What Happened to the Hollywood Musical?
I've been working on this piece, in one form or another, for about three years. It started as a note I made to myself after watching New York, New York in a film class in 2022 - I scribbled in the margins of my notebook, "why does this feel like a funeral?" - and it became, gradually, an obsession that sent me back through roughly a hundred films across eighty years of American cinema, more than a dozen books on film history and the Broadway musical, scattered academic papers about genre theory, and more hours of late-night TCM watching than I can honestly account for without embarrassment. The monthly recaps I write for this blog are one form of attention. This is another form: slower, messier, and aimed at a question that I kept not being able to answer to my own satisfaction.
The question is simple and the answer is not: what happened to the Hollywood musical?
Not why it declined - everyone agrees it declined, and the proximate causes are well-documented - but what the decline means, and whether "decline" is even the right word for what happened. The Hollywood musical didn't disappear. It mutated, subdivided, went underground, colonized other genres, and occasionally resurfaces in forms that are either triumphant or desperate or both simultaneously. To say it died is too easy. To say it's alive is dishonest. To understand what actually happened, you have to go back to where it started and trace what it was, at each stage, trying to do.
I've spent three years trying to do that. This is what I found.
~
Part One: What the Musical Was For
The Hollywood musical as a form was born from a technological accident and a cultural crisis that arrived in the same year. The Jazz Singer opened in October 1927 and by 1929 the silent film was commercially dead, replaced by something that could speak and sing and whose relationship with its audience was suddenly, disoriently different. The first sound films were musicals not because Hollywood planned it that way but because music was the most obvious demonstration that the technology worked - you could see a performer and hear them simultaneously, the body and the voice finally unified - and audiences went delirious for it.
The first years of the sound musical were a chaos of experimentation. The Broadway Melody (1929) won the second Academy Award for Best Picture and is barely watchable now. The Gold Diggers of Broadway (1929) was a massive hit and is lost. Hollywood Revue of 1929 is a variety show on film, barely coherent as a narrative object. The studios threw everything at the new technology and what emerged, by 1930 or so, was a set of conventions that would define the form for the next twenty-five years: the backstage musical, in which the story was about putting on a show and the numbers emerged naturally from that context; the operetta, in which European romantic conventions were transplanted to Hollywood and given to singers like Jeanette MacDonald and Nelson Eddy; and the integrated musical, in which the numbers were meant to emerge from the emotional situations of characters who sang because the emotion exceeded what speech could hold.
That last form is the one that matters most, and it took the longest to perfect. The backstage musical - 42nd Street (1933), Gold Diggers of 1933, Footlight Parade (1933), the whole Busby Berkeley cycle - solved the problem of the numbers by not pretending to solve it: of course people sing and dance in these films, because the film is about people who sing and dance for a living, and the stage becomes the space where the film's fantasy dimension can operate without pretending to be naturalistic. Berkeley's overhead camera transforming bodies into geometry, chorus girls becoming kaleidoscopes, the stage floor becoming infinite - none of this is meant to be realistic, and the backstage frame gives the film permission to abandon realism without apology.
The integrated musical asked something harder: for the songs to emerge from the drama, for singing to feel like a natural extension of speech rather than an interruption of it. The Rodgers and Hart, Rodgers and Hammerstein, and Cole Porter shows that Hollywood adapted across the 1940s and 1950s were already partly integrated on Broadway - Oklahoma! (1943) had transformed the form's relationship to narrative, to dance, to the idea that a musical could be about something rather than merely charming - and Hollywood learned from these shows how to make films where the songs carried dramatic weight rather than simply decorating it.
The peak of this form is the MGM musical of the late 1940s and early 1950s, and the peak of that peak is a group of films produced by Arthur Freed's unit at MGM: Meet Me in St. Louis (1944), On the Town (1949), An American in Paris (1951), Singin' in the Rain (1952), The Band Wagon (1953), Brigadoon (1954). These films represent the form at its most confident and most complete - the songs serving the characters, the choreography serving the story, the color and spectacle serving an emotional truth rather than substituting for it. They are what the musical was when it was most fully itself, and understanding what happened afterward requires understanding what made them work and why those conditions couldn't be sustained.
What made them work was a confluence of talent, infrastructure, and cultural moment that is almost impossible to reconstruct in retrospect without sounding like you're describing a golden age that never quite existed. But it did exist. The Freed unit at MGM had, at various points, Gene Kelly and Fred Astaire and Judy Garland and Cyd Charisse and Frank Sinatra; it had Vincente Minnelli and Stanley Donen directing; it had Betty Comden and Adolph Green writing; it had André Previn and Saul Chaplin arranging; and it had the studio infrastructure - the sound stages, the costume departments, the rehearsal time built into actors' contracts - to support the months of preparation that films like Singin' in the Rain and The Band Wagon required. You could make a film where Gene Kelly spent months choreographing a seventeen-minute ballet sequence because the studio system paid for it, built the stage for it, and understood it as a legitimate use of resources.
The cultural moment was equally specific: the postwar audience wanted beauty, wanted pleasure, wanted the demonstration that life could be choreographed into something more vivid and more joyful than the world they'd lived through in the 1930s and 1940s. The musical's utopian dimension - its argument that human bodies could move with this kind of grace, that human voices could fill a room with this kind of warmth, that love and community could be organized by music into something approaching the sublime - was not escapism in the pejorative sense. It was a genuine answer to a genuine need.
By the mid-1950s, both conditions were beginning to erode.
~
Part Two: The Television Problem and the Spectacular Solution
The Hollywood studios identified television as an existential threat by 1950 and their response was to do what threatened industries always do: double down on what television couldn't replicate. Television was small, black and white, broadcast at low resolution into living rooms. Hollywood's answer was to become bigger, louder, more colorful, more spectacular - CinemaScope and VistaVision and Todd-AO, widescreen processes that made the cinema image something television couldn't approximate. And the musical was the form most naturally suited to the spectacular response, because spectacle was already part of its DNA.
The Broadway adaptations of the 1950s - Kiss Me Kate (1953) in 3D, Oklahoma! (1955) in Todd-AO, Guys and Dolls (1955), Carousel (1956), The King and I (1956), South Pacific (1958), Porgy and Bess (1959) - were prestige productions, expensive and technically ambitious, and they performed respectably. But something was already shifting. The intimacy that had made the Freed unit MGM musicals so emotionally precise was harder to maintain at the scale these Broadway adaptations required. South Pacific is an example of what happened when scale overwhelmed the material: Rouben Mamoulian directing a film shot in a process so wide that the human face becomes a small element in a landscape, the famous colored filter sequences bathing the film in washes of magenta and yellow that Rodgers and Hammerstein detested, a technical ambition that produced something alienating rather than transporting.
The exceptions prove the rule. The King and I works because Walter Lang directs it with enough restraint to let Yul Brynner and Deborah Kerr's performances breathe. Guys and Dolls works in patches because Mankiewicz understood the material's irony - Guys and Dolls is fundamentally a satirical musical and Mankiewicz was a satirist - and because Sinatra and Brando's incompatible acting styles create a genuine energy that the film's formal competence couldn't have generated alone. These are films where the personal talent of the performers outran the industrial logic of the production, which is the opposite of how the Freed unit had worked.
The industrial logic of the production was increasingly what the 1950s Broadway adaptation cycle was about. Studios paid enormous rights fees for proven Broadway properties because the brand recognition reduced the financial risk. The choreography was imported from Broadway. The songs were already written. The sets were built to specifications that came from the original staging. The result was a form of prestige filmmaking that was simultaneously expensive and conservative - films that tried to preserve rather than transform, to deliver what Broadway had already accomplished rather than discover what film could do with the material.
There were two significant exceptions to this pattern in the late 1950s, and they point in opposite directions toward what would come next.
Gigi (1958) was the last great MGM Freed unit musical, and it is in some ways the form's final statement of confidence. Minnelli directing in Paris, Leslie Caron and Louis Jourdan and Maurice Chevalier and Hermione Gingold, Alan Jay Lerner and Frederick Loewe's score, and a film that uses its Parisian settings not as backdrop but as a moral environment - the Belle Époque world of courtesans and arranged relationships against which the love story defines itself. Gigi won nine Academy Awards including Best Picture. It was the last musical to win Best Picture for twenty-four years.
West Side Story (1961) - technically the early 1960s but the product of the late 1950s cultural moment - was the other exception, pointing in the opposite direction: toward a musical that used the form's conventions to address social reality, that refused to separate the dancing from the violence, that ended with two deaths instead of a wedding. Jerome Robbins and Robert Wise directing, with Robbins doing the choreography and Wise doing the dramatic scenes and the two halves of the film existing in genuine tension with each other. The Jets and Sharks sequences have a formal ferocity - the bodies in those opening minutes moving through the streets of New York with a kinetic aggression that still doesn't look like anything else - and the film dares to insist that what happens to Tony and Maria is a direct consequence of the world the dancing depicts, not a separate emotional register. It won ten Academy Awards and remains the most formally ambitious Hollywood film made from a Broadway source, and part of what makes it remarkable is that it was immediately clear that nobody could sustain the approach. West Side Story required Jerome Robbins, and Jerome Robbins was nearly impossible to work with and had already been removed from the production before it wrapped.
Between Gigi and West Side Story, the Hollywood musical of the 1960s was trying to hold two incompatible ambitions simultaneously: the prestige spectacle of the big Broadway adaptation, and the social seriousness that the culture was beginning to demand. It mostly failed at both.
~
Part Three: The 1960s and the Great Overextension
The 1960s Broadway adaptation cycle produced four films that are essential to understanding where the musical went wrong before it went anywhere interesting: My Fair Lady (1964), The Sound of Music (1965), Camelot (1967), and Hello, Dolly! (1969). Two of them are genuinely good films. All four of them contributed to a catastrophe.
My Fair Lady is the best of the group and the most purely competent. Cukor directing Audrey Hepburn and Rex Harrison, Cecil Beaton's Oscar-winning costumes, Frederick Loewe and Alan Jay Lerner's magnificent score - the film is the Lerner and Loewe Broadway production preserved on film with professional care and without particular imagination. It is the best possible version of the thing it was trying to be, which is a filmed record of a great stage production, and if you believe that filmed records of great stage productions are what Hollywood musicals should aspire to, My Fair Lady is your film. I find it graceful and slightly lifeless, which is the precise opposite of the Freed unit musicals that preceded it. Harrison won the Oscar for a performance that is technically brilliant and emotionally cold. Audrey Hepburn's singing was dubbed by Marni Nixon and she was not nominated, which remains one of the more egregious Academy failures of the decade.
The Sound of Music is the other good film, and if you say so in certain company people look at you like you've said something shameful, but it's true. Wise directing in Austria, Julie Andrews two years after Mary Poppins at the absolute peak of her particular quality - the warm-but-precise, joyful-but-controlled quality that made her the last great old-studio musical star - and a film that handles the Nazi subplot with more directness than the source material and uses the Austrian landscape with the visual intelligence that the earlier South Pacific process photography had missed. The songs are the best Rodgers and Hammerstein wrote for film specifically (as opposed to stage adaptations) and Andrews sings them with a clarity that is genuinely moving rather than merely accomplished. The Sound of Music made more money than Gone With the Wind and its success was the direct cause of the catastrophe that followed.
Because the lesson the studios took from The Sound of Music was not "make emotionally precise musicals with great songs directed by filmmakers who understand landscape" but "spend enormous amounts of money on Broadway properties and they will make enormous amounts of money back." The lesson was wrong and the films it produced were wrong. Doctor Dolittle (1967), Camelot (1967), Star! (1968), Goodbye, Mr. Chips (1969), Hello, Dolly! (1969), Darling Lili (1970), Paint Your Wagon (1969): a run of expensive, lumbering failures that nearly destroyed the major studios and created the conditions for the Hollywood Renaissance that followed. These films were made on the assumption that scale was sufficient - that big enough budgets and big enough stars and big enough source material would guarantee big enough audiences - and the assumption was catastrophically wrong.
Hello, Dolly! is the clearest example of how wrong it was. Gene Kelly directing Barbra Streisand and Walter Matthau, a $25 million budget in 1969 dollars (roughly $200 million today), a score by Jerry Herman that had been running on Broadway for five years, and a film that feels like a mausoleum for the form it's celebrating. The set piece on the staircase at the Harmonia Gardens is technically spectacular - hundreds of extras, a set of a scale that no musical since has attempted, Streisand descending in gold - and it produces almost nothing in the way of emotional response. The spectacle has swallowed the film. There is nobody left inside it.
Meanwhile, at the opposite end of the budget spectrum, something else was happening.
~
Part Four: The Underground Current
While the major studios were overextending on spectacular Broadway adaptations, a different kind of musical film was developing at the margins - less celebrated, less discussed, but in retrospect more important to where the form would eventually go.
A Hard Day's Night (1964) is the film that nobody describes as a musical but is entirely structured as one - the Beatles performing, the film moving between their numbers with a documentary looseness that the staged musicals around it didn't have, Richard Lester directing with a formal intelligence borrowed from the French New Wave. The film doesn't have a plot in the conventional sense because it doesn't need one; it has the music, and the music is the argument. What A Hard Day's Night demonstrated was that the integration problem the Hollywood musical had been solving through elaborate narrative machinery could be solved differently: by building the film around performers whose music was already integrated into who they were, whose songs didn't need motivation because the performer and the song were the same thing. You didn't need to explain why the Beatles sang. They were the Beatles.
This sounds obvious now, and in 1964 it was obvious - so obvious that nobody in Hollywood seems to have understood what it implied for the musical as a form.
Elvis Presley made thirty-one films between 1956 and 1969, most of them musicals of a sort, and almost all of them are terrible in ways that illuminate the same problem from the other direction. The Presley films tried to use his musical celebrity as a narrative premise - he's a racing driver who sings, he's a Hawaiian tour guide who sings, he's a crop duster who sings - and the results are films in which the performance of Elvis Presley singing is separated from the dramatic context by the clumsiness of the narrative machinery around it. The Presley films are a lesson in what happens when you have the right performer and the wrong approach: when you try to justify the music through plot rather than letting the music justify itself.
The right approach was being discovered elsewhere. The T.A.M.I. Show (1964), filmed in Santa Monica and featuring James Brown, the Rolling Stones, Marvin Gaye, and the Beach Boys, was not a Hollywood film in any conventional sense and barely registered as a cultural event at the time, but it pointed toward something: the concert film as a form that didn't need narrative justification, that could give you pure musical performance and make it cinematic through editing and camera placement rather than through story. James Brown performing "Please Please Please" at the T.A.M.I. Show is as purely cinematic a musical moment as anything in Singin' in the Rain - the body moving through space with a precision and an intensity that the camera had to work to follow - and it achieved this without any of the narrative apparatus the Hollywood musical considered necessary.
These parallel developments - the Beatles film and the concert film - were pointing toward a version of the musical that Hollywood wouldn't be able to make for another decade, because Hollywood was still trying to solve a problem that these films had dissolved by refusing to treat it as a problem.
Within the more conventional musical form, the most interesting work of the 1960s was happening in the films that took the form's conventions seriously enough to interrogate them. The Umbrellas of Cherbourg (1964) - French, but essential to the story - is the film that pushed the integrated musical to its logical conclusion by making every line of dialogue sung, without exception, and then used that total commitment to expose what the form had always been doing: creating a world in which emotion is so overwhelming that speech is insufficient, and discovering that a world that sings everything is not a utopia but a kind of beautiful desolation. The ending of The Umbrellas of Cherbourg, in which the two lovers who have been separated by circumstance meet again years later at a gas station in the snow and the music tells you everything their words are refusing to say, is one of the most devastating endings in the musical film and it works precisely because Demy has committed so completely to the form that the form's limitations become the film's emotional truth.
Funny Girl (1968) is the American film that comes closest to understanding what The Umbrellas of Cherbourg was doing, not formally - William Wyler directing a conventional backstage biopic - but in its willingness to end on failure. Barbra Streisand as Fanny Brice, and the film's final image of Streisand alone on stage singing "My Man" after Nicky Arnstein has gone to prison and the marriage is over, is the musical's utopian dimension turned inside out: the song survives, the voice survives, the world it was describing doesn't. Streisand won the Oscar for the role and Funny Girl represents the last moment where the traditional Broadway adaptation format produced a genuinely great film, because it found a performer so total that the conventional form could accommodate her.
The next year, the conventional form collapsed.
~
Part Five: The Crash
1969 is the year Hollywood lost its mind and the musical paid the largest portion of the bill. Hello, Dolly! lost $16 million. Paint Your Wagon lost $16 million. Sweet Charity, Bob Fosse's film debut, lost $10 million. Darling Lili was still being made and would lose more when it was finished. The studio system that had built the infrastructure for the Freed unit musicals - the rehearsal time, the sound stages, the months of preparation - was bankrupt, literally in some cases, and the independent producers and young directors who came to fill the vacuum had no interest in the form and no resources to sustain it even if they had.
Woodstock (1970) and Gimme Shelter (1970) are the punctuation marks on this period: two documentary films about rock concerts, one a celebration and one a catastrophe, both of them more cinematically alive than any studio musical made that year, both of them demonstrating that the energy of popular music in 1970 was as far from the Freed unit MGM musical as it was possible to get while still involving human beings and rhythm. The counterculture had produced its own musical forms, its own relationship between performer and audience, its own visual grammar - and the Hollywood musical had none of these things and couldn't acquire them without ceasing to be itself.
Hair had opened on Broadway in 1968 and the 1979 Milos Forman film is one of the more interesting failures in the musical's history - a director who had made One Flew Over the Cuckoo's Nest and Amadeus genuinely wrestling with the form's conventions and producing something with real energy that doesn't quite cohere. By 1979, Hair the stage show was already a period piece, and Forman's film has the melancholy quality of someone trying to capture something that has already passed. The hippie musical filmed by a Czech director in a post-Vietnam, post-Watergate moment is ten years late and it knows it.
The transition years of the early 1970s produced two films that point in different directions toward the musical's possible futures. Cabaret (1972) is the film that found a way to make the musical's conventions speak to the contemporary moment by treating the stage numbers as a theatrical space separate from the dramatic reality - Bob Fosse directing, Liza Minnelli performing, and the Kit Kat Klub stage as the place where the film's fascist nightmare is performed rather than disguised. Every number in Cabaret is performed on a stage, and every number comments on the dramatic reality surrounding it, and the relationship between performance and reality is the film's subject - which is also Fosse's subject, the person who spent his entire career thinking about what performance costs and what it conceals. Cabaret won eight Oscars and lost Best Picture to The Godfather, which is one of the more defensible Academy decisions of the decade. It is one of the greatest musicals ever made and it is the film that established the template for the serious musical: the form's utopian dimension acknowledged and then deliberately undermined, the songs beautiful and the world they're sung in terrible.
Nashville (1975) is the other film, and it's not really a musical in any conventional sense - it's an Altman film about the country music industry that uses twenty-four characters and their relationship with performance and celebrity to make a film about America - but it contains some of the most emotionally precise musical performance in any film of the decade. Keith Carradine's "I'm Easy" performed at a dinner club while four women in the audience each believe he's singing it to them specifically, and the camera moving among their faces as each one receives the same song as a private message, is a scene that understands what music does in the human emotional economy - how it functions as a vehicle for projection, for the listener's private meaning - in a way that no musical film that came before it had articulated so directly.
~
Part Six: Fosse and What He Understood
Bob Fosse made four film musicals: Cabaret (1972), Lenny (1974, which is not a musical), New York, New York (1977, which he didn't direct but deeply influenced through his relationship with Scorsese's use of the musical form), and All That Jazz (1979). Together they constitute the most complete statement about the musical in the late twentieth century, and the statement is essentially this: the form's utopian dimension is a lie that the form itself knows is a lie, and the most honest thing the musical can do is dramatize the cost of sustaining the lie.
Cabaret does this through history - the Kit Kat Klub's performed pleasure existing alongside the rise of fascism, the entertainment surviving as the world outside it becomes monstrous. All That Jazz does it through autobiography: Joe Gideon is Bob Fosse, a director and choreographer killing himself with work and amphetamines and cigarettes, and the film intercutsthe staging of a Broadway show with Gideon's hospitalizations and hallucinations and fantasies of death. The musical numbers in All That Jazz are simultaneously the most formally dazzling things in the film and the evidence of Gideon's self-destruction - he keeps putting beautiful things on stage while his body fails, and the relationship between creation and annihilation is not metaphorical. The finale, "Bye Bye Love" performed by everyone Gideon has ever known, is the musical form at its most complete and most bleak: the big production number as a goodbye to life, the spectacle as funeral.
Fosse understood something about the musical that the Freed unit had refused to understand: that the form's essential gesture - the moment when speech gives way to song, when ordinary movement gives way to dance - is not simply an expression of joy but an expression of need so extreme that ordinary means can't contain it. The Freed unit films used that transition as pure celebration. Fosse used it as diagnosis. The character who breaks into song in Cabaret does so because the world has become unbearable and the only available response is performance. The character who breaks into dance in All That Jazz does so because the alternative is confronting what the dance is costing him.
This is a darker understanding of the form, and it produced darker films, and it was the understanding that the 1970s' most serious musical filmmakers shared. New York, New York (1977), which Scorsese directed from a script by Earl Mac Rauch and Mardik Martin, is the film I keep returning to in this context - the note I made in 2022 that started this whole piece. It is a film that wears the visual language of the classic Hollywood musical - the studio sets, the studio lighting, the deliberately artificial production design - and uses that language to tell a story about a relationship that the musical's conventions would normally require to succeed and that instead falls apart. Jimmy Doyle (Robert De Niro) and Francine Evans (Liza Minnelli) are jazz musician and vocalist who love each other and cannot be together, and the film refuses to resolve this with a song or a reconciliation or a gesture toward hope. The final sequence, in which Minnelli performs "New York, New York" to a full house while De Niro watches from the audience and slips out before she can see him, uses the musical's climactic form - the star, the spotlight, the song - to enact a permanent separation. The song is the ending; the song is also the reason they can't be together. She has become what the song describes. He hasn't. They share a medium - jazz, music, performance - and it isn't enough.
The film failed commercially and has been slowly, incompletely rehabilitated over the forty-five years since. It's the most honest film the American musical produced in the 1970s about what the form could still do, and its commercial failure was part of what convinced Hollywood that ambitious musical filmmaking was too risky to attempt.
~
Part Seven: The Stage Musical Diverges
To understand what happened to the Hollywood musical from the 1980s onward, you have to understand what happened to the Broadway musical simultaneously, because the two forms had been in conversation since the beginning and that conversation now took a turn that changed both.
Stephen Sondheim is the central figure. Company (1970), Follies (1971), A Little Night Music (1973), Sweeney Todd (1979), Sunday in the Park with George (1984), Into the Woods (1987) - a body of work that took the integrated musical's premise about songs emerging from emotional situations and pushed it to a place where the emotions were ambivalence and regret and the failure of human connection, where the songs were too sophisticated and too uncomfortable for audiences who came to be moved in conventional ways, and where the theatrical experience was genuinely disturbing in ways that the commercial musical had never attempted. Sondheim's musicals are about people who can't have what they want and know why they can't and want it anyway. They end in death or compromise or the quiet defeat of understanding too late. They are some of the most important works of American art in the twentieth century.
Hollywood didn't know what to do with them for thirty years. The 1977 film of A Little Night Music, directed by Harold Prince who had directed the original stage production, is one of the most dispiriting failures in either medium: a film that transfers a stage production to screen without understanding what film and stage require differently, Elizabeth Taylor doing her best with material that requires a vocal instrument she didn't have. The Sondheim adaptation problem - how to make films from works whose sophistication is specifically theatrical, whose irony is specifically textual, whose music makes demands that movie stars can't meet - would not be seriously addressed until Sweeney Todd (2007), and I'll get there.
The Broadway musical in the 1980s fragmented in a different direction: the mega-musical. Andrew Lloyd Webber's Evita (1978), Cats (1981), Phantom of the Opera (1986), Claude-Michel Schönberg and Alain Boublil's Les Misérables (1980/87 London), Miss Saigon (1989) - enormous productions built around spectacle, around theatrical effects, around the experience of being overwhelmed rather than the experience of being moved. These shows worked on stage precisely because of the theatrical conditions: the live performance, the physical presence of the performers, the way the sets transformed before your eyes. The question of how to transfer them to film was the question Hollywood spent the 1990s and 2000s trying to answer.
~
Part Eight: The Eighties Wilderness and the Unexpected Detour
The 1980s produced almost no Hollywood musicals of consequence, which is itself a consequence worth examining. Annie (1982) was a commercial and artistic failure. A Chorus Line (1985) was a commercial and artistic failure. The Little Mermaid (1989) was a genuine triumph - the Disney animated musical finding a new format for the form's conventions - but it was not a live-action film and the success of the Disney Renaissance is a separate story from the live-action musical's continued difficulties.
What the 1980s produced instead was the music video, and the music video is essential to the story because it gave the musical's essential gesture - the body moving to music, the performer and the song unified - a new format and a new audience and a new visual grammar. Michael Jackson's "Thriller" (1983), directed by John Landis, is a fourteen-minute film with a narrative structure, dance sequences choreographed with the precision of a Busby Berkeley number, and an emotional arc from horror to comedy to exhilaration that the theatrical musical would have required ninety minutes and a Broadway budget to achieve. It was seen by more people than any film musical of the decade and it cost a fraction of the price.
The music video was doing what the concert film had done in the 1960s: proving that the musical's essential gesture didn't require the apparatus of narrative film to work. Prince's Purple Rain (1984) is the film that most fully absorbed this lesson - a semi-autobiographical star vehicle built around performance sequences that move between the concert stage and the dramatic story without the elaborate justification machinery the Hollywood musical had always required. Purple Rain doesn't need to explain why Prince sings. It builds a world in which Prince's relationship to music is the dramatic fact, and the songs illuminate the drama precisely because the drama is already about the songs. It's an impure film and a flawed film and it contains some of the most purely cinematic musical sequences of the decade.
Dirty Dancing (1987) is the other significant film of the decade and a more complicated case. A film about a young woman learning to dance and falling in love with her dance instructor, it uses the musical's basic architecture - the body's education, the romantic plot built around movement - without most of the musical's formal conventions. The songs are contemporary rather than integrated; the dancing is the point rather than an expression of dramatic emotion; the film is more romance than musical. But its success proved something important: that audiences would still come for films organized around dance and music if the films were emotionally direct enough and the dancing was good enough. Dirty Dancing made $214 million on a $6 million budget and remains one of the most efficient films ever made, in the sense of pure return on investment.
~
Part Nine: The Disney Solution and the Question It Answered
The Disney animated musical of 1989-1999 is the most commercially successful sustained run of musical filmmaking since the Freed unit, and it was successful for reasons that illuminate both what the form requires and why live-action has struggled to provide it.
The Little Mermaid (1989), Beauty and the Beast (1991), Aladdin (1992), The Lion King (1994), Pocahontas (1995), The Hunchback of Notre Dame (1996), Hercules (1997), Mulan (1998), Tarzan (1999) - Alan Menken composing for most of them, Howard Ashman writing lyrics for the first three (before his death from AIDS in 1991), and a studio that had reconnected with what the integrated musical required: songs that emerged from character rather than plot, that expressed what the characters needed rather than what the story needed to advance, that used the musical moment as a form of dramatic compression - three minutes of song that did the work of twenty minutes of dialogue.
"Part of Your World" in The Little Mermaid is the form's essential gesture as pure as it was in the Freed unit: Ariel singing about what she wants, the song expressing a desire that speech can't fully articulate, and the visual environment of the number - the grotto, the light, the human artifacts she's collected - doing the work of establishing character and world simultaneously. The song is the character. Howard Ashman learned this from Sondheim and from the integrated musical's long history, and what he understood - and what his successors at Disney have been trying to replicate since his death - is that the form works when the song is necessary rather than decorative, when the character would be incomplete without it.
The Disney animated musical solved the problem of the form's utopian dimension by using animation's inherent unreality as permission. In an animated film, characters can sing because the world they inhabit is already not the world - it operates by dream logic, by the logic of story, and the transition from speech to song is one transition among many. The problem of the live-action musical is precisely that the world feels real and the singing feels unreal, and the gap between the two is where audiences' skepticism lives. Animation closes the gap. The singing teapot and the dancing candelabra in Beauty and the Beast are unrealistic in exactly the same way as everything else in the film, so the unreality of the singing is absorbed rather than foregrounded.
This is not a permanent solution because it's not applicable to live-action. But the Disney Renaissance demonstrated something the live-action musical needed to hear: that the form's essential conventions - the integrated song, the dramatic dance sequence, the number that expresses what speech cannot - were not obsolete. They worked when the material was right and the craft was committed. The problem was not the form. The problem was live-action film's relationship with the form's conventions.
~
Part Ten: The 1990s Try Again
The 1990s live-action musical is a brief and mostly unsuccessful period of experimentation, with one significant exception and several instructive failures.
Evita (1996), directed by Alan Parker with Madonna as Eva Perón and Antonio Banderas as Che, is one of the more fascinating failures in the form's history - a film made with absolute technical competence that produces almost nothing in the way of emotional response. The problem is not Madonna, whose limited vocal instrument is used intelligently by Parker, and it's not the material, which is among Lloyd Webber's strongest. The problem is that Evita the stage show is built around theatrical distance - the story told by a narrator, the protagonist always slightly alien and slightly mythologized - and Parker's film makes the mistake of trying to close that distance, to make Eva human and comprehensible, which destroys the theatrical architecture the show requires. A film that tried to maintain the show's deliberate stylization might have worked. The realistic approach doesn't.
Dancer in the Dark (2000) - Lars von Trier's film with Björk - is the significant exception and the most radical thing the form produced in the period. Björk plays a Czech immigrant factory worker going blind who escapes into musical fantasy, and the film cuts between the Dogme-style handheld realism of her daily life and the elaborately staged musical numbers of her imagination, and the contrast is the film's subject: what music provides that reality can't, what imagination offers that the world refuses, and what happens when fantasy is the only available refuge from an unbearable situation. Von Trier uses the musical's utopian dimension not as a condition of the film but as his protagonist's coping mechanism, which is the most honest treatment of the form's escapist dimension since Singin' in the Rain ironized it forty-eight years earlier. Björk won Best Actress at Cannes. The film is genuinely disturbing and genuinely magnificent and almost no one working in Hollywood would have been permitted to make it.
Moulin Rouge! (2001) is the film that actually moved the needle on live-action musical filmmaking, and it did so by the most maximalist possible solution to the integration problem: refusing to integrate. Baz Luhrmann's film set in a Montmartre cabaret and built from pop songs repurposed as period romantic gestures - "Your Song," "Roxanne," "Come What May" - uses the jukebox musical format's anachronism as its own formal statement: these songs are already familiar to the audience, already emotionally loaded, and by putting them in the mouths of characters in 1900 Montmartre, Luhrmann argues that the emotions the songs express are not period-specific but universal, that love in 1900 had the same desperate quality as love expressed in any decade's pop music. It's a formally radical argument made through maximalist visual excess, and it works or doesn't work depending on whether you can surrender to the excess, which the film essentially requires of you in the first ten minutes.
What Moulin Rouge! proved was that contemporary audiences could engage with a film structured as a musical if the approach was declared as a stylistic choice rather than presented as naturalism. The problem of live-action singing - the suspension of disbelief required when a person in an otherwise realistic film suddenly begins to sing - could be solved by abandoning the realistic frame entirely, by making the artifice explicit and so complete that the singing becomes the world's condition rather than an interruption of it. This is the approach that La La Land (2016) would refine fifteen years later.
~
Part Eleven: The Sondheim Question, Finally Answered
Tim Burton's Sweeney Todd: The Demon Barber of Fleet Street (2007) is the most important Hollywood musical of the decade and the one that most directly addressed the Sondheim problem. Sondheim had been unadaptable to film for thirty years not because his shows were theatrical in ways that resisted translation but because Hollywood hadn't found a filmmaker whose sensibility aligned with Sondheim's in ways that would allow for genuine translation rather than mere transcription.
Burton and Sondheim are, it turns out, natural collaborators. Both are artists whose central subject is the grotesque and the beautiful occupying the same space without resolving, whose formal vocabulary is operatic rather than realistic, whose relationship with genre is always ironic and always committed simultaneously. Sweeney Todd required a director who would take the horror seriously and take the music seriously and not flinch from the film's demand that you find both beautiful, and Burton did exactly that. Johnny Depp's Sweeney is not a great singer in the technical sense, but his voice has a specific quality - contained, dark, controlled in the way a man controlling murderous rage would be controlled - that serves the material more honestly than a more technically accomplished voice might have. Helena Bonham Carter's Mrs. Lovett is the film's emotional center, a woman who loves a man incapable of loving her back, and the film's most devastating moment is her number "By the Sea," in which she fantasizes about the domestic life with Sweeney that she knows she will never have, and Bonham Carter performs the fantasy with a specificity that makes the delusion achingly visible.
Sweeney Todd restored something to the live-action musical: the possibility of genuine darkness. The Freed unit musicals and the Broadway adaptations of the 1950s and 1960s had kept the form essentially in the register of optimism or romance; even the troubled musicals of the 1970s were troubled about whether their optimism was justified. Sweeney Todd is a musical about serial murder in which the songs are where the real horror lives - where the characters articulate their rage and grief and desire more completely than the dialogue can, where Sondheim's language makes the audience hold the beauty and the violence in the same moment. It is the form doing something that only the form can do.
~
Part Twelve: The Contemporary Moment and What It Means
La La Land (2016) is the film that everybody has the most invested in as a referendum on the contemporary musical, and the debate it provoked - over its ending, over its politics of nostalgia, over whether its celebration of the musical form was sincere or ironic - is itself a form of evidence about what the form means now.
Damien Chazelle's film is a love letter to the Hollywood musical written by someone who knows the history and knows that loving it is complicated. Mia and Sebastian are dreamers in contemporary Los Angeles - she an actress, he a jazz musician - and the film uses the musical's conventions with a self-consciousness that is not irony but is not simple sincerity either: it's the sincerity of someone who understands why the conventions might seem naive and chooses them anyway, knowing what that choice costs. The opening number on the freeway is the form's utopian gesture stated as boldly as possible in the first five minutes: bodies dancing on car roofs in the Los Angeles sunshine, movement and music transforming a traffic jam into a celebration. The film earns the right to that gesture by spending the rest of its running time asking whether the gesture is sustainable.
The ending says it isn't. Mia and Sebastian succeed individually - she becomes a movie star, he opens his jazz club - and they don't end up together, and the fantasy sequence in which they imagine the life they might have had is the musical form's final statement: the song survives, the dream survives, the specific human reality of this particular love doesn't. It is the most sophisticated engagement with the form's utopian dimension since The Umbrellas of Cherbourg, and the fact that it was a massive commercial success suggests that the form has not exhausted its audience's patience but has, finally, found a way to address the contemporary relationship with it honestly.
The Greatest Showman (2017) is the other significant recent musical and the more purely commercial one, and I think it gets less credit than it deserves precisely because it doesn't have La La Land's self-consciousness. Hugh Jackman as P.T. Barnum, a story about a man who builds a spectacular entertainment out of people society has marginalized, and a film that uses the musical's utopian dimension without irony - that believes, unequivocally, that the spectacle of people performing together can be redemptive. It is an old-fashioned belief and an old-fashioned film and it connected with audiences in ways that more sophisticated musicals hadn't, which is its own argument about what the form still offers.
In the Heights (2021), Jon M. Chu adapting Lin-Manuel Miranda's early Broadway show, is the most recent entry that seems genuinely to have mastered the integration problem: songs that emerge from character, a community that sings because the community's emotional life is too full for speech, and a film whose visual grammar - shot on location in Washington Heights, the camera moving through the neighborhood with a specificity that the studio-bound musicals couldn't have - finds a way to make the contemporary reality and the musical's unreality inhabit the same space without apology. It underperformed at the box office during its pandemic release and deserves a second life.
Tick, Tick... Boom! (2021), also Miranda directing, is the Sondheim-adjacent film of the period - Jonathan Larson, before Rent, wrestling with what it means to make musicals in a world that doesn't value them, and Andrew Garfield giving a performance of musical ambition and self-doubt that is as personal as Fosse's All That Jazz in a different register. The film is about the desire to make musicals, which is a different subject than any musical before it has taken on directly, and its most devastating sequence - the "Sunday" number, a reference to Sondheim's Sunday in the Park with George - is both a tribute to the form and an honest account of what commitment to it costs.
~
Part Thirteen: What Actually Happened
The question I've been circling for three years, through a hundred films and a dozen books and more late nights than I can account for, is not answered simply. But I think I can answer it now.
The Hollywood musical didn't die. It dispersed.
The integrated musical's conventions - the song that emerges from emotion, the dance that expresses what movement cannot - didn't disappear when the Freed unit's infrastructure collapsed and the big Broadway adaptations bankrupted the studios. They migrated. They went into the music video and the concert film and the Disney animated feature and the pop star vehicle and the dance film. They went into Fosse's late work and Sondheim's stage shows and Demy's French cinema. They went into films that weren't labeled as musicals but were organized around the same relationship between music and emotion and the body - Nashville and Purple Rain and 8 Mile (2002) and Hustle & Flow (2005) and Inside Llewyn Davis (2013), films where music is the dramatic fact and the songs illuminate the dramatic situation rather than decorating it.
What happened to the Hollywood musical as a genre - as a recognized commercial category, as a reliable industrial product, as something studios made on a schedule and audiences expected on a regular basis - is that it became too expensive and too risky for the industrial conditions that replaced the studio system. The Freed unit worked because the studio system could absorb the risk of individual failures and amortize the cost of production infrastructure across many projects. The independent producer model that replaced it couldn't absorb those risks, and the Broadway adaptation failures of the late 1960s made the genre radioactive for a decade and a half.
But the form's essential gesture - the moment when the world becomes too much for speech and the body goes to music - is not something that human beings stop wanting or filmmakers stop finding ways to provide. It keeps appearing, in different containers, in different genres, in different industrial contexts. It appeared in Dirty Dancing and Purple Rain and Moulin Rouge! and Chicago and La La Land and In the Heights, films made across forty years in very different circumstances, all of which found some version of the integration the Freed unit had solved and then lost.
The question for the next decade is whether the form can reconstitute itself around a new industrial model - whether the streaming environment, which has different risk tolerances and different distribution logic than theatrical release, can support the kind of sustained musical filmmaking that the studio system supported at its peak. Hamilton (2020) as a filmed stage production distributed by Disney+ is one answer, a provisional and impure one. The adaptation of Sondheim's tick, tick... boom! on Netflix is another. These are not the same thing as the integrated Hollywood musical of the Freed unit, but they are the musical form finding its audience through the infrastructure that's available.
I don't know if the genre will reconstitute itself in anything like its classical form. I know that the desire it addresses - for music to organize emotion into something bearable, for the body to express what speech can't hold, for the world to become, briefly, the kind of place where the right song at the right moment makes everything make sense - is not a desire that human beings have stopped having. The form will keep finding ways to answer it. It always has.
The note I wrote in my margin in 2022 asked why New York, New York felt like a funeral. I know the answer now. It was the form mourning itself, which is what the form does when it's most honest about what it is: a beautiful lie told by people who know it's a lie and need it anyway.
New York, New York. It's a hell of a town.
~
This piece took three years and involved more films than I can count . If I've missed something essential, tell me in the comments. I probably missed something essential.
