When It Rains, It Pours: The 2008 Allinges (France) Level Crossing Collision
Background
Allinges is a municipality of 4938 people (as of 2022) in eastern France, located in the department of Haute-Savoie just south of Lake Geneva, 53km/33mi north of Annecy and 26km/16mi northeast of Geneva in Switzerland (both measurements in linear distance).
The municipality lies on the Léaz — Saint-Gingolph rail line, a 90km/56mi electrified single-tracked rail line connecting several towns along the Swiss-French border. The line crosses the municipality of Allignes between the towns of Mésinges and Margencel.
The Vehicles Involved
TER 884328 was a regional passenger service from Evian les Bains to Geneva’s Eaux-Vives station, offered under the SNCF’s (French national railway) TER-brand, which describes regional services run by the regional councils (Transport Express Régional). On the day of the accident the service was performed by SNCF Z 23512, a two-car bilevel electric multiple unit (EMU) introduced in 1997. Each Class Z 23500 consists of a powered four-axle bilevel car permanently coupled to an unpowered counterpart measuring 52.5m/172ft in length at an empty weight of 72 metric tons. The two-car units can carry up to 210 seated passengers at as much as 140kph/87mph.
Approaching the crossing on its way to Margencel was a Bova Futura FHX12, a 58-seat bus chartered by a local middle school (grades 5–9). The Futura was a streamlined two-axle bus introduced by the Dutch manufacturer Bova in 1982, finding international success until being replaced by the Futura 2 in 2010. They could be had in different lengths and fitted with different engines and interior configurations depending on the customer’s wishes, with the one involved in the accident being a 1999-model measuring 11.99m/39ft in length, 3.58m/12ft high and 2.49m/8ft wide at an empty weight of 12.5 metric tons. The bus was carrying 55 passengers and a driver at the time of the accident.
The Accident
The Theodore-Monod middle school in Margencel chartered a bus on the 2nd of June 2008 for a field trip to historic castles in nearby Allinges. The trip was the idea of Mister Jandin, a history- and geography teacher at the school. The trip had been cancelled twice due to poor weather, but the teacher decided that, despite the day showing rain too, he would not disappoint the students by postponing the trip again. 50 students and 5 teachers thus spent half the day walking around Allinges medieval structures before boarding the waiting bus to take them back to the school just after 1:00pm. They had arrived after an easy 9km/5.6mi drive along two main roads, but roadworks force the bus driver to take a different route through the town of Mésinges.
Both routes had the bus cross the Léaz — Saint-Gingolph rail line, but the main road on the way out saw the bus simply drive over a bridge while it will now have to navigate crossing 68 on the edge of Mésinges. The path demanded of the bus can be described as somewhat challenging: The bus will come down a narrow residential street, then go downhill as it crosses the rail line before having to watch for oncoming traffic as it heads uphill into a sharp turn exceeding 90° to the right. That turn would require the bus to temporarily occupy the oncoming lane, thus the need to watch for oncoming traffic. The driver will also have to watch for traffic coming out of a side-street on the far side of the rail line throughout the maneuver, which might not expect a vehicle occupying both lanes.
The bus approaches the level crossing at around 1:50pm in pouring rain, with the driver slowing as he nears the rail line. He stops as the front of his bus scrapes over the tarmac, continuing slowly after re-checking that the barriers are up and the stoplights at the crossing are off. He’s halfway across the tracks when a student in the back tells him that the barriers are closing against the bus. The driver stops, startled, before one of the teachers tells him to hurry and get out of the crossing as a train is coming. It was never determined to certainty what happened at that moment, but the driver either stalls the engine or picks the wrong gear by accident, keeping the bus stopped in the crossing.
Z 23512 has left Èvian-les-Bains at 1:51pm and is headed along the rail line towards Geneva at 93kph/58mph when the driver comes around a long right hand turn and suddenly sees the white bus materialize out of the rain up ahead, right in his path. The driver triggers an emergency stop before starting a retreat towards the passenger area, but it’s way too late to stop. The train is just 125m/410ft from the crossing as the brakes apply, and the train slams into the rear left corner of the bus at 1:52pm, still travelling at over 80kph/50mph. The impact obliterates anything behind the bus’ rear axle and rips the roof almost entirely off the vehicle as it’s thrown out of the crossing, spinning around in the process. Hitting behind the axle left so little resistance that the train doesn’t even derail, coming to a stop a short distance past the crossing. 7 people are killed in the accident, 33 more are injured (3 of which suffering severe injuries), all aboard the bus*. The train driver and his passengers are unharmed.
*This is the number from the official report, other sources claim 25 injured survivors
Aftermath
The train driver returns to his cab before the train has fully come to a stop and reports the crash before heading into the passenger area to tend to his passengers. He places another call shortly after the accident, demanding that the overhead catenary be shut down so responders can approach the wreckage.
Investigators initially examine the crossing, at least the parts the bus didn’t mow down as it got punted by the train. As far as they can tell everything about it worked as intended, the stoplights illuminated and the half-width barriers closed as they were meant to. The crossing was even fitted with breakaway barriers which would be especially easy to break through if one finds themselves stuck between closing barriers.
The bus driver survives the accident with minor injuries and can soon be interviewed by both the police and the rail accident investigators. He states that he initially stopped because he saw a car coming the other way and let it go through the crossing first, with another car turning off the road before reaching the crossing. He briefly stopped when his bus scraped the ground at the front before continuing until a student shouted that the barriers were closing. He was startled by the claim but unsure if she was serious or trying to make a bad joke, which may have delayed his reaction. The instruction from the teacher convinced him that there was urgency, so he tried to accelerate but reportedly found the bus to barely move, not exceeding a walking pace for the next several seconds before a sudden jolt marked the impact by the train. According to the driver he may have tried to downshift and accidentally picked a wrong gear, hampering the bus’ acceleration. The crossing was on his routes about twice a month, so he knew it and the difficult line to take over it fairly well, but usually used a different model of bus for his drives as he neither liked nor trusted the Futura FHX12. Apparently the type lacked acceleration in general and was plagued with breakdowns. Investigators further learned that he always took the crossing at extremely low speed to avoid damaging the bus from bottoming out, especially worrying about a drain plug at the rear (probably referring to the engine’s oil pan).
The bus does have a height-adjustable suspension, but the driver explained that he didn’t set it to “high” ahead of the crossing as that would have taken some more time, possibly forcing the bus to wait at the crossing, and that the high setting rode so rough that he didn’t pick it for the entire drive back. The company who owned the bus supported their driver’s statement, saying he was experienced with over 20 years at the company and had been one of their most attentive drivers.
The investigators also talked to the surviving students, trying to get a better image of what happened on the bus ahead of the collision. Everyone interviewed said that the atmosphere was relatively calm, but that various smaller conversations created a certain noise-level aboard the bus. A few students also recalled condensation on the side-windows limiting visibility outside. The survivors didn’t agree on the operations of the crossing, with students seated in the same row disagreeing on whether the stoplights were on or off as the bus entered the crossing. One student also recalled the bell ringing as the bus pulled into the crossing while 21 survivors said it only came on once the bus was in the crossing. Lastly, there was wild disagreement on the motions of the bus, proving the unreliability of eyewitness accounts:
- 38 survivors stated the bus stopped at least once
- 18 said the bus jerked or even reversed
- 8 believed that the bus slid while 2 stated it didn’t
- 5 recalled the engine stalling, two rejected that
- 2 passengers recalled the driver moving the gearstick
The investigators chose to reenact the accident in late June, organizing a matching bus and train as well as having the rails and road sprayed down with water to somewhat recreate the grip-conditions on the day of the accident. The bus was loaded with ballast to recreate the load conditions of the passengers it carried ahead of the accident. The reenactment procedure was attended by relatives of the victims, the bus driver from the accident, judges, lawyers and collision experts, who watched from the sidelines as the investigators conducted several tests. First was a brake test for the train, which, despite lacking the resistance of the bus, saw the train come to a stop just 20m/66ft past where it did on the day of the accident. This proved that the train driver really had no chance to avoid the accident, but also showed how little the bus had slowed the train down. The bus was then driven through the crossing several times, both by the accident driver (which must’ve been incredibly difficult) and another driver hired by the investigators in order to best recreate the motions of the bus.
The investigation found that the westbound crossing, while less tricky than an eastbound crossing would’ve been, saw the bus almost entirely in the oncoming lane as it went over the rails. This proved that the barriers had been open as the bus was halfway through the crossing, as it would’ve otherwise broken the western barrier. The bus was subsequently moved further through the crossing, making some contact with the road surface at both the front and rear end. Investigators placed the barriers against the bus as it moved across the rails, showing that the bus could have slipped underneath them without breaking the barriers. They lined the right hand front wheel up with a rubber mark on the curb and brought the rear into the calculated position where it was as the train hit the bus, causing the eastern barrier to close completely behind the bus.
The reenactment provides evidence indicating that the bus driver was indeed allowed to drive into the crossing when he did, with his slow speed dictated by the shape and uneven ground of the crossing meaning he would’ve required about 10 seconds to clear the tracks. The stoplight may have come on just as he passed it, likely too late to catch a glimpse of it, while the noise-level in the bus along with the required focus for the difficult line to follow through the crossing distracted the driver from the bell that accompanies the stoplight.
It couldn’t be determined to certainty what exactly slowed his exit from the crossing once the student (and subsequently the teacher) told him that a train was coming. The most likely theory is that he may have been in too high a gear from the descend down to the crossing, further slowing the bus. Alternative theories imply that the rear bumper settled on the road, taking load off the rear wheels and causing them to slip or that they happened to be on the wet metal rail and thus momentarily lost traction. The bus had also had past issues with injectors in the engine failing, which would lead to a loss in power, but the engine couldn’t be tested after the accident to check this.
The bus’ failure to move adequately (or at all) likely caused the urgency in the driver to grow into panic, explaining how he might have then failed to engage the correct gear (first or second). But even if he did, the evidence at the site indicated that he drove the forward right wheel against the curb, just about bringing the slow bus (the speed was calculated to be no higher than 5kph/3mph on impact) to a standstill. An engine stall may have also occurred at this point, which was also the point at which the teachers told the children to move to the front of the bus, which was past the tracks. A moment later the train struck the bus, obliterating the rear section and throwing the bus into the yard of an adjacent house.
The bus was struck when just the very rear section of it was hanging over the rail line, a structurally fairly weak section containing little that could provide resistance beyond the engine (which was carved from the bus by the train) as the train happened to miss the rear axle by a hand’s width. Teachers who had realized the danger had already ordered students to move forward in the isle, but students from the rearmost rows couldn’t move fast and far enough forward, causing 6 of the 13 students in the section obliterated by the train to perish in the crash. One student happened to sit in the center seat of the rearmost row and pick up quickly on what happened, allowing him to flee forward in a straight line and survive with minor injuries.
The report concludes that a combination of mechanical (gearbox, engine, wheel slip) and human (panic, distraction) factors caused the immobilization of the bus in the crossing, aided by the crossing’s unusually poor shape for long/heavy vehicles which required high focus on the road and very low speeds. The time span between activation of the lights and the train reaching the crossing, while up to the requirements, was also deemed to be quite short.
The report closes with three recommendations. Firstly, the department of traffic, infrastructure and infrastructure development (Sétra) is advised to consider the creation of a list of vehicles to be banned from this crossing and ones like it due to being unable to make it from one side to the other in an acceptable time span. This recommendation is followed up on soon after the accident, with large trucks and busses being banned from using crossing 68. A second recommendation suggests the expansion of the time between the lights/bell activating and the barriers lowering to a point where any vehicle allowed over the crossing can make it past the other side’s barriers before those lower. This was intended to reduce the risk of a driver feeling “caught” behind the far-side barrier, hesitating to break through it. Lastly, the installation of duplicated red warning lights on the far side of the crossing is to be evaluated, giving drivers an indication of the lights’ status even if they passed their side’s lights.
The driver of the bus is put on trial in April 2013, almost five years after the accident, and is found guilty of manslaughter by negligence and unskillfulness by June. He is sentenced to a suspended 2-year jail sentence along with a five-year ban on driving public transportation vehicles. The SNCF and SAS Philippe Transport, the operator of the bus, are also found guilty, and are sentenced to pay 5.6 Million Euros in damages along with legal fees. Mister Baptendier, the father of a victim and representative of the parents of the children, gave a statement after the sentencing that called the result perfect justice, seeming okay with the sentences handed out.
The level crossing is retired in November 2015, with a new routing being built for the road that leads the road over an overpass southeast of the site and completely negates the need to drive through Mésignes. The eastern approach is turned into a small parking lot, while the western side disappears under grass. A small seam in the asphalt to the east of the crossing and a patch of tarmac to the west of the track is all that indicates where the road used to cross the rail line. A small memorial stands where the road used to connect to the crossing on the western side. The plaque reads:
In memory of our children, students of Margencel College, whose lives ended here on June 2, 2008.
In the collision of their bus with a train on this level crossing which was removed on December 1, 2015, our seven stars smiled at life, their memory will forever remain in our hearts.
Another Life Lost
Mister Jandin struggled with severe guilt after the accident, feeling responsible for the deaths and injuries as he had decided to not cancel the trip again. In his eyes, his decision to have the field trip take place on that day seemed to be the root cause of the tragedy, making him the main person responsible. He left his home without notice on the 16th of July 2008, 1.5 months after the accident. His wife contacted the police and warned them about his unstable mental state, with her worries being confirmed when officers found his body the next day after he had committed suicide in the nearby woods. His death was so closely linked to the accident that, when the trial finally took place years later, the SNCF was found guilty of manslaughter in his death.
_______________________________________________________________
I have chosen to stay off Medium’s monetization offers to keep these stories as accessible as possible, but you can support me with a small tip via “Buy me a coffee” if you feel like it.
_______________________________________________________________
A kind reader is posting the installments on reddit for me, I cannot interact with you there but I will read the feedback and corrections. You can find the post right here.
