← All posts

The Song Nobody Wrote

I was hunting for a track for the soundtrack when I found out that Leonard Cohen did not write The Partisan.

I had assumed he did, the way most people do. He was clear about it himself, and rather nicely: he said he felt he had written it, and then said that he actually had not, which is a more honest account of authorship than most of us manage. The song reached him through a chain of hands, and once I started following it back I forgot about the soundtrack entirely.

The music belongs to Anna Marly. She was born Anna Betulinskaya in Petrograd in 1917, during the revolution that killed her father, and she left Russia with her mother before she was two. She danced in Monte Carlo, studied under Prokofiev, sang in the Paris cabarets, and fled to London when France fell. There she ran a hostel for French exiles and wrote songs for a resistance she was watching from the wrong side of a channel. In 1943 a Free French leader named Emmanuel d'Astier de la Vigerie gave her words for one of them, and it became La Complainte du partisan. The sheet music was published in Paris in 1945, with d'Astier credited under his code name instead of his own.

So far the provenance holds. Then it gets translated.

Hy Zaret published an English adaptation in August 1944. He went on to write the words to Unchained Melody, so he knew the work, and most of his version is faithful. He sanded off the specifics, the Germans and France, so that the song could be about any occupation anywhere. That is a defensible call. The last verse is where he did something else.

D'Astier ends it like this. The wind blows over the graves, freedom will return, everyone will forget us, and we will go back into the shadows. It is a song about people who accept that their sacrifice will be anonymous and permanent.

Zaret ends it like this. The wind blows through the graves, freedom will come soon, and then we will come out of the shadows.

Same wind, same graves, opposite song. One is about being forgotten and choosing it anyway. The other is about being vindicated. One verse, changed by one man in 1944, and the meaning of the whole thing turns over.

Nothing in the process caught it. There was no step at which somebody held the French and the English side by side and asked whether they still said the same thing. Zaret's version went into a songbook, the songbook went to a summer camp, and in 1950 a fifteen-year-old Canadian found it there. Eighteen years later he took it into a studio in Nashville, decided the English alone was not enough, and had French singers recorded in Paris and laid over the top of the translation that had replaced them. The record runs five verses of Zaret's English, three of d'Astier's French, and then returns to the English to close. Both endings are on it. The one that gets the last word is the mistranslation.

Every hand in that chain did legitimate work. There is no forger anywhere in it. The song simply travelled far enough that the thing arriving stopped being the thing that was sent, and no checkpoint existed where the difference would have surfaced, because at no single step did the difference look large.

I had already written a novel about a weapon assembled the same way, out of changes that each passed review on its own. I did not know about the song until afterwards, and it would have saved me a great deal of explaining.

There is a woman in it who opens a commit history and finds work she never did, signed with her own key. The key lives on encrypted drives. She has never spoken the passphrases aloud. She reads the signature line again, and again, waiting for it to resolve into someone else's.

It does not. It is hers.