Fools in Love

In Howard Hawks’s film Ball of Fire (1941), a group of bachelor scholars is writing an encyclopedia and working on the entries for the letter s. After an encounter with a loquacious garbageman (Allen Jenkins) at the Manhattan mansion that serves as the encyclopedists’ library and residence, Professor Bertram Potts (Gary Cooper), the literature expert, realizes he must rewrite his entry on slang: “Twenty-three pages compiled from a dozen reference books, eight hundred examples: everything from the idiotic combination absotively to the pejorative use of zigzag.” The garbageman “talked a living language”—gams, moola, boogie—while Potts has only “embalmed some dead phrases.”

S also stands for sex, about which Potts knows very little. During his search for the latest slang expressions, he meets the nightclub singer Sugarpuss O’Shea (Barbara Stanwyck), whose gangster boyfriend, Joe Lilac (Dana Andrews), is under investigation for murder. To evade the police, Sugarpuss decides to hide out with the scholars, most of whom have never spent much time with women, let alone one as sharp-witted and alluring as Barbara Stanwyck. Claiming she has nowhere else to go on a cold, rainy night, Sugarpuss pleads with them suggestively: “Look down my throat…. It’s as red as the Daily Worker and just as sore.” (The screenplay was written by Billy Wilder and Charles Brackett, masters at deflating ideology.) During her stay she leads the professors in a conga line around the library, and she and Potts eventually fall in love.

“Any intellectual endeavor with no pragmatic application is fair game for comedic ridicule,” the French film scholar Grégoire Halbout writes in Hollywood Screwball Comedy, 1934–1945, explaining the appeal of tweedy characters like Ball of Fire’s scholars. Is the attempt to define “screwball comedy” one such intellectual endeavor? Screwball comedy, like film noir, is a category invented by critics; there’s little evidence that Hollywood filmmakers self-consciously made screwball comedies the way they made, say, westerns or musicals. As a result, over the past half-century there’s been much debate over which romantic comedies from the 1930s and 1940s have a sufficiently high “ludicrous quotient,” as the film historian Wes Gehring puts it, to qualify as screwball.

Are screwball comedies fundamentally apolitical, as Gehring argues, or, as Thomas Schatz maintains in Hollywood Genres (1981), do their frequent interclass marriages carry a “sociosexual” political message? Does the love story have to be a battle of the sexes where the leads “make damn fools of themselves”—which Hawks claimed was the major innovation of his film Twentieth Century (1934), often cited as one of the earliest screwball comedies—or can it involve a happy couple in a bizarre situation, like the earnestly smitten fiancés of You Can’t Take It with You (1938)? How can you tell if something is screwball at its core or merely around the edges?

The hope is that in figuring out what we mean by “screwball comedy,” we might be better able to understand just what it is about these films that transports us. Halbout’s Hollywood Screwball Comedy, 1934–1945 is perhaps the most ambitious attempt to define the genre and catalog its examples. It is a dense academic work that’s full of jargon and that assumes readers know the difference between a typical MGM couple and a Warners Bros. one, but it’s carefully argued, contains a wealth of insight, and is refreshingly broad-minded.

Halbout resists as much as possible what he calls the “subjective appreciation and fetishism” of earlier critics whose overly rigid or idiosyncratic guidelines could lead to mincing conclusions. For example, Gehring and the film historian William K. Everson have argued that It Happened One Night (1934) isn’t quite zany enough to be a true screwball comedy, even though most writers view it as a cornerstone of the genre, and Everson bafflingly wrote of Ball of Fire, “As a comedy the film is always entertaining, often enchanting. As a screwball comedy, it is overrated.”

Halbout attempts not to settle such disagreements so much as render them irrelevant by presenting the genre as a spectrum. If a movie from the period Halbout covers has been called a screwball comedy by someone, he probably accepts that designation, even if others have contested it. In an appendix he lists 136 screwball films, many of them borderline cases, with forty grouped into what he calls a “primary corpus” containing the genre’s most famous examples along with delightful but less well-known movies like Midnight (1939), one of the best Cinderella stories Hollywood ever made, and The More the Merrier (1943), a romp about wartime housing shortages in Washington, D.C.

In view of the number of films the studios pumped out in the 1930s, there are bound to be omissions. On his longer list Halbout includes the slow and unsophisticated Annabel movies (1938), in which Lucille Ball plays a Hollywood starlet forced to undergo ever more embarrassing publicity stunts, but not Boy Meets Girl (1938)—starring James Cagney and Pat O’Brien as a mischievous pair of screenwriters based on Ben Hecht and Charles MacArthur—which has many screwball characteristics and may be the studio system’s most acerbic satire of itself. If the political content, foreign setting, and melodramatic aspects of Ninotchka (1939)—in which a Soviet envoy (Greta Garbo) falls in love with a Parisian gigolo (Melvyn Douglas)—don’t bother Halbout, why didn’t he also include The Baroness and the Butler (1938), a charming farce in which the butler to the Conservative prime minister of Hungary becomes the leader of the rival Social Progressive Party but insists on keeping his day job? In an interview last year with the critic A.S. Hamrah, Halbout said he should have knocked two or three films from the appendix, but he might consider adding others in subsequent editions.

In baseball a screwball is a pitch that moves in the opposite direction of a curveball; in film a screwball is a character who acts unpredictably. The term was first used by movie critics to describe Carole Lombard’s portrayal in My Man Godfrey (1936) of the airheaded heiress Irene Bullock, who thinks that being carried into a bathroom and forced under a running shower is a sign of love. The “screwball” label was in vogue from 1936 to 1938 and, according to the scholar Jane Greene, “often held a negative connotation” of “failed attempts to make up for a lack of clever dialogue and plotting.”¹ Reviewers who called Bringing Up Baby (1938) screwball did so in a spirit of exasperation at antics such as chasing leopards around the woods of Connecticut.

When the term was resurrected in later decades, it had become a mark of quality describing films—The Awful Truth (1937), His Girl Friday (1940), The Lady Eve (1941)—that felt fresher than other comedies from Hollywood’s golden age.² A number of common traits were agreed on: prominent romantic relationships treated unsentimentally and humorously, time of production (between 1934 and the early-to-mid-1940s), setting (contemporary, almost always urban and upper-class), defiant female leads, farcical plotlines, and a fast-paced mix of physical and verbal gags.

One of the strengths of Hollywood Screwball Comedy, 1934–1945 is that it clarifies the effects of censorship on the development of the genre. In 1934 the studios, prompted by backlash to Mae West’s salacious films, implemented the Motion Picture Production Code, which was enforced by the Production Code Administration (PCA). The PCA sought to minimize explicit treatment of sexual themes, and its censors scoured scripts for offensive language and gestures. The critic Andrew Sarris, who called the screwball genre “the sex comedy without sex,” imagined the dilemma filmmakers faced:

Here we have all these beautiful people with nothing to do. Let us invent some substitutes for sex. The wisecracks multiply beyond measure, and when the audiences tire of verbal sublimation, the performers do cartwheels and pratfalls and funny expressions.

Wit and physical confrontation became the chaste manifestations of desire.

The year 1934 is almost universally accepted as the start date for the genre, and while the Code was important to its rise, arguments linking the two tend to be oversimplistic. The PCA is often seen as hampering sophistication in Hollywood film, yet the movie industry had censored itself earlier. Ernst Lubitsch’s coy approach to self-censorship, which was closely studied and widely emulated, produced some of the pre-Code era’s most sophisticated movies. The difference between the treatment of sex in pre-Code and Code films is one of degree: pre-Code films could inch closer to taboo subjects, but even West had to rely on innuendo to get her point across. Despite the dates in Halbout’s title, he includes in his appendix two films from 1933, Three-Cornered Moon and Lubitsch’s Design for Living, as if to suggest the continuity in style between the pre-Code and Code periods. (In view of Halbout’s interest in borderline cases, he might have listed even more pre-Code movies.)

As for the slapstick, Halbout reminds us that many directors who worked in the screwball style had started in silent films. Physical comedy wasn’t merely a way to fill screen time when sexier material was forbidden; it was what they knew. (Sometimes from their personal lives: Leo McCarey—who helped establish Laurel and Hardy’s screen personas and later directed The Awful Truth—was particularly absentminded and clumsy. He once walked into an empty elevator shaft and broke both legs.)

Unlike Sarris, Halbout views the Code as having had mostly positive effects on the screwball genre. His point isn’t that censorship is good—he takes care to say that he’s not in favor of it—but that the threat of censorship tended to sharpen filmmakers’ wits and deepen their stories. The PCA and its bureaucratic procedures presented writers and directors with a challenge and a convenient target, a tiresome authority they had to placate but could also subvert. Halbout depicts the censors much as cops are treated in comedy: a looming, potentially catastrophic threat, but ultimately dim-witted and feckless.

The PCA, after all, didn’t have the power to prevent films from getting made; it worked for the studios, and its job was to ensure that Hollywood’s goods were protected from boycotts and government censorship. It helped writers and directors remove offending details from scripts while preserving their essence. Garson Kanin, the director of My Favorite Wife (1940), complimented the head of the PCA, Joseph Breen, on his “most courteous counsel” during preproduction: “I was rather surprised to find that your nose wasn’t blue. On the contrary, there may be a little too much red in it.” (That movie contains one of the Code era’s most positive depictions of an extramarital affair: Randolph Scott’s and Irene Dunne’s characters were castaways together on a desert island for seven years and refer to each other as Adam and Eve. They clearly had a lot of fun.)

It’s often impressive just how much sex filmmakers were able to get past the PCA, which focused on the sorts of silly things censors are usually concerned about: banned words (nuts to mean crazy, presumably because the word could also refer to testicles; lice, louse, and lousiest, for reasons that are unclear to me), the proximity of actors to beds and bathrooms. Halbout notes that writers put in jokes they knew would get cut in order to “divert the censor’s attention away from crucial plot points.” The PCA frequently missed crude, barely concealed references to genitalia, such as the pencil sharpener that Herbert Marshall’s lonely lawyer yearns for in The Good Fairy (1935), or the bone belonging to Cary Grant’s emasculated paleontologist in Bringing Up Baby that he and Katharine Hepburn’s character must find.

Halbout rightly points out that Ball of Fire, a movie about language, is one of the most glorious examples of the type of roundabout, suggestive speech characteristic of screwball movies: the use of slang creates “a sense of collusion between the audience and the characters…allowing film dialogues to circumvent the ban on explicit references to sex,” as when Jenkins’s garbageman tells the befuddled encyclopedists that he’s going to meet a “dish” for “some hoi-toi-toi.” To drive his point home (and give a knowing wink to viewers), he adds, “If you want that one explained, you go ask your papas.”

In many cases, all that’s needed to get across sexual meaning is a raised eyebrow or a sidelong glance. It isn’t especially difficult to discern that, for example, in Theodora Goes Wild (1936) Melvyn Douglas’s rakish illustrator steps up his efforts to sleep with Dunne’s best-selling romance novelist when he realizes that she’s a virgin; or that in Mr. and Mrs. Smith (1941) Robert Montgomery’s character is aroused by the possibility of having what would technically be premarital sex with his wife of many years (Lombard) after they discover that their marriage license is invalid; or that in My Favorite Wife Dunne’s character wants to punish her estranged husband (Grant) before their reconciliation by making him unbearably horny. As the film scholar Leger Grindon wrote, “Censorship fostered the artful implication that allowed the innocent to suspect nothing but provoked the imagination of the experienced.”³

Screwball stars, Halbout writes, “were a new species of clown, whose bodies were put in service of the purest slapstick tradition while their faces had to adapt to the subtlety of the dialogue and the tempo of back-and-forth delivery.” One of the best of these new clowns, William Powell, was a Hollywood veteran who had been typecast as a heavy in silent films and as a seedy seducer in early talkies. Like many other actors—Grant, Douglas, Lombard, James Stewart, Jean Arthur, Rosalind Russell—he struggled to attain stardom before becoming a screwball specialist. It was in the detective comedy The Thin Man (1934) that Powell first showed his genius for verbal and facial acrobatics. “Powell is to dialogue as Fred Astaire is to dance,” Roger Ebert wrote. His elegance in speech comes in part from always seeming like he’s about to trip over his words. Improbably, he turns stammering—ee, er, oh, um, ah—into lilting music.

The Thin Man’s wealthy, hard-drinking couple, Nick and Nora Charles (Powell and Myrna Loy, whose career had also been held back by typecasting), formed a pattern for later screwball couples. Even the Charleses’ wire fox terrier, Asta (played by a canine actor named Skippy), became a frequent presence in screwball comedies. The Charleses are “almost always playing, and they’re equals on top of it all,” Rob Kozlowski writes in Becoming Nick and Nora, a brisk dual biography of the stars. Their repartee is impeccable:

NICK: I was shot twice in the Tribune.
NORA: I read you were shot five times in the tabloids.
NICK: It’s not true. He didn’t come anywhere near my tabloids.

Kozlowski calls their relationship “the friendliest, most fun marriage ever captured on screen,” at the core of which is a lot of good-natured teasing. Loy frequently plays the straight woman, setting up Powell’s jokes, but she’s rarely overshadowed by him; she’s refined and dignified but impish, always eager to be led astray.

Halbout found about thirty detective screwball comedies that repeated the formula of The Thin Man, mixing romance, farce, and crime. The best of them include The Ex-Mrs. Bradford (1936), in which a detective novelist (Arthur) pressures her former husband (Powell) into solving a murder, and It’s a Wonderful World (1939), about a private detective (Stewart) who cracks a case with the help of a “famed poetess” (Claudette Colbert). Solving mysteries together, it turned out, made for happy pairs: a murder investigation, Halbout writes, “becomes a pretext for having fun…and getting sloshed.”

MGM milked the success of The Thin Man by casting its leads in five sequels as well as other excellent screwball comedies: Libeled Lady (1936, costarring Spencer Tracy and Jean Harlow), Double Wedding (1937), I Love You Again (1940), and Love Crazy (1941). Kozlowski notes that in these films

neither Powell nor Loy was ever already attached to anyone else…perhaps because even in an age when implausible narratives were the norm, the idea that either William Powell or Myrna Loy could ever be in love with anyone else…wasn’t even worth considering.

In 1941 the New York Times critic Bosley Crowther described the Powell-Loy films as a “steady progression toward insanity.” Powell’s character in Love Crazy fakes a mental illness to delay the divorce proceedings that Loy’s character brings after a misunderstanding. He escapes the asylum, shaves his mustache, and pretends to be his own sister until the plot literally unravels: one of the balls of yarn he uses as fake breasts gives him away.

I Love You Again, a rare small-town screwball comedy, is an outlandish tale of amnesia and Middle American ennui. Powell’s character gets knocked on the head and upon recovering finds that he can’t remember anything that happened since an earlier knock on the head during a robbery nine years before. A former con man, he learns that he’s been living in Habersville, Pennsylvania, as a businessman, civic leader, taxidermy enthusiast, and volunteer Boy Scout leader. He’s horrified by the squareness of this existence and also horrified that the one good thing about it, a wife played by Loy, is in the midst of divorcing the sexless husband she’s grown to despise. Powell must pretend to be upright and boring to the townspeople, against whom he plots a brazen swindle, while convincing Loy that he’s actually the charming, goofy man we all know Powell to be. “One of the most remarkable accomplishments of this film is how much it works even when the entire plot completely falls apart,” Kozlowski writes.

The Code’s disapproval of positive portrayals of divorce and adultery forced filmmakers to devise situations in which they could imply wedded couples’ sexual dissatisfaction and extramarital affairs. A technicality is discovered that renders a marriage license invalid, a couple dates other characters while waiting for a divorce to come through, or a divorced couple realizes they made a mistake; by the end of the film, husband and wife reunite, each better able to please the other. To quote Walter Burns (Grant) in His Girl Friday, “Divorce doesn’t mean anything nowadays…. Just a few words mumbled over you by a judge.”

The phrase “comedy of remarriage,” popularized by the philosopher Stanley Cavell in his influential book Pursuits of Happiness (1981), has become virtually synonymous with “screwball comedy,” even though he later complained that there was little overlap between the two categories. Halbout points out that remarriage plots long predate the Code—in the 1910s Cecil B. DeMille made both remarriage and divorce comedies—and that only about a third of the screwball films he’s identified contain remarriage plots. (The rest feature what he calls “new love.”) Nevertheless, Halbout counts Cavell as a major influence, and he thinks that Cavell’s view of remarriage as a prolonged, playful negotiation for mutual recognition can be applied to screwball relationships generally.

But Halbout goes further, constructing an elaborate account of what screwball romances have to say about marriage and society between the Great Depression and World War II. He argues that in early-twentieth-century America, the normalization of divorce (whose “existence and legitimacy are never disputed” in films despite the Code’s restrictions) and the growing emphasis on sexual fulfillment gave female characters the power to set the terms of their marriages and win greater freedom for themselves. Moreover, “the world built by men had been discredited” by the cataclysm of the Depression, and women had to help repair the damage, leading to “a reshuffling of classic male and female roles.” (The PCA had something to do with this shift, too, as Molly Haskell explained in her 1974 study From Reverence to Rape: “The proscriptions of the Production Code that were catastrophic to sexually defined, negligee-wearing glamour goddesses were liberating for active or professional women.”)

More and more in film, women worked in offices and retail stores, literally wore pants, and threw punches. Female stars brought new intelligence to comic roles, making women’s struggles visible at this time of transition and backlash. Jean Arthur conveyed a frightening ability to see sordidness and stupidity everywhere, which her characters covered up with tough talk or nervous patter. Beneath Irene Dunne’s huge smile and subtle seduction were fierce determination and a love of mischief. Even the “adventuresses” who appear in screwball comedies seem more substantial than their predecessors: in movies like Midnight and The Palm Beach Story (1942), Claudette Colbert “does for golddigging what Lombard does for craziness: she makes it seem like something liberating,” the critic James Harvey wrote in his excellent survey Romantic Comedy in Hollywood, from Lubitsch to Sturges (1987).

The men, by contrast, were often hapless and out of tune with daily life. They’re eggheads who seem afraid of or indifferent to women—Cooper’s Professor Potts, Grant’s David Huxley in Bringing Up Baby, Henry Fonda’s snake-obsessed Charles Pike in The Lady Eve. When the male characters in screwball comedies are more traditionally masculine—Clark Gable in It Happened One Night, Fred MacMurray in any number of roles—they’re diminished by pratfalls and cutting remarks from their female counterparts.

Halbout claims that screwball romances point the way toward a more democratic society with greater gender and economic equality: they bridge class divides, humble the proud, and give the timid courage to fight for what they deserve and to overcome the obstacles put up by disapproving snobs and elders. He calls a couple “the smallest possible democratic unit,” and screwball couples, with their inclusiveness and “logic of mutual consent,” offer a “model for society”—nothing less than a “Screwball New Deal” to aid “the construction and defense of America’s democratic regime.” If you’ve ever wondered why newspapers in old comedies are so concerned with the love lives of their main characters, his account of the genre offers an explanation: the union of the central pair, with its implications for civic and economic life, is a cause for mass celebration, and so “there must be media coverage to convey the news” to the wider public.

Cavell wrote that every genre has a myth that it retells again and again, and Halbout’s myth for the screwball genre—that romantic love can lead to economic and political regeneration—is a pleasing one, but it’s overwrought and overdetermined. He argues, for example, that father figures in screwball comedies can be linked to long-standing debates about “federal power” versus “local autonomy,” or to the economic recovery programs of the Roosevelt administration, as if the senex weren’t a stock character in any number of earlier comic traditions. It would be more accurate to say that certain comic conventions were well suited to 1930s America, rather than to suggest that the tropes and gags in screwball comedies arose out of conditions unique to that era.

It’s also a stretch to make screwball couples “pure hearts” who “triumph” over a corrupt gerontocracy. As Halbout seems to acknowledge elsewhere in the book, sometimes they’re less like civic exemplars than like Zeus and Hera in the Iliad, squabbling deities whose whims cause trouble for mere mortals. In contrast to Halbout’s hopeful interpretation, the film scholar Ed Sikov called screwball movies “a comic glorification” of the kernel of hatred that sometimes lies at the center of love, and of “the ugly little annoyances that bubble up” in relationships.

Many screwball movies end with the couple’s reconciliation, but in others—such as The Moon’s Our Home (1936), which cheerfully concludes with Fonda’s pompous travel writer putting Margaret Sullavan’s imperious movie star in a straitjacket—they merely accept that they’re stuck with each other. No one would take the warring pair at the center of, say, Twentieth Century—the scheming, self-pitying, hammy theater impresario Oscar Jaffe (John Barrymore) and the equally grandiose actress Lily Garland (Lombard)—as the basis for a utopian social program, yet no one would deny that they’re screwballs.

Halbout attributes the fall of the screwball genre to World War II: movie producers sensed that anxious audiences had less patience for farce, linguistic misdirection, and feminized male leads. In the late 1940s studio revenues declined, and filmmakers had less freedom than when the studios were run more like factories. The pace of comedy slackened as its style got broader and more overtly sexual. There were many bad remakes of screwball classics, as well as some successful attempts to channel their spirit, but what Pauline Kael called the pre-war studio system’s “steady flow of bright comedy” became a muddy trickle.

Preston Sturges, the first credited writer-director in the studio system’s sound era, was the last filmmaker to make significant contributions to the screwball style. In The Lady Eve, a film full of references to Eden, he elevated the pratfall into a symbol for the fall from innocence; the bratty heir Charles Pike is disabused of his priggish morality by the cardsharp Jean Harrington (Stanwyck), who, after he rejects her, pretends to be the titular British aristocrat. Fonda’s tumbles may be the most artful use of physical comedy in any screwball film.

But Sturges’s true genius was for dialogue, which stood out in an era of word-drunk films thanks to his ability to capture the symphony of American speech. “I spritz dialogue like seltzer water,” he told Darryl Zanuck. Jean’s father, Harry (Charles Coburn), a gentleman crook, convincingly commands both high and low registers. Here they are on the deck of an ocean liner, watching Pike board after he’s been “up the Amazon for a year” studying snakes:

JEAN: Gee, I hope he’s rich. I hope he thinks he’s a wizard at cards.
HARRY: From your lips to the ear of the Almighty.
JEAN: And I hope he’s got a big fat wife so I don’t have to dance in the moonlight with him. I don’t know why it is, but a sucker always steps on your feet.
HARRY: A mug is a mug in everything.

And here they are after Jean tells Harry, who plans to take Charles for tens of thousands of dollars at the card table, that Charles is falling in love with her:

HARRY: Of course he’s in love with you. Who is he not to be in love with you who beautify the North Atlantic?…
JEAN: No, I mean on the level, Harry.
HARRY: Are you suggesting that the others were on the bias?
JEAN: Oh, stop kidding…. You don’t get the point. I like him too.
HARRY: Why shouldn’t you like him? There’s as fine a specimen of the Sucker sapiens as I’ve ever seen.

Sturges was a natural farceur, and he proved that the Code’s attempt to protect the sanctity of marriage instead made a mockery of it. (The PCA nixed his preferred title for The Palm Beach Story: Is Marriage Necessary?) In The Miracle of Morgan’s Creek (1944) he took on the censors as well as the war effort, constructing a parody of the Virgin Birth with the Holy Ghost taking the form of a GI, possibly named Ratzkiwatzki, who marries and impregnates Trudy Kockenlocker (Betty Hutton) during a wild night just before he ships out. Joseph is her long-suffering admirer, Norval (Eddie Bracken), emasculated by Trudy and by the army, which won’t let him serve because of minor (and likely psychosomatic) vision trouble. The Christ Child multiplies into sextuplets, all male, who soon become a propaganda tool against the Axis. (“HITLER DEMANDS RECOUNT,” one headline reads.) Presumably the PCA approved the blasphemous script because the sextuplets were conceived in wedlock.

The film ends by quoting Twelfth Night (incidentally one of the ancestors of the screwball genre): “Some are born great, some achieve greatness, and some have greatness thrust upon them.” But if it’s Ratzkiwatzki—or was it Zitzkiwitzki?—doing the thrusting, then is greatness what Trudy and Norval get?

Sturges could do so much with the screwball comedy in part because his life had practically been one. From childhood till death, he bounced between shabby gentility and astounding wealth. His mother was the kind of eccentric grande dame 1930s Hollywood loved: a footloose bohemian who was best friends with the dancer Isadora Duncan, she started a cosmetics business that Preston was at times forced to operate. (He developed a no-smear lipstick called Red Red Rouge.) His marriage to his second wife, Eleanor Post Hutton (the daughter of Marjorie Merriweather Post and E.F. Hutton, who together built Mar-a-Lago⁴), was front-page news. He found out that he’d unwittingly committed bigamy with Hutton because the Mexican divorce secured by his first wife, Estelle Godfrey (née de Wolfe Mudge), turned out to be invalid.

“One of his greatest gifts was his knack for infusing old types with some of the messiness and circumstantiality of lived experience,” Stuart Klawans writes in Crooked, But Never Common, an incisive collection of essays on Sturges’s major films. Nabokov noted that Gogol “has a peculiar manner of letting ‘secondary’ dream characters”—which Nabokov called “homunculi”—“pop out…to flaunt for a second their life-like existence.” Sturges’s homunculi have a similar uncanniness.⁵ They were inspired by characters from earlier films—all those spinster aunts and tough cabbies and simpering salesmen who added color to studio fare—yet Sturges gave them new vitality, creating the parts with particular actors in mind. He developed a sort of despotic identification with the members of his “stock company,” feeling that his success depended on these actors with funny faces—brash William Demarest, feeble Jimmy Conlin, gravel-voiced Frank Moran—and he treated them as good-luck charms.

Klawans offers fresh analyses of Sturges’s movies, an impressive accomplishment considering how much has already been written about them. Like Halbout, he cites Cavell as a major influence, and Crooked, But Never Common is much closer in spirit and style to Pursuits of Happiness. Klawans thinks through Sturges’s work with love and careful attention, and he matches Cavell’s ability to ferret out previously unidentified allusions, as when Klawans suggests that Jean’s transformation into Eve may have been influenced by the myth of Lilith, Adam’s first wife, who was banished from Eden and who was in vogue in literary circles in the 1920s.

More than thirty years ago Geoffrey O’Brien marveled in these pages at the pandemonium in the Film Forum lobby during a Sturges retrospective, with “young and fashionable crowds spilling over into the street.”⁶ While I’ve never seen comparable crowds at more recent screenings of his films there, I have met many people under thirty-five who count The Lady Eve as their favorite movie. “However much you delve into [Sturges’s films] to understand what they mean and how they work,” Klawans concludes, “these movies in their wholeness behave as if they were alive.” Crooked, But Never Common shows us just how much more there is to discover in them.