They Borrowed Snow From A Cartoon And Ribs From General Motors.
Nine people died on a slope too shallow to avalanche. Two engineers modelled it with animation code from a Disney film and injury thresholds from cadavers crash-tested in the seventies. They say plausible. The headlines said solved.
A months-long reporting project. Documents cited below are held in The Vault and available to readers.
ZURICH — Nine people died on that slope in February 1959. They were students and they were experienced, and six decades of argument about what happened to them has been conducted mostly by people who never met them.
This story is about a model. Their story is somebody else's and is better told elsewhere; this newspaper has pointed at the piece it thinks does it properly.
What follows is where the model came from, which almost nobody has reported.
The objection everyone starts with
The slope is too shallow.
That has been the standing argument against an avalanche for sixty years, and it is not a foolish one. Slab avalanches are associated with steeper ground, and the ground there does not look like avalanche terrain to anybody standing on it.
What the two engineers did about it
They built a model, and to build it they went shopping.
For how snow moves, one of them went to an animation studio in California, having noticed that a film about a snow queen depicted moving snow better than anything in his own literature. He asked the animators for their code. They gave it to him. He modified it for a purpose the studio had not intended, which was to simulate what a moving slab does to a human chest.
For when ribs break, they needed thresholds, and thresholds come from bodies. They found them in a car company's archive.
That data exists because somebody wanted seatbelts to work. It was collected for a purpose that has nothing to do with mountains, by people who would have been baffled to learn where it ended up.
For what was found, they used the 1959 autopsies: rib and skull fractures recorded by a Soviet pathologist and left unexplained in the file for sixty years.
What it produced
A slab, released after a cut was made into the slope to pitch a tent, loaded overnight by wind coming down off the ridge, sliding as a block rather than as loose snow, and delivering the kind of blunt impact that breaks a rib cage without breaking the skin.
The injuries in the model resemble the injuries in the file.
The word in the paper
Plausible.
That is what the authors wrote and it is what they meant. The paper does not say solved, does not claim to know what happened on 1 February 1959, and offers no account at all of why the tent was cut open from the inside.
The coverage said solved. Some of it still does.
The part that persuaded this desk
Not the model.
After the press attention, three separate parties went to the pass and looked. Their reports record avalanche activity at that location.
That is a check nobody designed. The paper made a claim about a place, the claim was loud enough that people went to the place, and what they found is consistent with the claim. It is the closest thing to a test that a sixty-year-old event allows, and it happened as a side effect of publicity rather than as anybody's plan.
What our snow physicist wanted separated
Professor Vesna Aitmatova-Reinholt agrees with the mechanism and objects to how it has been received. Her statement is at the foot of this story and it makes a distinction this newspaper intends to keep.
There are three things here and they have been welded together. Whether the mechanism is possible: yes, and the work is careful. Whether it happened: unknown, and the paper says so. What the public heard: solved.
Her worry is not for the two engineers. It is that if somebody one day shows this was not an avalanche, the damage will land on the idea that a physical model can tell you anything — because people will remember the word solved.
The authors did not overclaim. Their profession's credit was spent on their behalf by other people, and none of those people will be asked to pay it back.
What is left unexplained
The tent was cut from the inside.
No model addresses that, the authors do not pretend otherwise, and this newspaper is not going to fill the gap with an adverb. Nine people left a shelter in the dark, in that cold, having made a hole in their own tent to do it.
A slab avalanche is a plausible reason to be very frightened of a tent. It is not a witness, and there were none.
The six o'clock edition
Three stories a morning, in your inbox before the coffee is made.
Sources & Method
We read the paper before the coverage of it, which turns out to be the whole difference. Then we traced each component of the model back to where it came from — an animation studio, an automotive safety programme, a Soviet autopsy table — because the provenance of a method is a fact about how much weight it can carry, and nobody was reporting it.
Who we spoke to
- The 2021 paper and its supplementary material, Peer-reviewed, published in an open-access journal. Read in full, including the sections on model calibration 2026 Published on the sixty-second anniversary, to the day, of the party setting out. The authors say this was not planned.
- The 1959 investigation file, Soviet prosecutorial record; released in stages since 1990. Read in translation, including the autopsy findings 2026
- Prof. Vesna Aitmatova-Reinholt, Snow mechanics; not an author of the 2021 paper and disagrees with part of it. Interviewed by video call twice June 2026 Thinks the mechanism is right and the confidence in the press coverage is not. Her objection is printed.
- Post-publication expedition reports, Published; three parties visited the pass after the 2021 coverage. Read; they record avalanche activity at the location 2026 This is the part that moved this desk from interested to persuaded, and it happened because of press attention rather than because of a plan.
Documents
- PX-2173 — The 2021 slab-avalanche paper, filed with the 1959 autopsy findings it is calibrated against and the post-publication expedition reports accepted
What we could not confirm
- What happened that night. A model that reproduces the observed injuries from the observed conditions is a strong candidate mechanism. It is not a witness and there were none.
- Whether this slope slid on that night. The 2021 work shows the mechanism is possible there; later expeditions record that avalanches do occur there. Neither establishes that one occurred on 1 February 1959.
- The nine minutes. Nothing in any model explains why a tent was cut open from the inside rather than unlaced, and the authors do not claim it does.
How Others Covered This
The same events, as reported elsewhere on the same day. We list what each outlet had that we did not, as well as what we had that they did not — including where we come off worse. Why we print this.
-
ApertureThe Pass
A long piece about the nine, who they were, and the families who have spent six decades being told different things.
Had that we did not
The people. Ours is a piece about a model and theirs is a piece about a party of students, and theirs is the one to read first.
Left out
The mechanics.
-
The Meridian TelegraphSCIENTISTS ADMIT THEY STILL CANNOT EXPLAIN IT
Reported the word 'plausible' as a confession of failure and the state-weapons account as the remaining candidate.
Had that we did not
That the paper says plausible. It does.
Left out
Everything about why a physical model that produces the observed injuries from the observed conditions is a different kind of object from a story.
-
They Buriedthis newspaperThey Borrowed Snow From A Cartoon And Ribs From General Motors.
Reported where each part of the model came from, and what the authors actually claim.
Had that we did not
The provenance of the method, which is the most interesting thing in the story and appears almost nowhere.
Left out
Our headline is about the tools and not about the nine people, and we made that choice knowing it reads as levity. — V. Ashcombe-Doyle, standards editor
Right of Reply
They Buried contacted Prof. Vesna Aitmatova-Reinholt, snow mechanics read the full draft; no deadline. Agrees with the mechanism and objects to how it has been received. Printed unedited.
I want to separate three things that have been welded together in public and should not be.
The first is the mechanism. A slab avalanche on a slope of that angle, with that snowpack, after a cut into the slope and a night of wind loading, is physically possible. I have no quarrel with that. It is careful work and the calibration against the injury data is ingenious.
The second is whether it happened. That is a different question and the paper is honest about it. The word in the title of the relevant section is plausible, and plausible is not a synonym for demonstrated.
The third is what the public heard, which is solved.
Here is what bothers me about that. If in ten years somebody produces evidence that this was not an avalanche, the damage will not be to the two engineers, who said plausible and meant it. It will be to the idea that a physical model can tell you anything at all — because people will remember the word solved and will conclude that science overclaims.
The authors did not overclaim. Their profession's ability to be believed was spent on their behalf by other people, and none of those people will be asked to pay it back.
So print plausible. Print it in the same size type as the cartoon.
Published unedited under our right-of-reply guarantee.
How was this story?
We publish the result, whatever it is. Reader verdicts appear on the front page and in our newsroom metrics.
Readers' Letters 0
Printed at once under the name you give and read by the desk afterwards; anything unfit is removed, with a note saying so, and nothing else is ever deleted — only corrected. Letters that changed something in the story carry a mark saying so, and there are 5 of those across the archive.