It was 11th September 1940. Exactly sixty-one years later, two planes would slam into the World Trade Center, and a third into the Pentagon, killing thousands and leading to the deaths of millions more. But on this day, a year before construction on the Pentagon began, although dark things were afoot in Vichy France, all was quiet in the Vézère Valley, where, outside the village of Montignac, eighteen-year-old Marcel Ravidat was walking his dog in the early autumn sun.
A year earlier, Ravidat had attempted to enlist with the French Army, but was turned away on account of his youth. If not for that accident of history, one of the greatest discoveries in the study of prehistory may never have been made at all. Pursuing a rabbit, Ravidat's dog Robot, a real organic dog named after the artificial organic workers in Karel Čapek's 1920 play R.U.R., was nosing around the base of an old, uprooted oak tree when it lost its footing, slipped and tumbled into a newly unearthed crevasse. Initially alarmed, Marcel Ravidat's concern turned to curiosity after he managed to retrieve Robot unharmed and in the process came to realize that the rabbit hole went very deep indeed.
Ravidat returned with three teenage friends and clambered down into the cave, which, they soon discovered, was filled with vibrant depictions of animals. In a manner reminiscent of the Famous Five, at first they had fantasized that they might have uncovered a legendary secret passage to Lascaux Manor, but they quickly came to realize that what they had on their hands was something altogether more ancient. They were put in touch with the priest and prehistorian Henri Breuil, who upon visiting the site on 21st September was able to establish that these artworks were thousands of years old. And that, in a nutshell, is the story of the discovery of the caves of Lascaux.
Except it isn't. The above story never happened. I lied to you.
The discovery of Lascaux didn't take place on the eleventh of September, but I liked the idea of juxtaposing this pastoral scene with 9/11. (Funnily enough, 11th September 1941 indeed seems to be the day ground was broken in the construction of the Pentagon.) I have no idea whether it was a sunny day, or whether the uprooted tree was an oak tree, though both things might have been true. The story of Ravidat trying and failing to join the Army is also one I made up wholesale. And these are just the aspects I intentionally misrepresented; I may well have got more wrong by accident.
Some of the stranger aspects of the story are true, as far as I can tell. The dog was indeed called Robot, although the English Wikipedia article mysteriously fails to mention this important fact. Why the dog was called Robot may have had something to do with Rossum and his Universal Robots, directly or indirectly, or it may not; I have no evidence of any direct connection. Breuil was indeed a priest, although by and large the era in which amateur antiquarian clergymen dominated the European intellectual scene was otherwise over by 1940.
Still other aspects of the story are untrue, but not made up by me. The idea that Robot fell into the cave is, according to Wikipedia, an embellishment made up by Ravidat himself. This is supported by the more extensive version of the story in Gregory Curtis's The Cave Painters (p88), which also makes no mention of the rabbit found in other accounts. Meanwhile, the actual date of the events described above is not clear: the Wikipedia page at least strongly implies that it was 12th September (not 11th), but this site says it was the 8th. But that site also says Ravidat was 17 at the time, whereas we know that his date of birth was 16th May 1922, in which case he must have been 18.
Misinformation also surrounds one of Ravidat's three human companions, Simon Coencas, a Jewish refugee who'd only moved to Montignac with his family in June of the same year. Less than a week after the meeting with Breuil, a German-led law enforced by the Vichy regime required all Jews to register, which they duly did, travelling to Paris for this purpose. In 1942 the whole family was arrested. This page writes of "Simon and his family, deported [plural agreement] to Auschwitz". But this page makes it clear that Simon and his sister, by virtue of their young age, were spared the fate of the rest of the family, who, like so many, did not survive Auschwitz.
Simon Coencas died on 2nd February 2020. Just over six years later, on 11th February 2026, my early morning bus crosses the river Vézère, its waters a turbulent chocolate brown, well above the level of the riverside path. After an avocado toast in town (well, I already own a flat), I trudge out to the site of Lascaux IV. It's an ugly concrete building, better set up for coachloads than it is for the lone Brit in February who, it turns out, is the only person signed up for the English-language tour that morning. I think my guide is recovering from a cold. Following a spiel about the discovery of Lascaux – repeating the "dog fell in" trope – she produces a key and unlocks the cave itself.
It's easy to see how the teenagers, and even Abbé Breuil himself, would have been impressed. The entrance area, the Bull's Chamber, is a spacious dome full of enormous bulls in the three colours of Lascaux: black (manganese oxide), ochre and red (iron oxide). Beyond it is a corridor-like space whose ceiling is covered in horses. It's equally easy to see why Breuil called the cave the "Sistine Chapel of Prehistory". Outside her script, my guide isn't very talkative, but silent wonder suits me too. I stride on, almost alone in this legendary space, through areas named the Nave and the Apse, past an image of two bulls facing away from one another, partly overlapping, demonstrating that Cro-Magnons twenty thousand years ago had an understanding of the art of perspective.
This, too, dear reader, is a lie. Well, sort of. What I called "the cave itself" isn't the cave itself, in the context of Lascaux IV, but rather an impressively precise reconstruction. Sixteen pixels per millimetre, according to my guide. Other than a few black connecting corridors to better funnel the hordes of gawkers onwards, it is picture perfect. Not olfactorily perfect, though: my guide, who has visited the original, tells me that the smell of the true Lascaux cave is really quite disgusting. They could easily recreate it, she says, but no one particularly wanted to do so.
If my mendacity hasn't driven you away yet, thanks! And sorry. I've just become fascinated with replicas and replication of all kinds. In 2025, at about this time of year, I visited Warsaw, whose baroque castle is an elaborate 1970s reconstruction (by a Communist government, no less), and Vilnius, whose Palace of the Grand Dukes has even less claim to authenticity. History is clearly important to people! The original Lascaux cave had to be closed to the public in 1963, once it became clear that the collective exhalations of a million visitors had a deleterious effect on the object of their awe. Twenty years of tourism threatened to destroy what had lain undisturbed for twenty thousand.
And so, in Warsaw, in Vilnius, in Montignac, around the world, people throng to visit impressive fakes. Lascaux IV is the third replica; Lascaux II, the first replica, is still popular with older generations who visited it in their youth, and Lascaux III is a sort of peregrinatory Baba Yaga's hut of a cave, hopping around the world to bring the replica to the people. But perhaps "fake" is unfair: people seem happy enough, in many such cases, to accept the replica as the real thing. The Ship of Theseus, but if the whole ship had been blown to smithereens by a German U-Boot and a replica had been built in the Cold War period.
With written histories, meanwhile, it's harder to tell real from replica. Perhaps this is because any history is a representation of the events it purports to narrate, rather than the events themselves: the map is not the territory. In this sense, history is replica by definition, and I've been writing recently, following Hayden White, about how it must take narrative form, which constrains it to share some of the properties of fiction. Seen this way, the replica-hood of history is virtually a trivial, necessary property. And yet clever people have felt the need, over and over again, to reassert the distinction: in linguistics, for instance, Ernst Pulgram's Real Proto-Indo-European versus Reconstructed Proto-Indo-European. This suggests that other clever people often fall into the trap of ignoring it, and of mistaking the replica for the original.
The cave painters of Lascaux must have been in a position to make the distinction, since they were capable of representation. The paintings of bulls in Lascaux are not bulls: ceci n'est pas un taureau. More generally, evidence of symbolic behaviour is everywhere in Lascaux, even if we don't know what the signified is in many cases: gate-like geometric structures, series of dots. In that sense, Lascaux IV is a replica of a replica, the main difference between the two processes being only that the second replication had a much higher degree of verisimilitude than the first.
(I find it interesting that this page refers to the four teenage discoverers as "inventeurs"; the related English word "inventors" is one we wouldn't use in this context, as it implies creation from scratch. But we can say something like "Edison discovered the electric lightbulb", as if he were out walking one day when his dog dug it up from under an uprooted tree in the Dordogne. Weird.)
Younger me was fiercely against intentionally deceiving anyone, ever, for any reason. So I'd get in trouble for saying things like (to my then girlfriend) "You're the second most beautiful person I've ever met". Older me thinks that, an age in which yet more clever people are being taken in by "AI", perhaps there's value in fakers like Alan Sokal and Jonas Bendiksen, as long as they reveal themselves – if only because their work reminds people of the difference between real and replica, and of the fact that the difference between types of replica largely boils down to creator intentions.
