A car meeting a train is not a collision between two vehicles. It is a can meeting the shoe that steps on it. Everything that makes an ordinary crash survivable, the give, the crumple, the shared exchange of energy, is missing here, because only one side gives.

Start with the weights. A loaded freight train runs into the millions of pounds; a passenger car is a few thousand. The ratio is roughly a car to a soda can. When they meet, the physics is not a negotiation. The car does not slow the train in any amount a person could measure. All of the energy of the impact is spent on the smaller object, because the larger one barely registers the contact. The people who study these collisions put the disparity plainly: to a freight train, a car is what an aluminum can is to a car.

The second fact is the one drivers never believe until it is too late. A train cannot stop the way a car stops. Steel wheel on steel rail is a nearly frictionless pairing, which is exactly why trains are efficient to move and exactly why they cannot brake hard. A fully loaded freight at highway speed needs something on the order of a mile to come to rest. So when an engineer finally sees a car on the crossing ahead, the train is almost always already too close to stop. He can throw the brake and sound the horn, and after that he can only watch. The distance he needed was spent before he could see the danger.

The third fact removes the last thing a driver would reach for. A train cannot steer. It is captive to two rails, and there is no version of events in which it goes around anything. A car in a bad spot can brake, swerve, accelerate, choose. A train has one line and one speed it can only bleed off slowly. The entire burden of avoiding the collision sits, every time, on the person in the car.

And there is a perceptual trap folded into all of it. A large thing far away and a small thing up close look alike to the eye, and a train is very large, so it reads as farther off and slower than it is. A driver glances, judges there is time, and rolls onto the tracks on a misread. The warning devices at a crossing exist to take that judgment out of the driver’s hands. The law sorts crossings into two kinds. A passive crossing gives only a crossbuck and asks the driver to look and decide for himself. An active crossing adds flashing lights, bells, and gates that come down when a train is detected. The national manual of traffic-control devices builds its hierarchy on a simple idea: the busier and faster the crossing, the less it should be left to a driver’s glance.

These collisions are not rare curiosities. Grade crossings are the second leading cause of rail-related death in the country, with more than two thousand incidents and around two hundred deaths in a typical year. And here is where a case lives. When one happens, the reflexive story is driver error, because the driver is the one who could have stopped. But the harder questions point at the crossing itself. Was a busy, fast crossing left with nothing but a crossbuck? Were the lights and gates working, and timed to give real warning? Was the sight line down the track swallowed by brush or a parked string of cars? Was this crossing already on a public authority’s list of dangerous ones, with money available to fix it and nothing done? The driver had one path. The people who designed and maintained the crossing had many, and the choices they made are evidence.

So when the story is a car and a train, do not accept that it begins and ends with the driver’s last glance. The train could not stop and could not turn; that was settled by physics before anyone woke up that morning. What was not settled was the crossing. Ask who decided how much warning the driver would get, and whether that decision is the reason there was no time left.

On the physics of the grade crossing, the mass and momentum disparity and the mile-plus stopping distance of steel wheel on steel rail, see How Railroad Crossings Work (Practical Engineering). On the toll, grade crossings as the second leading cause of rail-related death, see the Federal Railroad Administration. The warning-device hierarchy is set by the Federal Highway Administration’s Manual on Uniform Traffic Control Devices (Part 8, highway-rail grade crossings). This is general information, not legal advice.