Andrew Sarris wrote of Elia Kazan in The American Cinema that "his career as a whole reflects an unending struggle between a stable camera and a jittery one." Historically that's more or less been the rap on Kazan—a highly-acclaimed filmmaker with many strong titles, but one whose work was too simultaneously bland and conflicted for the critical establishment to elevate him to auteur. The son of Greek immigrants and eventually a famed Broadway director, Kazan began filmmaking with a group-directed short called People of the Cumberland, broke into feature directing with 1945's adaptation of Betty Smith's A Tree Grows in Brooklyn and left it 18 films later with a version of F. Scott Fitzgerald's The Last Tycoon. He came close to greatness on film, though rarely reached it: At his peak period he was at the high end of the middle bracket of several frankly liberal directors, many of whom had crossed over into movies from film and TV. He's lighter and earthier than the leaden, sententious cinema of Stanley Kramer and Richard Brooks, though he never achieves the pure ecstasy and reverie of the best Nicholas Ray.
In the 1950s Kazan made many enemies for naming names before the House Un-American Activities Committee (he then filmed On the Waterfront, his best-known work, in apology). The current generation of cinephiles also disdains him for his beliefs, but more their square-jawed sincerity than for any hurt feelings: We live in the age of irony, after all, and don't need movies to tell us that anti-Semitism is bad (Gentlemen's Agreement), that racism is worse (Pinky), that the media's out to get us (A Face in the Crowd), that family is important (A Tree Grows in Brooklyn), or that you should always do what you know is right (On the Waterfront). Yet, as Film Forum's recent Kazan retrospective showed, rather than raise an eyebrow or let burst a guffaw it's worth marveling at all the not just strong but terrific movies that Kazan actually made: Panic in the Streets, A Streetcar Named Desire, On the Waterfront, East of Eden, Baby Doll, Wild River. (Viva Zapata!, A Face in the Crowd, America America, and Splendor in the Grass all also have staunch supporters, the films' sluggish sentimentality be damned.) His claustrophobic compositions and squeezed, stumbling editing—a typical Kazan sequence shows people entering a room, followed by multiple back-and-forth close-ups as their words build to an argument, then a sudden cut to a long shot as they physically attack each other, followed by people entering another room—tend to suffocate me. Yet how is it a bad film director made so many good films?
The answer, I think, has to do with arrangement: Kazan had a superior ability for assembling talented collaborators, even if he often failed to integrate them into a focused, coherent whole. This applied to his offscreen fellow artists—Boris Kaufman's wintry On the Waterfront docks photography makes you want to watch with your coat on—but was especially true of his actors. Kazan directed 21 different actors to Oscar nominations, with nine winning. One could rightly respond that the Oscars are wrong, constantly, but would still have to concede career-best performances from James Dean, Carroll Baker, Vivien Leigh (Blanche DuBois bests Scarlett O'Hara), Karl Malden, Celeste Holm, Lee Remick, and above all, Marlon Brando. It's always a tricky question how much of a performance is due to the director and how much is due to the actor—in Kazan's case, frankly, I haven't read enough to judge. In his book Movie Love in the Fifties, James Harvey quotes Kazan refusing credit for Brando's work ("it was like directing some genius animal"). But Kazan was known for researching his character backgrounds exhaustively, and his notes for the original Broadway production of A Streetcar in Desire, in which he analyzes each character based on close reading, is vital for dramatists today. Yet as the bad, grotesquely hammy turns in his movies prove (Andy Griffith in A Face in the Crowd, Lee J. Cobb in On the Waterfront), the actor still needs to nail the part. Perhaps the bottom line is that actors always perform in collaboration. On A Streetcar Named Desire and On the Waterfront in particular, Kazan and Brando collaborated to create two of the most indelible performances in film history.
Many critics have spouted about how Brando brought the Method, a more overtly psychological, raw emotive style of acting than previous modes, to the screen (untrue—Montgomery Clift appeared in Red River two years before Brando's film debut in The Men), but I suspect that what wows most viewers even today is the actor's sheer physical energy. Oftentimes the physicality is explicitly sexual—while in the 1930s and 1940s Cary Grant and James Stewart walked alongside their leading ladies, hands firmly in pockets, in the 1950s Brando constantly violated their personal space. He often did so tenderly—picking the lint off of Kim Hunter's sweater in A Streetcar Named Desire, slipping Eva Marie Saint's glove onto his hand in On the Waterfront—so that throwing women onto beds or breaking their doors down later would seem all the tougher. The physical threat of sex wasn't new to Hollywood movies (a peek at Barbara Stanwyck stroking Henry Fonda's hair in The Lady Eve shows how screwball comedy thrived on it), but Kazan probably went further than any A-list director in making assaults a key part of his drama.
A good thing he did, since the wrestling matches are the most exciting parts of his movies: Carroll Baker paddling her husband's neck flab in Baby Doll, or James Dean throwing his brother at their mother in East of Eden, say far more about characters' relationships than the film's overwritten scripts do. The best moments in Kazan's films are inevitably full two-shots, bespeaking his theatrical training. Unlike the work of the great film stylists (Murnau, say, or Von Sternberg, or even a theaterphile like Renoir), we watch Kazan not for the shots but for the struggles in them. The acting style he favored doesn't work in abstraction—the actors need something concrete to work against. Finding fruitful ways for people to punch, block, and counterpunch each other can be difficult and exhausting, and in his movies it doesn't always come off.
Kazan faced an even greater, more basic challenge in staging his bouts—to entertain, the fighters had to be equally matched. One actor needed to be strong enough to assert what he or she wanted, and the other strong enough to deny it, for the scene to crackle. Skilled as he was at eliciting good performances, Kazan didn't consistently get good connecting performances. Three movies from throughout Kazan's career—a failed early film, a failed late film, and a successful middle film—show both how important the lead couple's chemistry is to his movies and why it doesn't always succeed.
1947's Gentlemen's Agreement, Kazan's fourth film, catapulted him upwards, winning Best Picture and Best Director awards at both the Golden Globes and the Oscars (as On the Waterfront would seven years later). The movie concerns a Gentile reporter (Gregory Peck) who goes undercover as Jewish to investigate anti-Semitism for a magazine story, and the strain this causes for his country-club girlfriend (Dorothy McGuire). Never mind that an actual Jewish character doesn't appear until 40 minutes in, or that the anti-Semites are almost all grinning walk-ons: The movie's chief fault isn't with its peripheral parts, but with its central relationship.
McGuire is a competently concerned-looking actress, given to wringing her hands and widening her eyes, but Peck is catatonic. Unlike the murder Peck's slow deliveries and stiff posture would commit, six years later, on what might have been a nice comedy called Roman Holiday (I dare anyone who likes Roman Holiday to substitute Cary Grant and Howard Hawks for Peck and William Wyler), his Gentlemen's Agreement casting is in a way useful to the movie's purposes—if you're going to have a black-and-white message, it's good to have the least morally ambiguous actor possible delivering it (cf., a decade later, Atticus Finch). At the same time, Peck's robotic rectitude drains the movie of human drama; when he learns late in the film that his mother's fallen sick, Peck delivers the line "A stroke" as he might "Coffee please." When McGuire grabs his arm to talk him out of the job, he just stares steely-eyed, not touching her back, no hint of sex whatsoever. The risk he takes by accepting the assignment stays purely intellectual because we don't sense he has anything physical to lose.
In The Last Tycoon the roles are reversed—here it's the man longing, and the woman out of reach. Robert De Niro plays a 1920s Hollywood studio head who lives only for pictures: As he says at one point, "They're my life." He sits at his desk. His hands idle. He stares out a window. His most animated moment comes when he's acting out a scenario for a writer: His step quickens, his hands race, his eyes laser in. As in many other roles, De Niro projects a brick wall, but here vulnerability and need shine through the chinks. J. Hoberman has called it De Niro's best performance in a non-Scorsese movie, and though I'm hesitant to agree, he has a point.
The movie flops because De Niro has no other person on whom to focus his energy. Despite a wealth of good supporting actors—The Last Tycoon also features Robert Mitchum, Tony Curtis, Jeanne Moreau, Jack Nicholson, Dana Andrews, and Donald Pleasance in small parts—the lead actress is an ethereal blond wisp named Ingrid Boulting, whose subsequent career highlight would be a 1980s thriller called Deadly Passion. Boulting, who plays a mysterious woman about to leave town to get married, is in less than half the movie, and when she appears tends to stand apart from De Niro, at the other side of the frame. Indeed, the film's very point is that her character is an unattainable ideal, the screen heroine De Niro searches for in his everyday life. Yet Kazan fails to move his hero's longing out of inertia. The movie begins with De Niro wanting, and ends with him still wanting. Kazan, whose motto may as well have been "No ideas but in things," thus has nothing specific to focus the movie on.
That's not the case with 1960's Wild River, Kazan's fourteenth film and one of the great underappreciated heterosexual screen romances (it's unavailable in the States on DVD). Montgomery Clift plays a 1930s Tennessee Valley Authority rep who comes to a small town to buy out a family's home so the TVA can build a dam. The family lives on an island that he has to row to, and as he's pulling away after a visit one of the group's young women (Lee Remick) leaps onto his raft. He stares at her, amazed, and she explains hurriedly: She barely ever leaves, and she's lonely.
Kazan contrasts performers to great effect. Remick was 25 and healthy, with a round, full face and deep water-blue eyes. Clift was 40 and angular; he'd been in a car crash four years earlier that left his lips pressed tightly together after facial reconstruction, his eyes wide and nervous, looking traumatized (an effect Stanley Kramer would vulgarly exploit by casting Clift as a Holocaust victim in the following year's Judgment at Nuremberg). The movie amazingly gives us none of the Clift character's back story, yet we still sense that we know all that there is to know about him. He's young, business-minded, and given to suit-wearing. "You're a hard man to love," the Remick character says.
She hopes he'll take her away, he hopes she'll marry him; as Arlene Croce wrote of Fred Astaire and Ginger Rogers, he gives her class, she gives him sex. But Wild River's tension is dramatized as being between an open person and a closed person. We frequently see Remick staring ahead, leaning forward, her mouth open, while Clift stands back, a statue doing his best not to crumble (I can't help but think of Clift's real-life homosexuality helping his unease). Unlike in both Gentlemen's Agreement and The Last Tycoon, where who was in the shot when seemed a matter of utility, here Kazan cuts back and forth between the pair early on, uniting them in the frame more frequently as the film progresses. A key image shows Remick resting her head on his shoulder. Her arms encircle his waist below the shot. His own arms are crossed, and his back is to her. His head is caught between his chest and his back, uncertain whether to look at her or look away. Unlike with Peck-McGuire, we sense that the people truly desire each other, and unlike with De Niro-Boulting, the conflict comes from their drawing towards each other despite their best attempts to pull away. Clift wants to stay uninvolved, above the locals' troubles, but Remick brings him down to Earth, literally. When she wraps her arms around him in another scene, he surrenders to her by collapsing to the floor.
In all three of these films, Kazan casts assertive supporting players as foils to his ambivalent leads. Because the central relationships don't sparkle, both the liveliness of John Garfield and Celeste Holm in Gentlemen's Agreement and the certainty of Theresa Russell and Jack Nicholson in The Last Tycoon end up overwhelming the stars. There is such a thing as being too ambivalent, to the point where Peck-McGuire and De Niro-Boulting evaporate into air. By contrast, though far from a perfect film (its treatment of race relations feels overly simplistic), Wild River finds the right balance. Clift and Remick's needs, wants, and fears are so well established that the convictions of Jo Van Fleet and Albert Salmi underline their struggles rather than distract from them. It's tough to say how much of a hand Kazan had in shaping the two lead performances, but Wild River ends up being his most successful movie—better than A Streetcar Named Desire, or On the Waterfront, or East of Eden—largely because it's the purest presentation of a situation he excelled at presenting: Two people who can only get what they want from each other.
Aaron Cutler has written about film for Slant Magazine and The Believer. He is working on a book about New York's repertory cinemas.