The Problem of the Crime Film

The crime film is the most enduringly popular of all Hollywood
genres, the only kind of film that has never once been out of
fashion since the dawn of the sound era seventy years ago. It
is therefore surprising to discover that, at least as far as academic criticism
is concerned, no such genre exists. Carlos Clarens’s magisterial
study Crime Movies (1980) begins by criticizing Robert Warshow’s
seminal essay “The Gangster as Tragic Hero” (1948) for its narrow definition
of the gangster film, based on liberal social assumptions that
“limited genres to one dimension apiece.” Yet Clarens’s definition of
the crime film is equally delimited by its pointed exclusion of “psychological
thriller[s]” like Shadow of a Doubt (1943), Laura (1944), and Kiss
Me Deadly (1955) from its purview on the grounds that their characters
are insufficiently emblematic of “the Criminal, the Law, and Society.”
1 Larry Langman and Daniel Finn place themselves outside the debate
over whether or not crime films include psychological thrillers
by announcing in the Preface to their encyclopedic reference, A Guide
to American Crime Films of the Forties and Fifties: “The American crime
film does not belong to any genre. . . . Instead, it embodies many
genres.”2 But their attempt to rise above the problem of classification
merely indicates how deeply entrenched that problem is.
None of this academic quibbling has prevented crime films from retaining
their popularity, or even from entering universities as the object
of closer scrutiny. But subgenres of the crime film, like the gangster
film of the 1930s and the film noir of the 1940s, have been more
often, and more successfully, theorized than the forbiddingly broad
genre of the crime film itself – this genre that is not a genre, even

though an enormous audience recognizes and enjoys it, and a substantial
following is interested in analyzing it critically. The unabated
popularity of mystery and detective fiction, the burgeoning of such recent
literary subgenres as the serial-killer novel and the novel of legal
intrigue, the efflorescence of true-crime books, and the well-publicized
criminal trials that keep Court TV in business all attest to the American
public’s fascination with narratives of crime. The crime film therefore
represents an enormously promising, but hitherto neglected, focus
for a genre approach to cultural studies.
To the question of whether the crime film is a genre or an umbrella
term for a collection of diverse genres like the gangster film, the detective
film, and the police film must be added another question: What
does it matter? After all, what difference does it make whether the film
noir is a genre or a subheading of a broader genre? To anyone but a
few scholars of genre studies, these questions might seem inconsequential
to the widespread understanding and enjoyment of crime
films.
It is exactly this understanding and enjoyment, though, that are at
issue in the definition of any genre. Raymond Bellour has pointed out
that viewers for Hollywood musicals like Gigi (1958) are able to put
aside their general expectation that each scene will advance the plot
because of their familiarity with the more specific convention of musicals
that successive scenes often present lyrical, tonal, or meditative
“rhymes” instead, so that a scene of Gigi explaining how she feels troubled
and baffled by love is logically followed by a scene in which Gaston
professes similar feelings, even if there is no causal link between
the two.3 On a more practical level, it is viewers’ familiarity with the
conventions of the musical that prevents them from cringing in bewilderment
or distaste when the story stops dead so that Fred Astaire
can dance or Elvis Presley can sing. Learning the generic rules of musicals
does not necessarily allow viewers to enjoy them more, but it
does allow them to predict more accurately whether they are likely to
enjoy them at all. It is therefore a matter of some importance to many
viewers whether or not films like The Wizard of Oz (1939) and Aladdin
(1992) are categorized as musicals, for their feelings about musicals
are likely to influence how much they will enjoy such films, or whether
they are likely to watch them in the first place. This is not to say that
only viewers who like musicals will like The Wizard of Oz and Aladdin.
Both films, in fact, are well-known for appealing to many viewers who
do not ordinarily watch musicals; but appreciative viewers who recog-

nize either film as a musical are more likely to be receptive to other
films that resemble them, confirming the importance of genre in accurately
predicting their enjoyment.
In the same way, asking whether films like Bonnie and Clyde (1967)
and The Wild Bunch (1969) are westerns, even if different viewers answer
the question differently, acknowledges the ways each film’s affinities
to the western – its similarities in mise-en-scène, action, and moral
problems to those of the western – places them in a context that
helps to sharpen and illuminate them. A familiarity with John Wayne’s
outsized heroic persona in westerns like Stagecoach (1939) and Fort
Apache (1948) deepens viewers’ understanding of the more problematic
but equally outsized heroes he plays in later westerns like Red
River (1948), The Searchers (1956), and The Shootist (1976). In each
case, the conventions of the western provide a context that may make
Wayne’s actions more ironic, tragic, or elegiac – certainly more richly
nuanced and comprehensible.
Viewers use many contexts, smaller or larger than established
genres like the western, to interpret conventions of action and performance.
Most viewers watching Stagecoach, for example, assume that
Wayne’s character, the Ringo Kid, will survive his climactic shootout
with the Plummer family, even though he is outmanned and outgunned,
because the survival of characters played by John Wayne is
statistically an excellent bet and because the conventions of classical
Hollywood narrative films4 like Stagecoach make it more likely that
Ringo will proceed to a rousingly heroic climax rather than survive a
hazardous attack by Geronimo’s braves only to be shot down on his
arrival in Lordsburg. Even more fundamentally, most viewers assume
that a climactic shootout will take place in the streets of Lordsburg
because the conventions of classical Hollywood narrative predicate
the resolution of the leading announced conflicts and an economy of
representation that requires each person traveling in the stagecoach
to fulfill the promise of his character and reveal his true nature. But
all these expectations are generic, based as they are on a knowledge
of the wider, though by no means universal, genre of classical Hollywood
narrative within which the western occupies a place that gives
its own conventions their special potency.
Because viewers understand and enjoy movies largely through their
knowledge of the generic conventions, the question of whether gangster
films have enough in common with whodunits and erotic thrillers
to constitute a single genre of crime films is important to many more

people than just film scholars. Even viewers who think they are interpreting
Brian De Palma’s remake of Scarface (1983) exclusively in light
of the conventions of the gangster genre – or, more narrowly, in light
of its departures from Howard Hawks’s 1932 film of the same title –
may well be seeing it in the context of the broader genre of the crime
film. The example of Stagecoach suggests that genres characteristically
nest in one another, the most sharply focused (the John Wayne
western, for instance) drawing their powers from their specific transformations
and adaptations of the conventions of broader genres like
the western or still broader genres like the classical Hollywood narrative.
Although viewers are most likely to be consciously aware of the
narrowest genres, the broader genres that are operating simultaneously
are equally, though less visibly, influential in directing their responses.
Because every genre is a subgenre of a wider genre from
whose contexts its own conventions take their meaning, it makes
sense to think of the gangster film as both a genre on its own terms
and a subgenre of the crime film.
If a genre can be as specific as the John Wayne western or as general
as the well-made Hollywood narrative, then it is clearly possible
to defend the crime film as a genre simply by installing it at a level of
generality somewhere between the gangster film and the classical
Hollywood narrative. But such a solution would prove nothing at all;
it would merely introduce still another category to a field already
crisscrossed with genre markers. The aim of this book is therefore not
simply to introduce a new generic category of the crime film but to
explain how such a category has already been operating to inform
viewers’ understanding and enjoyment of such apparently diverse
genres as the gangster film, the film noir, and the crime comedy.
Establishing the crime film as a genre as rich as those of the western
or the horror film – or, for that matter, the gangster film or the film noir
– raises the problems involved in defining any genre. Genre theorists
have long recognized this as a chicken-and-egg problem. If a genre like
the western can be defined only in terms of its members, but the members
can be recognized as such only by viewers who are already familiar
with the genre, how can viewers recognize any genre without
already having seen every film arguably within its boundaries?5 The
short answer to this question is that they can’t; hence the disagreements
that inevitably arise over whether The Wizard of Oz is to count
as a musical by viewers who have different ideas about what a musical
is. A contrary answer is that they can, despite the lack of theoret-

ical justification. Even if theorists were to demonstrate that the western
was a logically indefensible category, nonspecialist viewers would
go on referring to it because it is so useful and, except at its boundaries,
so easily recognized. Most people can recognize their friends
more easily than they can describe them because different skills are
involved in recognition and description, so that even Supreme Court
Justice Potter Stewart’s oft-ridiculed pronouncement that he couldn’t
define pornography, but “I know it when I see it,” makes sense.6
Recognizing genre conventions is clearly a developmental process.
Few children understand the conventions of Hollywood westerns,
but most adults do. Adults have gradually picked up the conventions
through exposure to particular examples of the genre, because their
understanding of the genre and of particular examples of it have been
mutually reinforcing. When revisionist westerns like Duel in the Sun
(1946) or Unforgiven (1992) appear, they are either dismissed as nonwesterns
or antiwesterns, sharpening the genre’s definition through
their exclusion, or they succeed in redefining the whole notion of the
western by exploring new possibilities implicit in the genre. The mutability
of generic conventions makes it clear that genres are best
thought of as contexts that evolve in both personal and social history,
the contingent results of ongoing transactions between viewers and
movies, rather than eternally fixed and mutually exclusive categories.7
Even given this transactional, evolutionary concept of genres, there
will always be debates about films on the margins of any particular
genre, since many viewers believe, for example, that Singin’ in the Rain
(1952) feels more like a musical than Fun in Acapulco (1963). Some fifteen
years ago, Rick Altman proposed a distinction between syntactic
and semantic definitions of genre to account for the phenomenon of
musicals that have many of the generic markers of musicals (a recognized
musical star like Elvis Presley sings several numbers) but not
others (Fun in Acapulco does not explore the thematic relationships
between performance and sincerity, public and private life, that are
central to musicals like Singin’ in the Rain).8 More recently, Altman has
suggested “a semantic/syntactic/pragmatic approach to genre” to incorporate
into his grammar of textual markers a more systematic
awareness of the multiple users and uses even the simplest films find.9
It is no wonder that Altman has expanded his earlier theory in the
light of the many films marked by conflicting, often shifting generic
allegiances. Most westerns from The Great Train Robbery (1903) to
Unforgiven are organized around stories of crime and punishment;
yet few viewers have called them crime films. If Sunset Blvd. (1950) is
to be counted as film noir because of its confining mise-en-scène, its
trapped hero, and its use of a fatalistic flashback, should Citizen Kane
(1941) be counted as noir too? Is Something Wild (1986) [Fig. 1] a crime
film or a screwball comedy gone wrong? Critics have often coined
nonce terms like “superwestern” and “neo-noir” to describe films that
transform or combine elements from different genres, but these terms
raise as many problems as they solve. If Outland (1981) is an outerspace
western – High Noon (1952) in space – is Assault on Precinct 13
(1976), John Carpenter’s homage to Rio Bravo (1959), an inner-city
western?

This problem of cross-generic allegiances persists even within the
crime film.10 Is The Thin Man (1934) a private-eye story or a crime
comedy? Is The Maltese Falcon (1941) a hard-boiled detective story or
a film noir? The Usual Suspects (1995) combines elements of the gangster
film and the whodunit; how is it to be classified? What to make of
police films that are also studies of criminals, like The Untouchables
(1987) and The Silence of the Lambs (1991) [Fig. 2]? And what about
White Heat (1949), which combines a gangster hero, a film-noir heroine,
an undercover cop, and an extended prison sequence that borrows
the conventions of many another prison film? These problems
are not solved by using the genre of the crime film to dissolve all distinctions
among its long-recognized subgenres; nor are they solved by
declaring one subgenre the categorical victor and ignoring the claims
of others. It makes sense, in such a work of classification as the bibliography
to Barry Grant’s Film Genre: Theory and Criticism, to exclude
gangster films from the crime-film genre on the grounds that “that
group of films is clearly defined to the extent that it can be understood
as comprising a distinct and separate genre.”11 But the distinctiveness
of the gangster film’s conventions cannot support an argument for any
essential distinction between gangster films and crime films, because
there is no reason to assume that distinctive genres are parallel and
mutually exclusive. The caper film, for example, has its own distinctive
generic rules, but those rules do not prevent it from being widely
recognized as a subgenre of an even more well-established genre, the
gangster film, whose gangsters have been assembled in caper films on
an ad hoc basis for a particular job.
Instead of attempting to construct genres that are mutually exclusive,
it would be more judicious to agree with Janet Staiger that “Hollywood
films have never been pure instances of genres,”12 from D. W.
Griffith’s combination of historical epic, war movie, domestic melodrama,
and racial propaganda in The Birth of a Nation (1915) to George
Lucas’s revitalization of science fiction in Star Wars (1977) by recycling
the story of Akira Kurosawa’s samurai comedy-drama The Hidden Fortress
(Kakushi toride no san akunin, 1958), itself based largely on the
conventions of the Hollywood western.
The multiple generic allegiances of most films, however, are obscured
by the fact that some such allegiances have historically overridden
others. Any story presented in animated form, from the musical
romance Beauty and the Beast (1991) to the epic Lord of the Rings
(1978), will automatically be classified as a cartoon because the animated
cartoon is a stronger genre than the genres of romance and
epic. Virtually any story with a setting in nineteenth-century western
America will be classified as a western, because the claims of the western
override the claims of competing genres. Films like Harlan County,
U.S.A. (1976) and Hoop Dreams (1994) are commonly classified together
as documentaries rather than distinguished in terms of their subject
matter. In the same way, films like Blazing Saddles (1974) and The

Naked Gun: From the Files of Police Squad! (1988) are classified as parodies
rather than as members of the various genres whose conventions
they mock, because their parodic intent trumps their affinities
with the specific genres they are sending up.
What makes a genre strong? The example of the cartoon, the strongest
of all popular genres, suggests that the most powerful generic
claims are based on mise-en-scène. Crime-and-punishment tales like
Winchester 73 (1950) and Rancho Notorious (1952) are classified as
westerns rather than crime films because their setting takes precedence
over their story. Any movie set in outer space, from Buck Rogers
(1939) to Alien (1979), becomes a science-fiction movie. The reason
that film noir is such a strong genre, or subgenre, despite the lack of
any clear consensus about what sort of stories it tells, is the powerfully
homogeneous sense of visual style that unites such diverse noirs
as The Killers (1946), Force of Evil (1948), and The Big Combo (1955).
Almost equally powerful as a generic marker is intent.13 Any movie
whose stated aim is to entertain children will be classified as a children’s
film or a family film, whatever its plot or characters or setting
– unless, of course, it is animated, in which case it will be classified as
a cartoon. Comedy, which seeks to make viewers laugh; horror, which
seeks to make them scream; documentary, which seeks to inform
them about some real-life situation; and parody, which seeks to make
fun of other genres – all these are such strong genres that critics have
long categorized Arsenic and Old Lace (1944) and Married to the Mob
(1988), for example, as comedies about crime, rather than crime films
with some laughs; and reviewers who saw Mars Attacks! (1996) as more
imitation than parody unanimously dismissed the film as a failed parody
rather than a successful imitation because they agreed that a parody’s
first duty is to be funny rather than faithful to its sources.
Weaker genres are based on typological situations (boy meets girl,
ordinary characters get into ridiculous scrapes), characters (zombies,
monsters, oversexed high-school students, attorneys), or presentational
features (the story is periodically interrupted or advanced
by dance numbers). Such genres are most likely to be overridden by
stronger genres whose claims conflict with theirs. Thus Abbott and
Costello Meet Frankenstein (1948) is a comedy rather than a monster
movie, and the transsexual science-fiction horror parody The Rocky
Horror Picture Show (1975), however it is categorized, is rarely described
as a musical. When Brian Henderson argued that The Searchers’s
story of rescuers attempting to save a victim who did not want

to be saved actually crossed the boundaries of the western to constitute
“an American dilemma,” in films as different as Taxi Driver (1976)
and Hardcore (1979), his premise did not have the effect of establishing
a new genre of unwelcome-rescue films because the common story
he described did not have the power to override the conflicting generic
allegiances of the examples he cited.14 The disaster genre that
flourished early in the 1970s (Airport, 1970; The Poseidon Adventure,
1972; Earthquake, 1974; The Towering Inferno, 1974) shows that small
numbers do not necessarily make a genre weak; but the disaster genre
is easily overridden by the conventions of the parody, as in Airplane!
(1980), or the action blockbuster, as in Jaws (1975), originally marketed
as a disaster movie until it was recognized as inaugurating a far
more profitable, hence stronger, genre.

Lacking the box-office potential of such recent blockbusters as Independence
Day (1996) and Titanic (1997), most genres can best display
their strength by articulating the central problems that endow
their stock characters and situations and spectacles with power and
meaning. Even apparently unproblematic genres like the musical and
the cartoon can be seen as organized around problems based on their
distinctive presentational features. Musical performers like Fred Astaire,
Gene Kelly, and Judy Garland typically act out rituals dramatizing
the complex relationship between realism and artifice, sincerity
and performance, both while they are performing their song-anddance
numbers and in their characters’ more private moments. Their
films use production numbers to raise questions about public and
private identities and the dynamics of self-presentation, particularly
within the ritualistic context of romantic courtship. Similarly, just as
cartoons are defined pictorially by a tension between the highly stylized
two-dimensional space in which they are drawn and the more
realistic third dimension they imply, they are defined thematically by
the tension between the requirements of realism (empathetic comingof-
age rituals for Disney heroes from Pinocchio to Simba) and magic
(from the constant transformations of shapes and animated objects
typical of all Disney cartoons to the playful self-reflexiveness of Warner
Bros.’ Duck Amuck, 1953).

No matter how it is defined, the crime film will never be as strong a
genre as the cartoon, the horror film, or the parody. It lacks both the
instantly recognizable mise-en-scène of the animated film (or even the
compellingly stylized visuals of the film noir) and the singleness of intent
of the horror film or the parody. But the crime film is a stronger
genre than theorists of subgenres like the gangster film and the film
noir have acknowledged. In fact, it is a stronger genre than the criminal
subgenres that have commanded more attention, not only because
its scope is by definition broader than theirs, but because the
problem it addresses as a genre, the problem that defines it as a genre,
places the film noir and the gangster film in a more sharply illuminating
context by showing that each of those is part of a coherent larger
project.

The defining problem of the crime film is best approached through
the specific problems involved in establishing it as a genre. Should the
crime film be defined in terms of its subject, its effect, or its visual
style? Many crime films adopt the visual conventions of film noir (lowkey,
high-contrast lighting, unbalanced compositions, night-for-night
exterior shooting), but others do not. If the noir visual style is a defining
feature of the crime film, how are color films like Leave Her to
Heaven (1945), Chinatown (1974), and Pulp Fiction (1994) [Fig. 3] to be
categorized?
If the noir visual style seems to produce too narrow a definition
of the crime film, its characteristic subject, crime, and its frequently

3. Pulp Fiction: A noir world of criminals like Jules Winnfield (Samuel L. Jackson)
without the noir visual style.
sought effect, suspense,15 are impossibly broad. Both crime and suspense
have an important role in a very great number of movies. The
English Patient (1996) presents several important crimes, from robbery
to murder, and a detective figure in David Caravaggio (Willem
Dafoe); do those elements make it a crime film? Every classical Hollywood
narrative depends on some disruption of the social order for its
conflict, and an enormous number of social disruptions (e.g., the fire
in The Towering Inferno, which is started by the illegal installation
of substandard wiring) are rooted in crimes. It would surely be impractical
to call every film in which a crime produces the central dramatic
situation a crime film. The touchstone of suspense is even more
hopelessly vague, since suspense might be called a defining feature
of the well-made Hollywood narrative. Even Jane Austen adaptations
from Pride and Prejudice (1940) to Emma (1996) depend on the suspense
generated by the questions of who will marry whom, and how
the anticipated happy ending can be compassed. How can the crime
film be distinguished from the broader category of the classical Hollywood
narrative, and how useful is such a vaguely defined genre likely
to be?

The problem of defining the crime film is exacerbated by three problems
implicit in its subject. John G. Cawelti has noted that popular narrative
genres almost by definition package “the ultimate excitements
of love and death” within the most reassuring generic formulas in order
to appeal to both viewers’ flight from ennui and their love of security.
16 In crime film, this paradox is linked to the question of crime’s normalcy.
By definition crime is an aberration, a disruption to the normal
workings of society; yet crime films invariably treat crime as normal
even as they observe the ways it undermines the social order. Gangsters
do nothing all day long but smuggle or steal. Police officers pursue
criminals for a living. Every single case a private eye like Philip
Marlowe takes on turns criminal; every adaptation of a John Grisham
novel of legal intrigue, even if the initial proceeding is a civil one, explodes
in violence sooner or later. Crime films all profess to solve the
criminal problems they present by means of a happy ending; yet the
frequency of crime in such films suggests that the more general problems
posed by crime will never be solved. Is criminal behavior in these
films abnormal or all too normal?

The second problem cuts even deeper. In distinguishing between
the heroes of thrillers, who “almost exclusively represent themselves,”
and the heroes of crime films, who “represent the Criminal, the Law,
and Society,” Carlos Clarens implies a distinction between crime as an
isolated event (the province of the thriller) and crime as a metaphor
for social unrest (the province of the crime film).17 But how solid is
this distinction? In Clarens’s terms, the work of Alfred Hitchcock, the
filmmaker most closely identified with crime, includes only thrillers
rather than crime films; yet critics from Eric Rohmer and Claude Chabrol
to Robert Corber have recognized that the criminal plots of all
Hitchcock’s films, from The Lodger (1926) to Psycho (1960), have obvious
moral and social implications that range far beyond the plight of
the characters themselves.18 When is a cinematic crime a metaphor
for an enduring moral dilemma or social upheaval or ideological critique,
and when is a crime just a crime?
The third problem concerns what may seem like the most straightforward
components of the crime film: its stock characters. Every
crime story predicates three leading roles: the criminal who commits
the crime, the victim who suffers it, and the avenger or detective who
investigates it in the hope of bringing the criminal to justice and reestablishing
the social order the crime has disrupted. The three roles
could hardly be more clear-cut, yet they everywhere overlap and melt
into each other. Gangsters like Vito Corleone are devoted family men
concerned only to protect and provide for their loved ones. Victims
like Paul Kersey, the bereaved hero of the Death Wish franchise (1974–
94), turn vigilante in order to avenge their loved ones. Maverick cops
like Harry Callahan, in Dirty Harry (1971) and its sequels Magnum Force
(1973), The Enforcer (1976), Sudden Impact (1983), and The Dead Pool
(1988), break the law in order to catch criminals they know are guilty.
A critique of the justice system is obligatory in Hollywood movies
about lawyers, police officers, or private eyes. When the hero is a good
cop, he is set against an entire corrupt department, as in Serpico
(1973), or ends up battling vigilante demons inside himself, as in The
Untouchables. And Hollywood movies about victims who merely suffer,
as opposed to taking arms against their oppressors, are virtually
unheard of. Evidently crime films both believe and do not believe in
the stock characters at their center; they seem determined to undermine
and blur the boundaries of the typological figures that might
otherwise stake their surest claim to the status of a single genre.
Although these problems might seem to present insuperable obstacles
to the definition of the crime film, they are in fact at the heart of
such a definition: for the crime film does not simply embody these
problems; it is about them. Crime films present as their defining sub-
ject a crime culture that depends on normalizing the unspeakable, a
place where crime is both shockingly disruptive and completely normal.
Crime may have different metaphorical valences in different criminal
subgenres – it can demonstrate the fragility of the social contract
in thrillers about innocent men on the run, attack the economic principles
of the establishment in gangster films, express philosophical
despair in films noirs, test masculine professionalism in private-eye
films – but it is always metaphorical. Every crime in every crime film
represents a larger critique of the social or institutional order – either
the film’s critique or some character’s. Finally, crime films dramatize
not only the distinctive roles of criminal, victim, and avenger but also
their interdependence and their interpenetration.
The problem at the heart of crime films, then, is their attempt to mediate
between two logically contradictory projects. Like all popular
genres, crime films work primarily by invoking and reinforcing a cherished,
but not entirely convincing, series of social bromides: The road
to hell is paved with good intentions, the law is above individuals,
crime does not pay. Crime films need to reinforce these beliefs, just as
viewers want to have them reinforced, in order to confirm the distinctiveness
of the moral and legal categories that allow viewers to maintain
their sense of social decorum and their own secure place in the
social order as law-abiding citizens who know right from wrong, identify
with the innocent, and wish to see the guilty punished. It is no surprise
that the Hollywood film industry is eager to endorse these bromides,
since the industry’s continued success depends on the health
of the capitalist economy. The moral certitudes on which the industry
and its audience agree depend on a series of categorical distinctions
among the roles of victim, who ought, according to Hollywood’s
official morality, to be their natural identification figure; the criminal,
who ought by the same token to be the target of their fear and hatred;
and the avenging detective, who ought to express the law in its purest
yet most personal form.
Viewers for crime films know that these three figures – the innocent
victim, the menacing criminal, the detective who incarnates the law –
never exist in such pure incarnations, not only because of the requirements
of realism and narrative complexity but because they would
be utterly uninteresting. The ritual triumph of avenging heroes over
criminals is compelling only as ritual; to succeed as narrative, it requires
complications and surprises in the conception of the leading
roles and their relationships. The fascination of crime films arises pre-
cisely from the ways they test the limits of their moral categories, engaging
and revealing contradictions in the audience’s fantasies of identification
by mixing elements from these three different positions, the
primary colors of crime films that never occur in isolation. Although
crime films typically move toward endings that confirm the moral absolutes
incarnated in each of their three primary figures, an equally
important function crime films share is to call these primary figures,
and the moral absolutes that inspire them, into question by making a
case for the heroic or pathetic status of the criminal, questioning the
moral authority of the justice system, or presenting innocent characters
who seem guilty or guilty characters who seem innocent. Even
when the endings of crime films endorse a reassuringly absolutist view
of crime and punishment, the middle of such films puts absolutist categories
like hero, authority, innocent, guilty, victim, criminal, and avenger
into play, engaging the doubts and reservations about these labels
that make them fit subjects for mass entertainment as well as moral
debate, and so raising questions that the most emphatically absolutist
endings can never entirely resolve.
Crime films always depend on their audience’s ambivalence about
crime. The master criminal is immoral but glamorous, the maverick
police officer is breaking the law in order to catch the criminals, the
victim is helpless to take any action except capturing or killing the
criminal. It is therefore inevitable that they both insist on the distinctions
among criminals, crime solvers, and victims, and that their obsessive
focus is on the fluid and troubling boundaries among these
categories. Crime films are about the continual breakdown and reestablishment
of the borders among criminals, crime solvers, and victims.
This paradox is at the heart of all crime films.
Crime films operate by mediating between two powerful but blankly
contradictory articles of faith: that the social order that every crime
challenges is ultimately well-defined, stable, and justified in consigning
different people to the mutually exclusive roles of lawbreakers, law
enforcers, and the victims who are the audience’s natural identification
figures; and that every audience member is not only a potential
victim but a potential avenger and a potential criminal under the skin.
The audience’s ambivalence toward both these premises, and the
shifting identifications crime films therefore urge among the fictional
roles of lawbreaker, law enforcer, and victim, are the defining feature
of the genre, and the feature that indicates the place each variety of
crime film has within the larger genre.

Hence the genre of crime films includes all films that focus on any
of the three parties to a crime – criminal, victim, avenger – while exploring
that party’s links to the other two. What defines the genre,
however, is not these three typological figures any more than a distinctive
plot or visual style, but a pair of contradictory narrative projects:
to valorize the distinctions among these three roles in order to
affirm the social, moral, or institutional order threatened by crime,
and to explore the relations among the three roles in order to mount
a critique that challenges that order. This contradictory double project,
which has often been obscured by the predominance of subgenres
like the gangster film and the film noir over the crime film, underlies
the ambivalence of all the crime film’s subgenres, including
several this book will not consider in detail. White-collar crime films
like Wall Street (1987) explore the paranoid hypothesis that American
capitalism is at its heart criminal; caper films like The Asphalt Jungle
(1950) present a criminal culture more admirable in its honor and professionalism
than the official culture it subverts; prison films from
Brute Force (1947) to The Shawshank Redemption (1994) explore the
nature of legal and moral guilt in order to consider how individual humanity
can survive the dehumanizing rituals of the prison system.
One final apparent omission deserves fuller mention because, as
Carlos Clarens has acknowledged, it goes to the heart of the crime
film’s definition: the thriller. The crime film has much in common with
the thriller; but following Charles Derry’s brief definition of the thriller
as “films in the shadow of Alfred Hitchcock”19 reveals that the thriller
is not, as Clarens argues, a parallel alternative to the crime film but a
subset of it. Although every crime film postulates the same three pivotal
figures, different figures predominate in different criminal subgenres.
The criminal is most prominent in gangster films and films
noirs; the avenging crime solver in detective films, police films, and
lawyer films; and the victim in the man-on-the-run films of which Hitchcock
made such a specialty. In a larger sense, however, all of Hitchcock’s
films are about victims. The types of crime films Hitchcock never
essayed – films about professional criminals, about ordinary people
sucked into committing crimes, about heroic agents of the justice system
– make up a virtual catalog of the types of films about criminals
and avengers. Despite Hitchcock’s bromide, “The more successful the
villain, the more successful the picture,”20 he never makes a criminal
the hero of a film without recasting that criminal, from Alice White in
Blackmail (1929) to Marnie Edgar in Marnie (1964), as a victim. Hitch-
cock’s distaste for the police is even more well-known; he regards legal
authorities of any sort with suspicion and fear. His abiding interest
therefore remains with innocent people who are unjustly suspected
of crimes (North by Northwest, 1959), or who must confront criminals
without any help from the authorities (Shadow of a Doubt), or who
turn detective in order to clear themselves or save their country (The
39 Steps, 1935). Hitchcock’s thrillers, indeed thrillers generally, are
essentially crime films that focus on the victims of crimes, or of the
criminal-justice system.
Including in the definition of crime films all films whose primary
subject is criminal culture, whether they focus on criminals, victims,
or avengers, may seem to make the genre too broad to be truly useful
or distinctive. But the test of this definition, like that of any genre, is
neither its narrowness nor its inclusiveness; it is its ability to raise
questions that illuminate its members in ways existing modes of thinking
about crime films do not. If all genres, as Staiger and Altman suggest,
are contingent, evolving, and transactional,21 the question they
raise is not whether or not a particular film is a member of a given
genre, but how rewarding it is to discuss it as if it were. Nearly any film,
from The Wizard of Oz to The English Patient, might be considered a
crime film. The model of ambivalence toward the categories represented
by the criminal, the victim, and legal avenger is not meant to
distinguish crime films from non–crime films once and for all, but to
suggest a new way of illuminating the whole range of films in which
crimes are committed.

Crime Films/Thomas Leitch/ http://www.cambridge.org