Just after nine A.M. on September 24, 2015, during the annual Muslim pilgrimage known as the hajj, an accident occurred near the holy city of Mecca, in Saudi Arabia, that stands as the deadliest in the long history of hajj disasters. The numbers are disputed, but by reasonable estimate more than 2,400 pedestrians were trampled and crushed to death in a period of about 10 minutes. The event was widely reported as a stampede, a term that evokes visions of panicked herds and zealots, but the opposite was actually the case. There was indeed a giant herd, but the zealots within it could not escape, let alone run, and the panic that broke out was the result and not the cause of the carnage.
The hajj consists of a circuit of tightly scripted rituals at Mecca’s Grand Mosque and four other locations several miles away. It takes place across five consecutive days in the 12th month of the Islamic lunar calendar and is mandatory at least once in a lifetime for all Muslims who are physically capable of making the trip and can support their families during their absence. Non-Muslims are forbidden to enter the holy cities of Mecca and Medina, and the penalties for violation may include death. September 24 was a Thursday, and three days into the ritual. Two million registered pilgrims had descended onto the scene, along with perhaps another 200,000 who had sneaked in. They wore simple white garments meant to symbolize equality in the eyes of God. The women covered their heads but left their faces exposed. The gathering was not the largest known. Nonetheless, more than two million people all trying to do the same thing in the same place on the same day makes for a dangerously large crowd.
On this Thursday the action was not in Mecca but in the narrow Mina valley, three miles to the east. Mina is the site of the Jamarat, three immense pillars set into in a four-level pedestrian bridge, where pilgrims stone the pillars with pebbles in symbolic rejection of the Devil. Mina is also home to a tightly packed grid of more than 100,000 air-conditioned, fire-resistant fiberglass tents, where most pilgrims spend the nights. It contains hundreds of pedestrian alleys, many larger side streets that all look alike, and several major pedestrian arteries that lead in parallel to and from the Jamarat Bridge. On the morning in question, the temperature was approximately 110 degrees. The pilgrims had arrived around dawn after a mandatory overnight stay in the open desert and had been dispersed to their quarters to await their slotted departure times for the stoning ritual. They came from more than 180 countries, spoke dozens of mutually incomprehensible languages, and as a general matter had little experience with following rules. Consider, for example, that 62,000 Egyptians were among them, including no doubt a fair representation of cabdrivers from Cairo, who are famously unruly.
By 8:45 A.M., just before the tragedy, hundreds of thousands of pilgrims were on the move, streaming through the alleys, joining into larger flows on the side streets, and emptying into the main channels inbound toward the Jamarat Bridge. Those channels by then were dense with pilgrims. At the same time, a heavy return flow of pilgrims who had already completed the ritual was moving through separate channels in the opposite direction, outbound to the tents in Mina. By design, those two flows, the inbound and the outbound, were never meant to mix. The heaviest inbound flow was down a channel called Street 204, which was flanked by high steel fences. The movement there was slow but inexorable, regulated by the pace of the oldest and most infirm, and forced forward from behind by miles of advancing foot traffic. Toward the front the crowd compressed until people were walking nearly chest-to-back—a density that is inherently dangerous.
Why this occurred remains a question. Security forces are stationed at key points to regulate the flow. After the accident it was claimed—primarily by hostile Iran—that the severe crowding was due to a blockage caused by the movement of a Saudi prince or some other V.I.P. The attraction of this claim is that it provides a simple explanation and lays the blame squarely on the haughtiness of Saudi Arabia’s elites. The drawback is that it is probably not true. In any case, by nine A.M. the situation on Street 204 was critical: the crowd pressures were so great that people had lost all physical autonomy and were being propelled forward by unstoppable forces. There was no panic, but many of the pilgrims were growing anxious, and for good reason. In such conditions the slightest hiccup—someone tripping, someone fainting—can have catastrophic consequences.
What happened next in Mina was more than a hiccup. Eight hundred yards from the entrance to the bridge, a short side road made a right-angle connection with Street 204. The side road is called Street 223. It was supposed to be empty, but just after nine A.M. a large throng of disoriented pilgrims came down it, undeterred by police. The crowd was propelled from behind into the thick of the moving people on Street 204. The identity of the new arrivals remains in question. They may have been pilgrims headed for the bridge who had taken a parallel route, Street 206, that emptied onto the side road, Street 223, that in turn emptied into the crowd on the main route, Street 204. On the other hand, some evidence suggests that they were people returning from the ceremony who somehow had gotten confused and had split off from the outbound flow. Either way, their sudden arrival on Street 204 represented a major failure of the Saudi authorities—the self-professed guardians of the hajj.
The effect was to jam the flow on the main street, stopping any further movement toward the bridge and causing pressures to build rapidly as the trailing crowds continued to move forward with no awareness of what was happening ahead. No video recordings have surfaced publicly, and the memories of survivors are limited by confusion and trauma, but what is certain is that, for those in the middle of the intersection, escape was not possible. The pressures grew so great that some pilgrims were lifted out of their sandals, and many had their clothing torn off. Those caught with their hands at their sides could not raise them to protect their chests for breathing. The shouting and screaming began. Within a few short minutes the first victims died, some of them while standing. Compression asphyxia was the cause: the pressure on their chests may have exceeded 1,000 pounds. That same pressure was pushing people against the steel fences, which unfortunately did not give way. Some young men were able to free themselves and climb over, or to pass children across to safety, but most people lacked the strength, and survived or died in a condition of helplessness.
It got worse: a chain reaction began when one or several pilgrims fell down. This created a void into which the crowd pressures pushed the immediate neighbors, in turn expanding the void, turning a small crowd collapse into a massive one that progressed upstream on both streets, and in places stacked the victims 10-high. The primary cause of death was approximately the same—asphyxiation due to the sheer weight of bodies, though skulls were crushed as well and lungs were pierced by broken ribs. Some witnesses later reported seeing torsos that had been torn apart. The collapse ended relatively quickly on the side street but progressed for minutes up the main artery, Street 204. It ended only after urgent calls brought the upstream flow to a halt. Tangled among the dead were more than a thousand injured, many of them moaning or calling out for help or water. The heat was intense. Emergency crews began to move in quickly but found access difficult because of the crowds, and were overwhelmed by the scale of the carnage they came upon. It took 10 hours for the evacuation to be accomplished. Much effort was wasted on the removal of the dead even as the injured lay mostly unattended and continued to die.
The street was closed for another day, but the hajj proceeded as ordained, and even pilgrims who had barely escaped with their lives went on to stone the Devil after all. True to form, the Saudi government announced that 769 people had died—an undercount that it has stuck to ever since, but that was soon given the lie by all the people from 42 countries who weeks later were still missing because the bodies were never identified and, given the dictates of Islamic writ, were buried quickly. Saudi Arabia’s great Shiite rival, Iran, was the worst-hit. It lost 464 pilgrims. Mali lost 312; Nigeria, 274; Egypt, 190; Bangladesh, 137; Indonesia, 129; and the list goes on. What had just occurred was the most lethal crowd crush in history. It did not escape the world’s attention that the second-worst had also been during the hajj—1,426 dead in 1990—and that a string of other mass fatalities had taken place during the stoning of the Devil. The Saudis take great pride in hosting the hajj, and they felt embarrassed—even threatened, as they tend to feel under even the best of circumstances. They have vast wealth but little else, and live amid religious and geopolitical forces that one day will likely tear the kingdom apart. In the meantime they act with the arrogance of people in control. The government responded with typical obfuscation, promising a thorough and open investigation—meaning a cover-up—and blaming the tragedy on the pilgrims for not having followed instructions. The man in charge of the hajj was the crown prince and minister of the interior, Mohammed bin Nayef. The day after the accident, Saudi Arabia’s highest religious authority, the grand mufti, Abdul Aziz bin Abdullah al-Sheikh, helpfully assured him that he was not to blame, and attributed the deaths to God’s will.
Such reactions frustrate G. Keith Still, a professor of crowd science at Manchester Metropolitan University, in Manchester, England, and arguably the pre-eminent expert in the field. Still is an affable Scotsman with a love for performing magic tricks, riding his Harley-Davidson, and playing the jazz saxophone. He has a Ph.D. in mathematics and came to crowd science through his knowledge of complex modeling and computer simulation. He has since grown wary of such tools because of the need they impose to make assumptions that may be false, and the difficulty of predicting human behavior. He now advocates only narrow uses of simulation at certain stages of planning, and a broader, more practical approach to accommodating large crowds. He said, “I came to realize that the people who make the life-and-death decisions—no disrespect—but they are soldiers and police, or ex-soldiers and police, and they don’t come through academia. That’s putting it politely.” On the other hand, he said, “computer scientists are the worst guys to try and talk to, because they have a God-like ability to play with dots on a screen as if they are their children. But I’ve never, ever seen a crowd behave the same as a simulation.” More than a decade ago he spent several years shuttling to Riyadh to help the Saudis improve safety during the hajj, and particularly to reduce the recurrence of crowd crushes on the Jamarat Bridge. He said, “I had to try to get into the mind-set of the pilgrims. The people I was working with said I was four-fifths Muslim, because I could never get past the alcohol bit. Being from Scotland, you see.” In other ways, too, it was an unsatisfying experience. He went on: “Yeah, the ‘will of God’ the pre-destination argument, kept coming out. To which I replied, God didn’t build this system. I don’t remember him at any of the bloody project meetings. We built it! You need to understand the dynamics of the risks!’ ” Then he said, “Needless to say . . .”
Needless to say, the Saudis were not impressed by his views. At one point, he says, they confiscated his passport and held him in a ministry building. Meanwhile, they were cutting off dissidents’ heads.
But so what? There is plenty of business for Keith Still in the world. Dense crowds assemble in nearly every country. Over the past 20 years alone, death by crowd crushing has occurred in Afghanistan, Angola, Austria, Bangladesh, Belarus, Benin, Brazil, Bulgaria, Burkina Faso, Cambodia, China, Congo (Brazzaville), Congo (D.R.C.), Denmark, Egypt, England, Germany, Ghana, Guatemala, Haiti, Honduras, Hungary, India, Iran, Iraq, Ivory Coast, Japan, Kenya, Liberia, Libya, Malawi, Mali, Mexico, Morocco, Nigeria, North Korea, Pakistan, the Philippines, Portugal, Saudi Arabia, Scotland, Senegal, Slovenia, South Africa, Spain, Tanzania, Togo, the United States, Yemen, Zambia, and Zimbabwe. In those crushes, more than 7,943 people have died.
The venues and activities that create dangerous crowds are well known: big rock concerts, big sporting events, popular nightclubs, mass pilgrimages, and the funerals of demagogues. In that last category, John J. Fruin, a former Port Authority of New York and New Jersey research engineer and the father of modern crowd science, has written that in 1953, when a crowd of three million gathered in Moscow for Joseph Stalin’s funeral, hundreds and possibly thousands were crushed to death by forces sufficient to lift horses off their feet (and crush the horses as well). The Soviets suppressed the news. A more recent case occurred in 1989 at Hillsborough Stadium, in Sheffield, England, at the start of a semi-final championship soccer game between the Liverpool and Nottingham Forest football clubs. Because of grave errors by the local police, thousands of eager Liverpool fans were allowed to enter two stoutly fenced standing-room pens that were already fully packed with spectators. The resulting crush killed 96 people, with most of them dying upright on their feet. About 300 others were seriously injured. The crush was worsened by police on the field who misread people’s attempts to escape by climbing the fence, and initially struggled to keep them contained. Then came the insult. The police defended themselves by altering field reports, blaming the fans, and planting false stories in the press about their behavior. This was widely believed because of the existence of soccer hooliganism, but in Sheffield the accusations were false. Investigations gradually uncovered the truth, and in April of 2016 a coroner’s inquest issued a finding of fact that the victims had been unlawfully killed, that they had not contributed to their own deaths, and that gross negligence by the police was primarily to blame.
Two forms of crowd movement lead to crushes. The first form is known as a “craze,” when large groups of people move forward in the rational hope of attaining a benefit—food handouts, proximity to a band on a stage, discounts at a big-box store, or, for that matter, the completion of a ritual during the hajj. The second form is known as a “flight response,” when large groups move away from a perceived threat. The word “flight” evokes images of people running and fits neatly with the misnomer “stampede,” but the record shows that if there is any running it soon ends because of the crowding, and that people in such cases are generally calm before the crushing begins. The problem is crowd density. In the 1970s, Fruin calculated that the average pedestrian takes up about 1.5 square feet. At densities of 15 square feet per pedestrian, people can move freely. At 10 square feet, according to Fruin, “excuse me” becomes necessary. At 2.75 square feet, involuntary contact with others begins, but there is still little risk of a crush. In a crowded elevator where there is contact all around and movement is impossible, the space is reduced to 1.6 to 1.8 square feet per person. Those are the densities where, on a larger scale, crowd crushes happen.
Keith Still has taken that work and expanded upon it through computer simulation and experiments with volunteers. He uses a measure of people per square meter—nearly the same as a square yard—and differentiates the requirements for a crowd that is moving and one that is not. At two people per square meter, even a moving crowd is fine. Add two more and movement becomes awkward. Add another, resulting in five people per square meter, and you begin to flirt with disaster. At six people per square meter, no space is left between individuals, and people are hemmed in and unable to control their movements, whether to stop or go. No one would willingly enter into such a mob, but crowds of the unwilling are compacted by the progression of the masses behind them and by physical constraints such as walls, fences, gates, doorways, stairways, up ramps, and slight turns or changes in direction. As the crowd in a given space exceeds 80 percent of the space’s capacity, the compression accelerates. In the real world, densities of seven, eight, or nine people per square meter are not uncommon.
Even at that extreme, people are not yet dying, but beyond five people per square meter the crowd has effectively formed into a single mass through which energy can be transmitted. It is more like a liquid than an assembly of solids, and the laws of fluid dynamics start to apply. Someone shoves, someone stumbles, and the effect is amplified by others. The impulses move through the crowd and rebound with increasing intensity. They are a prelude to death. From within the crowd they appear as sudden mass movements, impossible to resist, 10 feet in some direction, 10 feet in another. People caught up in them are in serious trouble. They need to leave, but cannot. They need to raise their hands into a boxing position to protect their chests, and turn 90 degrees to the flows, because from side to side the rib cage is less compressible than it is from front to back. If they are strong and lucky, they may succeed in this, though not in the highest-density crowds. Above all, they need to stay on their feet, although if a progressive crowd collapse occurs, this will be impossible to do. Then it’s a question of luck—whether they end up at the top of a pile or the bottom.
Shock waves are implicated in most crowd crushes, but not all. For instance, large crowds moving down stairways have repeatedly suffered mass casualties because someone tripped: 354 dead in 1942 on the stairs leading to an air-raid shelter in Genoa, Italy; 173 dead in 1943 on the stairs leading to another air-raid shelter, in the London Underground station at Bethnal Green; 21 dead and more than 50 injured in 2003, during an urgent exit from a second-floor nightclub in Chicago. Shock waves are a more insidious matter. They capture people long after the possibility of avoidance has vanished. Shock waves certainly accounted for the soccer deaths in Sheffield. They also accounted for the deadliest day of the war in Iraq—August 31, 2005—when a million Shiite pilgrims gathered at a Baghdad shrine and rumor spread of an impending suicide attack. The crowd did not respond to the rumor by panicking, as was widely reported, but quite reasonably began to leave the area. Thousands tried a bridge over the Tigris River, only to find that on the far side the exit from the bridge was heavily gated. In the crush that developed as people continued to cross, the shock waves grew so powerful that the guardrails gave way, dropping hundreds into the river. The fall to the river amounted to a lucky escape, but only for those who could swim. In all, 965 people died, most on the bridge, and by compression asphyxiation.
Admittedly, that was in the hell of Iraq during a chaotic time. But the problems exist even in the most orderly societies. In Duisburg, Germany, for instance, 21 people died and more than 500 were injured in 2010 at the entrance to a music festival called the Love Parade. A huge crowd was trapped in a sheer-walled concrete channel that the event’s organizers—who were worried about gate-crashers—had stupidly designated as the way in. The police were almost as incompetent. Their attempt to control the crowd added to the pressures. Fruin was the first to make the point that police are often poorly prepared to handle such masses of people, because their emphasis is on maintaining public order, and it is crowd management, not officious control, that is needed. In this case proper management would have entailed metering the pedestrian flow far upstream of the potential choke points; instead the police waded into the thick of things and tried to set up blockades. Inevitably they were overwhelmed. Videos exist on YouTube that show the shock waves developing and capture the screams of the victims. The point is that these were neither zealots following the dictates of an ancient prophet, nor even die-hard soccer fans. They were fresh-faced Germans who just wanted to celebrate life. But the density of the crowd condemned them.
The obvious solution is to avoid big crowds. When it comes to the hajj, however, Muslims do not have a choice. This places the rulers of Saudi Arabia in a typically Saudi-style bind—one that is largely of their own making, and impossible to undo. The Saudis are conservative Wahhabis, true believers, and they take their hajj responsibilities seriously, for both religious and geopolitical reasons. Their problem goes back to the Prophet Muhammad, who was not only a big-picture man but also a micro-manager who issued edicts on all manner of subjects: how to go about one’s day; how to dress; how and what to eat; how to have sex; how to wash; when to pray. His words on any subject became law, subject to relatively little interpretation over the centuries because he was the final prophet.
The issues here concern the creation of a hajj and the requirement that all able-bodied Muslims perform a pilgrimage to Mecca at least once in their lifetime if they can afford it. At first it was a unifying idea that anticipated the vast geographic expansion of Islam. Then pick a date—say, a thousand years ago. Muslims were numerous in large parts of the world, but few of them could afford the long and arduous trip, and most were therefore let off the hook. Crowd crushes were not a problem. By 1926, when the House of Saud gained possession of Mecca and the kingdom of Saudi Arabia was effectively born, pilgrims on the hajj still numbered only about 100,000 a year—a volume that was easily accommodated by Mecca’s 16th-century Grand Mosque, and by the open land of the Mina valley and beyond. No changes were made until 1955, when the first Saudi expansion of the mosque was begun. The country’s founder, His Majesty King Saud, had 38 wives and concubines and more than 100 children. He initiated the expansion later in life. The purpose was largely to consolidate his family’s prestige and power. Saudi Arabia was strapped for cash at the time—its oil wealth lay in the future. The head of the Saudi Binladin Group—a friend of the king, and the father of Osama bin Laden—advanced the necessary funds in return for exclusive development rights in and around Mecca. The expansion continued for the next 18 years. It destroyed much of historic value and replaced it with poorly conceived designs, many of which in turn were soon torn down. A willingness to destroy ancient structures is as fundamental to the Saudis as it has been to ISIS and is rooted in an aversion to any hint of idol worship—the sort of reverence that turns objects into shrines. In any case, by the time it was finished, in 1973, the expansion allowed the mosque to accommodate 500,000 pilgrims at a time. For a brief period, that seemed enough.
But globalization was coming. It first touched Mecca with a mass killing that had nothing to do with crowd crushes. In November 1979 a group of at least 500 rebels demanding a return to a purer Islam and an end to Westernization invaded the Grand Mosque, took thousands of hostages, and proceeded to hold off Saudi forces for more than two weeks, at the cost of at least 255 dead. The siege was finally broken with help from French commandos who hastily converted to Islam in order to enter the city. Sixty-eight of the rebels were captured, sentenced to death, and publicly beheaded in a stern display of the king’s displeasure. Nonetheless, apparently because he believed that the attack was God’s punishment for a society grown lax, the king then moved in the direction the rebels had demanded: shuttering movie theaters and music stores, banning public images of women, enforcing stricter separation of the sexes, increasing religious studies in schools, and eliminating classes on world history.
The kingdom found itself yearning to modernize and at the same time hurtling backward in time. The dichotomy was nowhere more visible than in Mecca, a sacred city where nonbelievers had never been allowed, and would not be now, even though the technical expertise necessary to build it up resided primarily among the atheists, Christians, and Jews of Europe and the United States. The pressures reached a peak every year during the five days of the hajj. In the 1980s, with a fast-growing Muslim population worldwide, and inexpensive air travel suddenly a reality, the number of Muslims who could afford to fulfill the obligation soared, and for the first time the crowds in Mecca surpassed one million. It became obvious that Mecca’s capacities would never meet the demands. But rather than thinking the problem through, the Saudi king, whose name was Fahd, began a second expansion plan, and then doubled down in 1986 by expanding his formal title from “His Majesty” to include “Custodian of the Two Holy Mosques.” Fahd was the second-richest man in the world. He had a 482-foot yacht and a private Boeing 747, both equipped with medical facilities and doctors. He also had a problem with the hajj, but apparently did not understand it. His change of title demonstrated that there is no cure for stupidity. This is a basic fact of life in Saudi Arabia. There are problems you can’t just buy yourself out of.
The first crush occurred the following year, in 1987. It was not a craze, but a flight response. A large group of Iranian pilgrims were demonstrating against the United States and Israel, as they had routinely done in previous years. Much as they hated Iranians, and were supporting Saddam Hussein in his war against them, the Saudis had generally let such demonstrations pass because the protests were not directed against the Saudis themselves. This time, however, Saudi security forces blocked the path, the demonstration grew violent, and gunfire erupted. As the protesters fled, some were shot and killed, and others were crushed. More than 400 people died, including 275 Iranians. Afterward, Iran boycotted the hajj for three years, and Saudi Arabia instituted a quota system, still in effect, that tried to limit the crowds by allotting one hajj visa for every thousand Muslims by country. This created long waiting lists and resentment, raised religious concerns, spawned corruption in countries such as Indonesia and Pakistan, and provided an excuse for hundreds of thousands of worshippers to ignore official permission and sneak in uncounted and uncontrolled.
By the late 1980s a second expansion was under way. It was primarily focused on enlarging the Grand Mosque to achieve the present capacity of nearly a million pilgrims at a time, but it also involved infrastructure improvements elsewhere along the routes of the hajj, and especially in Mina, where canvas tents were organized into a tightly packed grid. As usual the improvements were designed by distant consultants who were not allowed on the actual site. The construction was done by the Saudi Binladin Group. One of the improvements was a 600-yard air-conditioned pedestrian tunnel that passed through a small mountain between Mecca and the Mina valley. Spanning its exit was an overhead pedestrian bridge. In 1990, on the final day of the hajj, disaster struck when crowd pressures on the overhead bridge caused a railing to collapse and dropped seven pilgrims into the throng below, blocking the tunnel exit, and causing the tunnel to fill beyond its capacity. In the crowd collapse that ensued, 1,426 pilgrims died. Nearly half were Indonesian. The Custodian of the Two Holy Mosques, His Majesty King Fahd, said, “It was God’s will, which is above everything.” He also blamed the dead for not following the rules, and added, “God willing, we will see no tragedies in the coming years.”
God was unwilling. In 1994, a crowd crush killed at least 270 pilgrims during the stoning of the Devil at the Jamarat pillars, in Mina. Since the 1950s, each pillar had been surrounded by a low concrete wall, creating basins into which the thrown pebbles fell for later removal. In the 1960s a simple one-story bridge had been built around them, allowing the slowly moving crowds to fire off from either ground level or the bridge above. That design had increased the throughput of the site to about 100,000 people per hour, but by now the numbers arriving were nearly double that. The deaths there had been predicted by outside consultants, and ignored. The Jamarat had become a bottleneck.
In 1997 a fire broke out in Mina, incinerating 70,000 tents. More than 300 people died, most by crushing as huge crowds fled the flames. Typically, the Saudis did not address the core issues of density and overcrowding, instead turning to a narrow, off-the-rack solution and rebuilding Mina as tightly as before, only with fire-resistant fiberglass tents. That fixed the fire part, but nothing else. The nearby Jamarat Bridge continued to stand out as a problem. In 1998, 118 pilgrims were crushed to death there. In 2001, the toll was 35. In 2003, it was 14. The next year, it was 251. The Saudis repeatedly blamed the dead, but every mass fatality was an embarrassment that called the king’s stewardship into question. The hell of it was that, in 2001, they had already decided to build a larger Jamarat Bridge. The design and construction phases took six years and led to the bridge that stands today—a structure that can be traversed on one of five stacked levels, with multiple entry and exit routes, helipads, a control tower, and new pillars five floors high. A conveyor belt at the bottom of the pillars whisks away the pebbles (about 50 million of them a day) to waiting dump trucks for re-use on the next hajj. The new bridge is capable of handling 400,000 pilgrims an hour and, with additional levels soon to be added, is meant to handle twice as many in the future.
Why, then, is there a sense that little has been solved? Keith Still has opinions on the matter. He was first engaged in the project (remotely—from Riyadh) at the start, in 2001, when he was brought in to run computer simulations of crowd flows. He recommended modifications to certain parts of the new bridge and also determined the optimal dimensions and characteristics of the three new pillars, which were to be elliptically shaped to streamline the flow, and made of a special composite material to absorb energy and cause the pebbles to drop rather than bounce back into the crowds. Still was pleased with the work, but largely unimpressed by the Saudis. Over time he grew frustrated by the narrowness of their approach. He made the obvious point that the hajj is a tightly coupled system that has to be addressed as an inter-related whole, and that changes to any of its components will reverberate throughout, possibly with deadly consequences.
The Saudis didn’t want to be bothered. They kept concentrating on the Jamarat Bridge, and therefore so did he. It was to be pre-fabricated off-site, and made of sections that could quickly be assembled and installed. As usual, the Saudi Binladin Group had the contract. The first concrete was poured in 2004, with two hajjs still to go before the installation. After the huge crush that occurred that year, the question was how to prevent any further disasters until the new bridge could be put into use. The Saudis turned to Still and several others to come up with a plan. They installed three temporary elliptical pillars and took measures to regulate the inflow. This worked well enough in 2005, when no one was killed. That summer Still wrote a report that predicted a potential crush at a certain narrow entrance to the bridge, and expressed the danger in blunt terms. The Saudis rejected it. A group of German consultants had arrived and gained the upper hand with impressive computer simulations which predicted that flows onto the bridge could be handled with an electric sign—a “verbal messaging system”—to signal Stop or Go. Still insisted that this would not work, particularly for a crowd in which more than a hundred languages are spoken and many people are illiterate, or are old and have lost their vision. He was overruled. The Saudis did away with the previous measures and hung the electric sign directly over the entrance, where soldiers would establish a crowd-control line. The problem was that neither the soldiers nor the front ranks of pilgrims could see the sign when it was directly overhead. Still tried to get the sign repositioned 50 yards deeper into the bridge, where at least the front ranks could see it. Again he was overruled. He left the country. Then, for the 2006 hajj, 2.5 million pilgrims went to Mecca, and on the morning of the third day, when the sign said “Stop,” the soldiers, sagging backward, managed to halt a crowd at the entrance to the bridge. When the sign then said “Go,” neither the soldiers nor the front ranks saw it, but thousands of pilgrims farther back understood and began to move forward. Nearly 350 people died.
Still was called back to Saudi Arabia for the investigation. It lasted two days and came to the usual conclusion: the collapse was the fault of the dead and was God’s will. Still left Saudi Arabia and has not returned. No sooner had the hajj of 2006 ended than the Saudi Binladin Group demolished the old Jamarat Bridge and began installing the new one. By now, Saudi Arabia was swarming with foreign consultants who were supplying expensive equipment and advice but were still unable to enter Mecca. The Saudis were proud. The annual number of hajj visitors now exceeded three million. All this was happening as Mecca, by royal decree, was being transformed into a gaudy Las Vegas-style religious-tourism city, with numerous shopping malls and luxury hotels, chain stores, souvenir and fast-food outlets, and clusters of skyscrapers, including the world’s third-tallest building, the widely reviled Makkah Royal Clock Tower—an absurdity modeled on London’s Big Ben that rises to 1,972 feet across the street from the Grand Mosque. The reason for these developments was not to accommodate pilgrims on the hajj but to profit from the much larger number of ordinary visitors who come to Mecca year-round for a lesser pilgrimage known as the umrah. Those pilgrims, who confine their rituals to the mosque, will soon amount to 15 million a year.
The problem for the Saudis is that performing an umrah does not lessen the responsibility for performing the hajj. By 2012, the historic peak of hajj attendance, six years had passed since the last fatal crowd crush, the revamped Jamarat Bridge was proving its worth, and a new, high-capacity rail system had been installed to cover the 11 miles between Mina and Mount Arafat, the most distant point on the hajj circuit. The Custodian of the Two Holy Mosques, now a king named Abdullah, launched a major new expansion of the Grand Mosque meant to accommodate five million pilgrims by the hajj of 2020. The planning had been done under cloak of secrecy and at great expense by some of the largest engineering and architectural firms in the West. It had involved extensive crowd simulations and much thought about practical matters such as air-conditioning, shade, drinking water, food, garbage, and sanitation. No detail had been overlooked. The placement and orientation of toilets had provoked long theological debates, but had finally been settled. But now all that was done, the Saudi Binladen Group had the contract, and work soon got under way.
The project was not limited to the mosque. It included expanding the crowd capacities at every stage of the circuit except for one—the tent city of Mina and the routes to and from the Jamarat Bridge. This was a glaring omission, but the Saudis had placed surveillance cameras throughout the valley, linked them to optical counting software in a control room, and invested in an impressively complex scheduling plan backed by simulation and designed by the German consultants. The scheduling was described in a recent paper co-authored by one of the consultants, a professor of computational social science named Dirk Helbing, who took pains to say that others, and not he, were responsible for the planning in 2015. Helbing believes in simulation to the extent that in 2011 he applied (unsuccessfully) for a one-billion-euro grant from the European Commission to build a simulation of the entire world. His paper on his efforts in Mina is an unabashedly German artifact—an impressive description of the use of math and simulation to schedule optimal departure times (to the nearest minute) from the tents, generally to coincide with trains running perfectly on time. It ignores the reality that many of the pilgrims are illiterate, disoriented, or decrepit, and that almost none of them come from countries where people stand in orderly lines. It certainly did not help that he had never been to Mecca.
Still said, “Simulation? Little dots on a screen are just one method of testing a set of assumptions. If I change the weather conditions, are your assumptions still true? If there is suddenly a loud noise or a bad smell, are your assumptions still true? You have to understand the limits to mathematical models. You can’t really reduce an individual’s mind-set to an algorithm.” He went on: “The Saudis are always looking for a technological solution—you know, read the meter, pull the lever, make it work.” And in the meantime they are keeping their mouths shut. Recently I wrote to the Saudi Embassy in Washington, D.C., and directly to a ministry in Riyadh, requesting information about the official investigation of the most recent disaster. I did not ask for conclusions, merely for a description of the investigation itself—who is conducting it, what methods are being used, and when a report might be issued. I received no answer.
The truth is we already know what we need to know. The crush of 2015 stands for all of Saudi Arabia, a country condemned to mutually destructive impulses—the urge to go forward, the desire to go backward; the urge to lead, the need to follow; the compulsion to suppress, the knowledge of where suppression will lead. Its arrogance, its insecurity, its dishonesty, its cowardice. Its pampered, fleshy weakness dressed up as purity and strength. Its fundamental dependence on people it despises. The country is at the mercy of forces beyond its control—whether it be the hajj or its position in the Middle East. I spoke to the leading crowd specialist in the United States, Paul Wertheimer, a man with a fine sense for realities. He said, “There are 1.6 billion Muslims in the world, and it’s the fastest-growing religion. All the Saudis know how to do is to make things bigger. But you can never build big enough. The hajj is much more than just a crowd-management problem. What’s needed is enlightenment. The thinking has to change.” But that is not the Wahhabi position, and the thinking may never really change. If there is a God, that must be God’s will.
