Robert Doisneau’s iconic photograph of young love in Paris sold thousands of posters, but the identity of the couple remained a mystery for decades. In 1988 Jean-Louis and Denise Lavergne saw it on a magazine cover and thought they recognized themselves: They’d been on the rue de Rivoli on April 1, 1950, and had a diary to prove it, and Lavergne still had the skirt and jacket she’d worn that day. They contacted Doisneau, who greeted them warmly but did not offer to share any of the five-figure income he’d been making each year from the poster.

When they sued him, he revealed that he’d posed the shot using Françoise Delbart and Jacques Carteaud, a couple he’d seen kissing in the street but had not dared at first to photograph. Finally he’d approached them and asked them to repeat the kiss. Delbart said, “He told us we were charming, and asked if we could kiss again for the camera. We didn’t mind. We were used to kissing.”

A thousand bubbles burst, the Lavergnes lost their suit, and Delbart eventually sold the print Doisneau had given her to feign a spontaneous kiss. She didn’t share the proceeds with Carteaud — they’d broken up nine months after the photo was taken.

What is Chopin’s B Minor Sonata? What constitutes its identity? Not the fact that it’s part of Chopin’s conscious experience, because it continues to exist after his death. And not the fact that it’s part of any listener’s experience, because it continues to exist when those experiences have ended. It can’t be identified with any particular performance, and it’s different from its score, since the sonata is a sounding work and the score is an arrangement of graphic signs.

If the sonata is not material, and if it’s different from the experience of both the composer and the listener (in fact, it continues to exist if no one takes any conscious interest in it at all), how can it exist? How do we discern the same “original” work in a hundred different performances?

Is the sonata an ideal object, immutable and atemporal, like a mathematical concept? Well, no, because Chopin created it at a particular time. Perhaps there is no sonata, only individual performances? But then there’d be no sense in distinguishing a performance from the work itself, or in talking about the identity of a work (“Chopin’s B Minor Sonata”), or in arguing over whether a given performance was faithful to the original.

“For what is the point of saying that one performance rather than another gives a more nearly accurate account of the B Minor Sonata when the sonata does not in fact exist and when there is nothing real with which these performances may be compared?” asks philosopher Roman Ingarden. “Are we really going to agree that such judgments concerning the sonata itself and its performances are all false and stupid?”

(Roman Ingarden, The Work of Music and the Problem of Its Identity, 1986.)