in a particular section of highway are not working (Caltrans reports anywhere from 65 to 75 percent of its twenty-eight thousand statewide loops are working on a given day), or a section of highway will have no loops at all.

This is why, each day in Los Angeles, there is a frantic search for the truth. It’s called the traffic report. Traffic news is the sound track of daily life in Los Angeles, a subliminal refrain of “Sig Alerts” and “overturned big rigs” always on the edge of one’s consciousness. Occasionally the story is that there is no story, according to Vera Jimenez, who does the morning traffic on KCAL, the CBS affiliate in Los Angeles. “Sometimes it’s funny,” she said one morning at the Caltrans building. “The story is not that the traffic’s really heavy, but, oh my gosh, it’s surprisingly light. It’s not a holiday, there isn’t anything going on, it’s just really light. Everyone’s driving the way they should, everyone’s merging, and believe it or not, look how nice it is.”

No city in the world has more traffic reports or traffic reporters than Los Angeles, and to spend time with them is to see the city, and traffic, in a new way. Early one morning, I drive to Tustin, an Orange County suburb that is home to Airwatch, a Clear Channel subsidiary and one of America’s largest traffic-reporting services. In a room filled with banks of televisions, computer monitors, and police scanners, Chris Hughes is several hours into the morning rush hour. Armed with a stopwatch and jittery from caffeine, Hughes rattles off a fast, well-calibrated, flow: “Heavy traffic in Long Beach this morning on the North 405 through Woodruff to the 710 then again from the 110 Freeway heading up to Inglewood…”

For each of the different radio stations for which Hughes reports, he must change the length of his report, as well as the way he says it. One station wants “upbeat and conversational,” while another wants a precise robotlike diction they call “traffic formatics.” Some stations have advertisements for Hooters Casino, but the Christian stations do not. Some stations actually want him to be someone else. “Good morning, I’m Jason Kennedy with AM 1150 traffic brought to you by Air New Zealand,” I suddenly hear him say. “They’re sort of competing stations,” he explains sheepishly, “even though we own them both.”

Hughes has an instinctual understanding of Los Angeles’ highways. He can tell which way a rainstorm is moving by looking at the real-time traffic-flow highway maps. He knows Fridays heading east out of the city can be particularly bad. “Everyone’s going to Las Vegas—all the way to ten p.m. that’ll be backed up.” He knows that people drive slower on highway stretches that have sound barriers to either side. He knows that mornings with heavy rains often lead to lighter afternoon traffic. “Maybe a lot of people got scared of the rain and disappeared,” he says. He notes that while traffic information is easily available to the public, often the trick is in understanding it. “It’s kind of like The Matrix,” he says. “You’re looking at the map and you can pick out what looks right and what doesn’t. I can look at the map now and say, ‘Hey, there’s something wrong on the 101. A big-rig fire at Highland, probably.’”

There is no limit to the things that can disrupt the flow on Los Angeles highways. “Do you want to know the number one specific item dropped on the freeway?” asks Claire Sigman, another Airwatch reporter. “The most recorded item is ladders.” Trucks, just like in the Beverly Hills Cop movies, also spill avocados and oranges. Portable toilets have been dumped in the middle of the freeway. In 2007, a house, replete with graffiti and a “For Rent” sign, sat for weeks on the Hollywood Freeway, abandoned during the course of its move after it struck an overpass (the owner had taken a detour onto an unauthorized route). People hold apocalyptic signs on overpasses, or threaten to jump. Wildfires break out. Out in the high desert, tumbleweeds cause problems. “People swerve out of the way, rather than just drive through it,” Hughes says. A computer screen at the Airwatch office ticks off a steady flow of traffic incidents, ranging from the absurd to the horrifying, as recorded by the California Highway Patrol (CHP). Codes are used to disguise the presence of stalled female drivers, who might otherwise be preyed upon by unsavory men listening to police scanners. Not atypical of the stream is incident 0550, which describes a “WMA,” or white male, wearing a plaid jacket and “peeing in middle of fwy.” It adds a noteworthy detail: “No veh in sight.” (Now, where was that wayward Porta Potti?)

CHP officers are the foot soldiers in the daily battle to keep Los Angeles’ traffic from collapsing. The sophisticated computer modeling and fiber-optic cable that the traffic generals in the bunker have at their disposal are of little use when a car has stalled on Interstate 5, as I learned one afternoon when I went out for a patrol in a CHP cruiser with Sergeant Joe Zizi, an easygoing former trooper now doing public relations. CHP patrol officers begin each day by “cleaning their beat,” or removing any abandoned vehicles or hazards from the road. “That way there’s nothing that people have to look at when they’re driving,” Zizi says as he drives along the 101. Something as simple as a couch dumped in a roadside ditch can send minor shudders of curiosity through the traffic flow. A standard-issue black pump-action shotgun sits between the front seats. To enable drivers to carry out their traffic triage duties, patrol cars are outfitted with reinforced bumpers, designed to let them push cars off the road rather than wait for a tow truck. Their trunks are filled with a dizzying array of equipment for dealing with traffic contingencies, ranging from baby-delivering kits (“definitely a spectacle for rubberneckers”) to dog snares.

“For some reason, dogs are attracted to the freeway,” says Zizi. “They get on there, get completely freaked out, and start running down the center.” According to CHP statistics, these Code 1125-As (traffic hazard—animal) peak on July 5, presumably from dogs scared by the previous night’s fireworks. When traffic is moving, CHP officers pass the time by looking for stolen vehicles (screwdrivers in the ignition are a telltale sign) and, of course, writing traffic tickets. Does Zizi have any advice for beating tickets? “I have a lot of officers who say that women crying will get them out of tickets, while other officers say that if someone does cry they’re getting the ticket,” he says. “Of course we have a lot of men who cry trying to get out of tickets, but that really doesn’t work on the heartstrings of officers.”

For all the Caltrans cameras and loops wired into the road, for all the CHP officers flagging incidents, the highway system running through Los Angeles is so vast and incomprehensible that, sometimes, the only way to really understand what’s happening is to pull way back and view the whole system from above. That is why there is still a place for people like Mike Nolan, KFI’s “eye in the sky,” a longtime L.A. traffic reporter who, twice daily, will take off in his Cessna 182 from Riverside County’s Corona Airport and cover a swath of ground from Pasadena to Orange County.

“The learning curve is being able to read a freeway,” he explains, banking his plane over a new subdivision carved into green hillside. “I know what’s normal. I know where it should be slowing down and where it shouldn’t. When I see something out of the ordinary, then I investigate it.” Nolan, whose navigational mantra is “Keep the freeway to your left,” knows traffic patterns like a grizzled fishing guide knows the best bass holes. A stalled Volkswagen in East Los Angeles is worse than an overturned oil truck in La Cañada (“More spectacular does not necessarily translate into worse,” he says). Mondays, especially during Monday Night Football, tend to be a bit lighter. Thursday, congestion-wise, is now looking like the new Friday, traditionally the busy “getaway day.” There are also strange blips in the pattern, like sunrise slowdowns. “The very first day of standard time, when we go from daylight saving time to darkness, everybody just locks up,” he says. “The traffic goes from bad to horrendous.” Rainy days can be bad, but the first rainy day in a while is even worse. “There’s a buildup of oil and rubber if it hasn’t rained in a while. It’s like driving on ice, literally.”

Nolan says people have long been predicting, because of ground sensors and in-vehicle probes that can detect the speed of traffic, that there will no longer be a need for aerial traffic reports. Indeed, on his instrument panel he has attached a TrafficGauge, a Palm Pilot-sized device fed by Caltrans data, that shows congestion levels on L.A. freeways. But he says that data rarely tell the whole story, or the correct story. “In my mind there’s no substitute for looking out the window and telling people what you’ve got,” he says. “The sensors in the road are delayed, they’re inefficient. They’re working half the time, not working half the time. There’s no substitute for saying, ‘It’s in the right lane, I see it right there, right at the fill-in-the-blank overpass.’ Or that the tow truck is in heavy traffic. The sensor can’t tell you the tow truck’s a block away, or ready to hook up and pull away. It can’t give you the substantive info that comes from looking at it directly.”

Indeed, that afternoon of flying around the city, accompanied by an Airwatch reporter receiving ground reports, seems to be an exercise in chasing ghosts. The jackknifed tractor-trailer on the 710 is not there, or never was there. The blockage on the 405 was a rumor. Nolan is the one who must try to make sense of the strange reports that come in, like the one that announced a dead dog was “blocking lanes one, two, three, and four.” The most remarkable traffic event he ever saw was during the L.A. riots of 1992. “I remember seeing people stop at a stoplight in Hollywood. They would get out and loot a store. The light would turn green and they’d get back in and drive away. That was the most incredible thing I’ve ever seen.”

Flying over a city like Los Angeles, it is easy to glance down and think, for a moment, that the people below, streaming along trails, look like ants. If only it were that simple.

When Slower Is Faster, or How the Few Defeat the Many:
Traffic Flow and Human Nature

You hit the brakes for a second, just tap them on the freeway, you can literally track the ripple effect of that action across a two-hundred-mile stretch of road, because traffic has a memory. It’s amazing. It’s like a living organism.

Mission: Impossible III

At some point you may have come to a highway on-ramp, expecting to join the flow of traffic, only to be stopped by a red light. Such devices are called ramp meters, and they are found from Los Angeles to South Africa to Sydney, Australia.

Ramp meters often seem frustrating because the traffic on the highway appears to be moving just fine. “People ask me, ‘How come you’re stopping me at the ramp meter? The freeway is free-flowing,’” says Dawn Helou, the Caltrans engineer. “The freeway is free-flowing because you’re stopping.”

This is one of the most basic, and often overlooked, facts about traffic: That which is best for an individual’s interest may not be best for the common good. The game traffic engineers play to fight congestion involves fine-tuning this balance between what is “user optimal” and what is “system optimal.” This happens on several different levels, both having to do with congestion: how traffic moves on roads and how larger traffic networks behave (an idea I’ll return to in a later chapter).

The reason why highway ramp meters work is, on the face of it, simple once one knows a few basic facts about traffic flow. Engineers have been trying to understand, and model, traffic flow for many decades, but it is a huge and surprisingly wily beast. “Some puzzles remain unsolved,” declares Carlos Daganzo, an engineer at the University of California, Berkeley. The first efforts merely tried to model the process known as “car following.” This is based on the simple fact that the way you drive is affected by whether or not someone is in front of you, and how far away or close they are. Like ants responding to the presence of pheromones on the trail, you’re influenced by the driver ahead, a constant, unsteady wavering between trying not to get too close and trying not to slip too far back. Now imagine those interactions, plus lane shifts and all the other driving maneuvers, a fluctuating mix of vehicle speeds and sizes, a wide range of driver styles and agendas, a dizzying spectrum of differing lighting and weather and road conditions; then multiply all this by the thousands, and you can begin to appreciate the higher-order complexities of traffic modeling.

Even the most sophisticated models do not fully account for human weirdness and all the “noise” and “scatter” in the system. Traffic engineers will offer caveats, like the disclaimer I saw at one traffic conference: “This model does not account for the heterogeneity of driver behavior.” Do you feel uncomfortable driving next to someone else, and therefore speed up or slow down? Are you sometimes willing, for no apparent reason, to ride quite close to the car in front, before gradually drifting back? All kinds of strange phenomena lie outside easy capture by the traffic sensors. Car following, for instance, is filled with little quirks. A study that looked into how closely passenger-car drivers followed SUVs found that car drivers, contrary to what they said they did—and despite the fact that the SUV was blocking their view of the traffic ahead—actually drove closer to SUVs than when they followed passenger cars.

Or take what Daganzo has called the Los Gatos effect, after an uphill stretch of highway in California. You may have experienced this: Drivers seem reluctant to abandon the passing lane and join the lane of trucks chugging uphill, even when they are being pressured by other drivers, and even when the other lane is not crowded. What’s going on? Drivers may not want to give up the fast lane for fear of having trouble returning to it. They may also be unsure whether the person behind truly wants to go faster or is just keeping a tight space to prevent someone else from passing. A tight “platoon” forms, but for how long? We all see these odd patterns. One of the idiosyncrasies I have noticed in traffic flow is something I call “passive-aggressive passing.” You’re in the passing lane when suddenly the driver behind you pressures you to move into the slower right-hand lane. After you have done so, they then move into your lane, in front of you, and slow down, thus forcing you to pass them.

The basic parameters of how highways perform have been gradually hammered out. One of the key performance measures is volume, also called flow, or the number of vehicles that pass a buried sensor or some other fixed point on the highway. At four a.m., before rush hour, cars may be zipping along a highway at 75 miles per hour. The volume is measured at 1,700 cars moving past a point in one hour. As rush hour begins, the volume quite naturally begins to rise in an upward curve, reaching a theoretical maximum of 2,400 cars traveling at 55 miles per hour. System-wise, this is traffic nirvana. Then, as additional vehicles enter the highway, the curve begins to drop. Suddenly, the volume is back at 1,700. This time the cars are going 35 miles per hour. “So you have the two 1,700s,” Helou says. “Same volume, completely different situation.”

Because traffic moves in time and space, measurements like volume can be deceiving, as can the highway itself. Solo drivers sitting in a highly congested lane may look to the HOV lane next to them and think that it’s empty—a psychological condition so prevalent it even has a name, “empty lane syndrome.” Many times it just seems empty because of the large headways between vehicles moving at much higher speeds. That lane may actually be achieving the same volume as the lane you are in, but the fact that the drivers might be going upward of 50 miles per hour faster creates an illusion that it’s being underused. Of course, neither of these positive or negative individual outcomes—the driver whisking along at 80 miles per hour or the people stuck at 20 miles per hour in the congested lanes—are what’s best for the entire system. The ideal highway will move the most cars, most efficiently, at a speed just about halfway.

Even as rush hour kicks in and the speed-flow curve begins to drop, traffic can perk along at what has been called “synchronized flow,” heavy but steady. But as more vehicles pile onto the highway from on-ramps, the “density,” or the number of cars actually found in a one-mile stretch (as opposed to passing a single spot), begins to thicken. At a certain point, the critical density (the moment, you will recall from before, when the locusts began their coordinated march), the flow begins to break down. Bottlenecks, fixed or moving, squeeze the flow like a narrowing pipe. There are simply too many cars for the road’s capacity.

Ramp metering aims to keep the highway’s “main-line flow” below the critical density by not letting the system be flooded with incoming on-ramp cars. “If you allow unimpeded access, then you have a platoon of vehicles that are entering the main line,” says Helou. This means not only more cars but more cars jockeying to merge. Studies have shown that this is neither predictable nor always cooperative. “That [merging] eventually breaks down the right lane,” she says. “This overflows to the next lane, because people try to merge left before they get to it. And then the people in the second lane try to merge to the next lane before they get to it, so you break down the whole freeway.” A line of cars waiting to exit an off-ramp can trigger this same chain reaction, one study showed, even when all the other lanes were flowing nowhere near critical density.

If done properly, ramp metering, by keeping the system below the critical density, finds that sweet spot in which the most vehicles can move at the highest speed through a section of highway. Engineers call this “throughput maximization.”

A simple way to see this in action involves rice. Take a liter of rice and pour it, all at once, through a funnel and into an empty beaker. Note how long it takes. Next, take the same rice and pour it not all at once but in a smooth, controlled flow, and time that process. Which liter of rice gets through more quickly? In a demonstration of this simple experiment by the Washington DOT, it took forty seconds for one liter of rice to pass through the funnel using the first method. The second method took twenty-seven seconds, nearly one-third less time. What seemed slower was actually faster.

Rice has more to do with traffic than you might think. Many people use water analogies when talking about traffic, because it’s a great way to describe concepts like volume and capacity. One example, used by Benjamin Coifman, an engineering professor at Ohio State University who specializes in traffic, is to think of a bucket of water with an inch-wide hole in the bottom. If the inflow into the bucket is half an inch in diameter, no water will accumulate. Raise it to two inches, however, and the water rises, even though some water is still exiting. Whether we drive into a jam (or a jam drives into us) depends on whether the “water”—that is, the traffic trying to flow through a bottleneck—is draining or rising. “As a driver, the first thing you encounter is the end of the queue,” Coifman told me. “The first thing you encounter is wherever the water level happens to be that day.” The bucket metaphor also teaches us something else about traffic: No matter how much capacity there is in the rest of the bucket (or on the roads), the size of the hole (or the bottleneck) dictates what gets through.

At places like bottlenecks, however, traffic acts less like water (it does not speed up as highway “channels” narrow, for one) and more like rice: Cars, like grains, are discrete objects that act in peculiar ways. Rice is what’s called a “granular media,” a solid that can act like a liquid. Sidney Nagel, a physicist at the University of Chicago and an expert in granular materials, uses the analogy of adding a bit of sugar to a spoon. Pour too much, and the pile collapses. The sugar flows like a liquid as it collapses, but it’s really a group of interacting objects that do not easily interact. “They do not attract one another,” says Nagel. “All they can do is scatter off one another.” Put a bunch of granular materials together, and it is not easy to predict how they will interact. This is why grain silos are the building type most prone to collapse, and it’s also why my box of Cascadian Farm Purely O’s cereal begins to bow outward at the bottom after several pours.

Why does the rice jam up as you pour it into the funnel? The inflow of rice exceeds the capacity of the funnel opening. The system gets denser and denser. Particles spend more time touching one another. More rice touches more rice. The rice gets “hung up” from the friction of the funnel walls. Sound familiar? “That’s like cars on the highway,” says Nagel. “And when you get narrowing of traffic, then that becomes very much stuff trying to flow through the hopper.”

Pouring less rice at a time—or moving fewer cars—keeps more space, and fewer interactions, between the grains. Things flow faster. As intuitive as the “slower is faster” idea is, it’s not always easy for a driver stuck in traffic to accept. In 1999, a state senator from Minnesota, claiming that ramp metering in the Twin Cities was doing more harm than good, launched a “Freedom to Drive” proposal that called for, among other things, shutting down the meters. The legislation died, but under another bill a ramp-meter “holiday” was declared. For two months the meters were turned off. Drivers could enter the highway at will, on so-called sane lanes, unfettered by troublesome red lights. And what happened? The system got worse. Speeds dropped, travel times went up. One study showed that certain highway sections had double the productivity with ramp meters than without. The meters went back on.


The “slower is faster” idea shows up often in traffic. The classic example concerns roundabouts. Many people are under the mistaken impression that roundabouts cause congestion. But a properly designed roundabout can reduce delays by up to 65 percent over an intersection with traffic signals or stop signs. Sure, an individual driver who has a green light may fly through a signalized intersection much more quickly than through a roundabout. Roughly half the time, however, the light will not be green; and even if it is green there is often a rolling queue of vehicles just starting up from the previous red. Add to this such complications as left-turn arrows, which prevent the majority of drivers from moving, not to mention the “clearance phase,” that capacity-deadening moment when all lights must be red, to make sure everyone has cleared the intersection. Drivers do have to slow down as they approach a roundabout, but under typical traffic conditions they rarely have to stop.

In the 1960s, experiments were made at the Holland Tunnel, one of the main arteries for traffic coming into and leaving New York City. When cars were allowed to enter the tunnel in the usual way, with no restrictions, the two-lane tunnel could handle 1,176 cars per hour, at an optimal speed of 19 miles per hour. But in a trial, the tunnel authorities capped the number of cars that could enter the tunnel every two minutes to 44. If that many cars got in before two minutes were up, a police officer made the next group of cars wait ten seconds at the tunnel entrance. The result? The tunnel now handled 1,320 vehicles per hour. (I will explain why shortly.)

On streets with traffic signals, engineers set progressions with a certain speed in mind that will enable the driver to hit a line of constant greens. To drive faster than this only ensures that the driver will be forced to come to a stop at the next red light. Each stop requires deceleration and, more important, acceleration, which costs the driver in time and fuel. A queue of drivers stopped at a light is a gathering of “start-up lost time,” as engineers call it (in an appropriately forlorn echo of Proust). The first cars in a queue squander an average of two seconds each, two seconds that would not have been lost had the car sailed through at the “saturation-flow” rate. The first driver at a light that turns from red to green, because he must react to the change, make sure that the intersection is empty, and accelerate from a standstill, generates the most “lost time.” The light is green, but for a moment the intersection is empty. The second driver creates a bit less lost time, the third driver less still, and so on (assuming everyone is reacting as soon as they can, which is not a given). SUVs, because they are longer (on average, 14 percent longer than cars), and take longer to accelerate, can create up to 20 percent more lost time.

Some of the start-up lost time could be “found” if drivers approached at a slower, more uniform speed that did not require them to come to a stop. (If they came too slowly, however, time would also be lost, as green signal time would be wasted on an empty intersection.) Much of the time being lost these days is “clearance lost time,” the time between signals when the intersection is momentarily empty. This is because traffic engineers are increasingly lengthening the “all-red phase,” meaning that when one direction gets the red, the competing direction has to wait nearly two seconds before getting a green. They do this because more people cannot seem to stop on red.

Now picture a highway during stop-and-go traffic. Like those drivers stopped at the light, each time we stop and start in a jam we are generating lost time. Unsure of what the drivers ahead are doing, we move in an unsteady way. We are distracted for a moment and do not accelerate. Or we overreact to brake lights, stopping harder than we need to and losing more time. Drivers talking on cell phones may lose still more time through delayed reactions and slower speeds. The closer the vehicles are packed together, the more they affect one another. Everything becomes more unstable. “All of the excess ability for the system to take in any sort of disturbance is gone,” says Coifman. He uses the metaphor of five croquet balls. “If you put them a foot apart and tap one lightly, nothing happens to the other four. If you put them all up against one another and tap one lightly, the far one then moves out. When you get closer to capacity on the roadway, if there’s any one little tweak, it impacts a lot of the cars.”

When the first in a group of closely spaced cars slows or stops, a “shock wave” is triggered that moves backward. The first car slows or stops, and the next one slows or stops a little farther back. This wave, whose speed usually seems to register at about 12 miles per hour, could theoretically go on for as long as there was a string of sufficiently dense traffic. Even a single car on a two-lane highway, by simply changing its speed with little rhyme or reason (as people so often seem to do, in what I like to call “speed-attention-deficit disorder”), can itself pump these waves back down a stream of following vehicles. Furthermore, even if that car’s average speed is fairly high, the fluctuations wreak progressive havoc. This was the secret behind the Holland Tunnel experiment: With cars limited to “platoons” of forty-four vehicles each, the shock waves that were triggered were confined to each group. The platoons were like croquet balls spaced apart.

Many times we find ourselves stuck in traffic that seems to have no visible cause. Or we make it through a jam and begin to speed up, seeming to make progress, only to quickly drive into another jam. “Phantom jams,” these have been called, to the annoyance of some. “Phantom jams are in reality nonexistent,” thunders Michael Schreckenberg, a German physics professor at the University of Duisburg-Essen so noted for his traffic studies that he has acquired the epithet “jam professor” in the German media. There is always a reason for a jam, he says, even if it is not apparent. What seems to be a local disturbance might just be a wave pumped up from downstream in what is in reality a big, wide moving jam. It is wrong, says Schreckenberg, to simply call the whole thing stop-and-go traffic: “Stop-and-go is the dynamic within a jam.”

We fall for the phantom-jam illusion because traffic happens in both time and space. You may be driving into a space where a jam has been. Or you may not be driving into a jam—instead, the jam might be driving into you. “In my bucket analogy,” says Coifman, “the driver would be a water molecule. If the water level’s rising, then the jam’s coming to us.” We are also driving into history—or, perhaps more accurately, we are being driven back into history. By the time we actually arrive where something triggered the shock wave, in all likelihood the event will be only a memory. It may have been an accident, now cleared. “The queue’s going to persist for a while as it’s dissipating,” says Coifman. “It’s that water sitting in the bucket. In this case you’ve enlarged the hole in the bucket, but it does not disappear instantaneously.”

Or the hiccup in heavy traffic that passes through you might be the echo of someone who, forward in space and backward in time, did something as simple as change lanes. The car that changes lanes moves, eating up capacity in the new lane and causing the driver behind to slow; it also frees up capacity in the lane it has left, which triggers a bit of acceleration in that lane. These actions ripple backward in a kind of seesaw effect. This is why, if you pick one car in the neighboring lane as your benchmark, you will often find yourself passing that car and being passed by that car continuously. This is equilibrium asserting itself, the accordion of traffic flow stretching and compressing, the lingering chain reaction of everyone who thought they could get a better deal.

Since it takes so long for traffic to resume flowing freely once it has plunged past the critical density, it would seem the best way to avoid the ill effects of a jam would be not to drive into it, or let it drive into you, in the first place. This is the thought that occurred one afternoon a few years ago to Bill Beatty, a self-described “amateur traffic physicist” who works in the physics laboratory at the University of Washington. Beatty was on State Highway 202, returning from a state fair. The road, a “little four-lane,” was thronged with traffic from the fair. The traffic was “completely periodic,” as he describes it. “You’d drive real fast and then almost get to sixty and then you’d slow down and come to a stop, for almost two minutes,” he says.

So Beatty decided to try an experiment: He would drive only 35 miles per hour. Rather than let the waves drive into him, he would “eat the waves,” or subdue the wildly varying oscillations of stop-and-go traffic. Instead of tailgating and constantly braking, he would try to drive at a uniform speed, leaving a large gap between himself and the car ahead. When he looked in his rearview mirror, he saw a revelation in the pattern of headlights: Those behind him looked to be in a regular pattern, while the other lane had clusters of clumped stop-and-go vehicles. He had “damped” the wave, leveled off the extremes. “It cuts off the mountains and puts them in the valley,” he says of his technique. “So instead of getting to drive at sixty miles per hour briefly, you’re forced to drive at thirty-five miles per hour. But you don’t have to stop, either.”

Without analyzing the total traffic flow of the highway, it would be hard to know for sure what good Beatty’s experiment did. People may have just merged in front of him, pushing him back (if he wanted to keep the same following distance), while those behind him who thought he was going too slow may have jumped into the next lane, causing additional disturbance. But even if Beatty’s technique did little more than take a tightly congested traffic jam and stretch it backward, so that a car spent the same amount of time traveling a section of road, it would still save fuel and reduce the risk of rear-end accidents—two added benefits for the same price. Only how do you get everyone to cooperate? How do you prevent people, as so often seems to happen, from simply consuming the space you have left open? How, in essence, can we simulate ant-trail behavior on the highway?

One way is the “variable speed limit” system now being used on any number of roads, from England’s M25 “controlled motorway” to sections of the German autobahn to the Western Ring Road in Melbourne, Australia. These systems link loop detectors in the road to changeable speed-limit signs. When the system notices that traffic has slowed, it sends an alert upstream. The approaching drivers are given a mandatory speed limit (enforced by license-plate cameras) that should, in theory, lessen the effects of a shock wave. Even though many drivers suspected it was the lowering of speeds to 40 kilometers per hour that was causing the congestion, a study of the M25 found that drivers spent less time in stop-and-go traffic, which not only helped lower the crash rate by 20 percent (itself good for traffic flow) but cut vehicle emissions by nearly 10 percent. As drivers adjusted to the system, their trip times declined. Again, slower can be faster.

Smart highways also require smart drivers. The sad truth is that the way we drive is responsible for a good part of our traffic problems. We accelerate too slowly or brake too quickly, or the opposite; since we do not leave enough space between vehicles, the effects are often magnified as they move back up the line. Traffic is what is known as a nonlinear system, meaning most simply a system whose output cannot be reliably predicted from its input. When the first car in a long platoon comes to a stop, one cannot exactly predict how quickly or how far back each car behind it will stop (if they come to a stop at all). And the farther back, the harder it is to predict.

A driver’s overreaction (or underreaction) may amplify a shock wave that snaps, like the crack of a whip, several cars back, helping to cause a collision in the space that the originating driver has since left. One study examined a crash on a Minneapolis highway involving a platoon of seven vehicles that had been forced to come to a sudden stop. The seventh car in the group crashed into the sixth. Since we normally assume that cars keeping an adequate following distance should be able to stop in all conditions, that should be the end of it.

But the researchers, examining the braking trajectories of the vehicles in the platoon, found that the third car arguably bore a considerable responsibility for the crash. How so? Because the third car was overly slow to react, it “consumed” a larger portion of the “shared resource” of braking distance allocated among the cars. This left the cars farther down the line with progressively less time and space in which to stop—to the point where the seventh car, even though it reacted faster than the third, was following too closely to the sixth car to stop under the amplified conditions. Had the third car’s reaction been faster, the crash might have been prevented. For these sorts of reasons, the researchers pointed out, people who tailgate—that is, do not follow at the “socially optimal” distance—increase their risk not only of striking the vehicle they’re following but of being struck by the car following them.

What if drivers’ reaction times could be predicted with mathematical precision? The ultimate answer may be to combine smart highways with smart cars. It’s probably no accident that whenever one hears of a smart technology, it refers to something that has been taken out of human control. L. Craig Davis, a retired physicist who worked for many years in the research laboratories of the Ford Motor Company, is one of a number of people who have run simulations showing how equipping cars with adaptive cruise control (ACC), already found on many high-end models, can improve traffic flow by keeping the distance between cars at varying speeds mathematically perfect. This would not kill traffic waves entirely, says Davis. Even if a line of stopped cars could be coordinated to begin accelerating at the same time, he says, “if you wanted to get them up to speed with a normal distance between them at sixty miles per hour, you would still have this wave effect.”

Remarkably, the simulations show that if just one in ten drivers had ACC, a jam could be made much less worse; with as few as two in ten drivers, the jam could be avoided altogether. In one experiment, Davis located the precise moment the jam was avoided, just as one additional manual car was given ACC. This putative straw that broke the camel’s back brings to mind the example of the locusts. When the locusts reached critical density—one more locust—they began to behave entirely differently.

Just one problem has arisen in Davis’s simulations. Since the simulated vehicles with ACC like to keep very tight gaps between themselves, it may be difficult for a non-ACC car entering from an on-ramp to find a safe space between them. Also, like human drivers, ACC cars may not feel obliged to yield to entering drivers. These problems can surely be solved scientifically, but in the meantime, as we suffer the effects of our failure to always act cooperatively on the highway, we can draw one comforting lesson: Even machines sometimes have trouble merging.