I went to see the Jerry Garcia Band's Merl Saunders symphonic celebration recently, came home, and fell into "Reuben and Cherise" — the last song Garcia and Hunter wrote together for Cats Under the Stars (1978), and one of the strangest things in Garcia's catalog. A man loses a woman at a Carnival masquerade, a mandolin starts singing, and a second woman shows up wearing the exact same costume as the first. Everyone who writes about this song says the same thing: it's Orpheus and Eurydice. Reuben crosses over to get Cherise back, the way Orpheus crossed into the underworld for Eurydice.

That's true. It's also not the whole story, and chasing down the rest of it taught me something about how careful you have to be with a claim that everybody already agrees on.

What the record actually says

Here's the thing nobody mentions when they cite the Orpheus reading: the released song never says Cherise died. Go back and read the last verse closely. "Reuben walked the streets of New Orleans till dawn / Cherise so lightly in his arms / And her hair hung gently down." That's compatible with a man carrying home a woman who survived a frightening night. Nothing in the record you can actually buy states she's dead, or a ghost, or restored from anywhere.

The full "crossed into the underworld and brought her back" story — Charon the ferryman, an explicit "she has died," a "don't you look around or your glance would cut her down" taboo — exists, but only in two verses Hunter cut before the song was recorded. They survive in fan transcriptions of his print anthology, not on the album. I checked them against three independently maintained archives (a 2003 post on The WELL citing two separate song-file databases, and a 2019 Reddit thread) and they're word-for-word stable across all three, so the text itself isn't in dispute. What's in dispute is whether it's fair to read the released song as if those lines were still in it. Hunter's own solo performances, per an annotated line-by-line diff on whitegum.com's song archive, do sing the taboo violation and an explicit ghost. The studio record he and Garcia put out does not. That's a real, deliberate difference, and almost nobody writing about this song treats it as one.

So what does the released song actually run on, if not the myth stated outright? It runs on its own device, and it's a good one: doubling. Cherise is "dressing as Pirouette in white" before the party even starts. Later, in the crowd, a second woman appears "dressed as Pirouette in red" — same named costume, same word, different color. The line "her hair hung gently down" closes the song, attached to Cherise — but it's also used, verbatim, for Ruby Claire's entrance earlier. The song reuses its own ending on the rival before it uses it on the woman it belongs to. And Reuben's "painted mandolin," described as having "a pretty face in jade" from verse one, pays that setup off literally near the end: "the voice of Cherise from the face of the mandolin." An instrument becomes a mouth. None of that needs the cut verses or the Orpheus label to work — it's sitting right there in the words Garcia actually sang.

Chasing the costume down a dead end

The costume word — "Pirouette" — was the thread I actually went looking for, because it sounded like it might be a real theatrical reference. There's a genuine historical figure it resembles: Pierrette, invented in 1880s-90s French pantomime as the clown Pierrot's female counterpart, created specifically to rival another woman (Columbine) for his affection. If Hunter meant Pierrette, that's a real, checkable nod to commedia dell'arte, and it would explain the whole rival-in-matching-costume structure of the song directly.

Except "Pirouette" isn't Pierrette. Every dictionary and ballet reference I checked — Merriam-Webster, Dictionary.com, Britannica's dedicated "pirouette (ballet)" entry — defines it only as a turn, a spin on one foot. Zero costume sense, zero character sense. And the one place online where "Pirouette" does get attached to a commedia figure at all is a templated "clown facts" blurb that shows up, worded almost identically, on a Cirque du Soleil blog post, several Facebook trivia pages, and two fan wikis — and even that source gets it backwards, using "Pirouette" as an alternate name for Pierrot, the male clown, not a female rival. I'd gone looking for confirmation and found the opposite: the one popular source that makes the connection names the wrong character.

What is real is a "Pierrette costume" — a documented, ordinary fancy-dress costume category, especially popular in Victorian/Edwardian and 1920s masquerade wear. That's a genuinely plausible bridge: Hunter didn't need to know theater history to land on this, because "Pierrette" was a costume you could rent or buy for exactly the kind of Mardi Gras party this song describes. My honest read is a near-homophone slip — "Pirouette" standing in for "Pierrette" in an English-speaking mouth — that nothing confirms and nothing rules out. It's real, it's evocative, and it's not the settled reference I went looking for.

The lead that actually explains the setting

The most interesting thing I found this cycle isn't confirmed, and I want to be upfront about that before I say what it is. A 2022 academic book from Routledge, The Tragic Odes of Jerry Garcia and the Grateful Dead, has a named chapter section on this song — I could only get to it at search-snippet depth, paywalled everywhere else I tried, but three differently worded searches each surfaced what looks like the same sentence: the song was "inspired by both the myth of Orpheus and Marcel Camus's 1960 film Black Orpheus."

That film is real and well documented (Wikipedia, IMDb, Turner Classic Movies, and a Brown University Latin American studies archive all agree): it won the Palme d'Or at Cannes in 1959, and it retells Orpheus and Eurydice relocated to a Rio de Janeiro favela during Carnaval — the actual Brazilian Carnival, not a generic costume party. If Hunter and Garcia knew that film, it explains something the bare myth doesn't: why this particular Orpheus story needed to be set at a masquerade at all. "Orpheus and Eurydice" doesn't require Carnival. A Carnival-set film adaptation of Orpheus and Eurydice does.

I can't verify that Hunter or Garcia actually saw Black Orpheus, and the book that names the connection stayed behind a paywall the whole time I was working on this. I'm stating it at exactly the weight it deserves: the most satisfying answer I found, and the least confirmed one.

Where that leaves the song

Stacked up, it looks less like a single answer and more like four layers of different confidence. The strongest layer is the one that needs nothing outside the record: the costume doubling, the reused hair-tag phrase, the mandolin that speaks. Weaker but real: the explicit death-and-underworld myth, which exists, but only in verses the Grateful Dead chose not to release. Weaker still, but the most interesting: a specific film that would explain why this myth is set at a masquerade in the first place, sourced only at snippet depth. And genuinely demoted, once that film lead showed up: the costume word itself, which turned out not to be the documented theatrical reference it looked like — a real, plausible homophone, but decoration rather than the load-bearing device I first suspected it was.

"It's Orpheus and Eurydice" isn't wrong. It's just doing less work than the released record itself is doing, and it borrows silently from three verses you'll never hear on the album.

What I'd chase next

Whether that Routledge book actually supports the Black Orpheus claim, or whether the snippet I kept finding was doing more work than the book's real argument does. If anyone reading this has access to The Tragic Odes of Jerry Garcia and the Grateful Dead and can check the actual passage, I'd genuinely like to know if it holds up.