The Heist Film: Stealing with Style

There are several ways to approach a history of the heist film, each with its limitations. One way of answering the question is to trace the ‘parentage’ and ‘evolution’ of heist movies over time. But we can only do this as long as we do not get trapped working through the entire corpus of films with theft as a central plot element. This would fall short of providing a precise conceptual account since plenty of films with petty thieves, or genre films with exciting robberies, let alone idiosyncratic films like the documentary Man on Wire (James Marsh, 2008), are partially identifiable as a caper. There is a reason critics refer to the ‘big caper’ film. They emphasise the centrality and grandeur of the crime in the narrative. Tracing the heist’s filiation would helpfully tease out features shared with well-defined genres, criminal or otherwise, such as 1930s gangster films. Yet, where the gangster film has remained relatively stable over time, a sign of its enduring form, the heist film has been taken as a weak or unstable genre by critics who nevertheless know one when they see it. From another point of view, early heists find themselves just as often in a historical lineup with film noir, which has been variously understood as a cycle, a visual style, or a discourse of crime film, but in any case is lacking generic moorings. A heist can be both a type of gangster film and a narrative variant of film noir. Or more problematically for the parentage approach, neither of them – many heists are purely comical, have bumbling thieves as opposed to gritty tough guys, or are devoid the visual qualities of noir.

In addition, any account of the heist film must take into account its evolving critical conceptualisation, paying close attention to a fluctuating taxonomy and terminological differences proposed by film theorists, historians and industry players. When Stuart M. Kaminsky revised the second edition of his influential study American Film Genre a decade after the first edition (1974, 1985), he expanded it but also dropped the chapter ‘Variations on a Major Genre: The Big Caper Film’. Kaminsky’s analysis stands as the most important of the genre’s formula and is a crucial reference for this study. But its purview ends in the mid-1970s. For Kaminsky, the big caper had already enjoyed a ‘limited but distinct history’ (1974: 74) to this point, but had ‘not continued as a significant generic form’ (1974: vii) from the 1970s through the mid-1980s. That claim resonates with another made in a much-read 1984 essay by the genre theorist Rick Altman in which he argued that the big caper was among a set of genres that had failed to develop a ‘coherent’ or ‘stable syntax’ (1984: 16). This view persisted tacitly when Altman included the same essay in his study Film/Genre (1999: 225) that appeared just as the heist began a resurgence around 2000. With few notable exceptions, heist films continued to appear sporadically between the mid-1970s and the late 1990s, such that Kaminksy and Altman’s claims have merit. But I wonder to what extent those claims may be viewed as historically contingent, since starting around the mid-1990s, and particularly since 1999, a proliferation of heists have solidified the heist genre’s syntactic patterns while modifying its social function or message by emphasising certain virtualities of the genre that have existed since its inception around 1950.

By ‘textual structure’, to begin with, I have in mind the ‘semantic/syntactic approach’ of Rick Altman. Semantics, for Altman, refer to recurring ‘building blocks’ of the same film genre, ‘common topics, shared plots, key scenes, character types, familiar objects or recognizable shots and sounds’ (1999: 89). Stuart Kaminsky was the first to carefully analyse those recurring semantic features of the heist that provide it a lexical ground. Among the heist’s most salient semantic features are its character types, which are, according to Kaminsky, of two broad categories. First, the gang leader, defined as a man of action (rarely a women), and a mentor with experience of the criminal world (1974: 80). In John Huston’s The Asphalt Jungle (1950), Dix Handley (Sterling Hayden) is the man of action, mentored by the mastermind of the job, ‘Doc’ Riedenschneider (Sam Jaffe); in The Italian Job (F. Gary Gray, 2003 version), John Bridger (Donald Sutherland) is the mentor, Charlie Croker (Mark Wahlberg) the man of action. Sometimes the mentor and man of action are blended into a single character. Next, there are the team members with ‘individual skills and crafts which command no great social respect and which have little or no chance of making those who possess them wealthy by any legal method’ (Kaminsky 1974: 79). This group of social misfits and societal castoffs traditionally embraces serious craftsmen, such as safecrackers, mechanics, drivers, demolitions experts and other technicians. Semantics, however, has its limitations as a means for identifying genres – at least since Jules Dassin’s caper team in Topkapi (1964), made up of peculiar members like mute acrobats and a maker of automata. The quirky cases of Virgil (Casey Affleck) and Turk (Scott Caan) Malloy, the puerile brothers from Provo, Utah, in Ocean’s Eleven (Steven Soderbergh, 2001 version), are an example of this sort too. The serious craftsmen and the curious oddball, as opposed as they might appear on the surface, occupy the same role as thieves in a heist film. Yet another binary divides the heist film’s types between a ‘thuggish prole’ and cool caper characters of the sort found in the original Ocean’s Eleven (Newman 1997: 71). There are seeming differences in the sleek professional thieves of Michael Mann’s Heat (1994) and in the bumbling professional ones in Woody Allen’s Small Time Crooks (2000), but then there are amateur thieves in many a comic heist such as Mad Money (Callie Khouri, 2008). Semantics may be necessary for generic understanding, but it is far from sufficient. In the aforementioned examples, as thoroughly dissimilar as the heroes might appear, they perform the same task in heist plots and the audience is led to cheer for their victory.

Heist films afford a powerful screen identification with criminals breaking the law, providing ‘escapism and voyeurism, or in other words the pleasure of watching stories about illicit worlds and transgressive individuals … that may appeal to our fantasies and desires’ (Thompson 2007: 4). The heist encodes in story form a particular desire to elude the oppressive aspects or limitations of contemporary mass society. We recognise the necessary cooperation among team members and the rivalry between competing criminals, and want to see whether the social microcosm remains resolute or dissolves in the face of pressures (see Mason 2002). Will the team overcome obstacles in order to execute the plan, pull off that one last job, and break free at last from whatever constraints bind them? Risk and reward, bound in an equation invariably distorted by the ‘foolproofness’ of the plan, induces apprehension in the spectators and the hope that they will experience an affective payoff at the end. The audience’s reception of the heist film is thus also a part of its utility. But in this we are encroaching on Altman’s notion of generic ‘syntax’.

Syntax works in alignment with audience reception and serves to identify ‘generic affiliation’ more tellingly than semantics ‘because a group of texts organizes those building blocks in a similar manner (as seen through such shared syntactic aspects as plot structure, character relationships or image and sound montage)’ (Altman 1999: 89). The syntactic dimension of Altman’s model for analysing genre is crucial: vocabulary (semantics) varies with relative ease, as I have suggested, but syntax often matters more in figuring out form and meaning, especially over time. At the heart of the heist film is the extraordinary robbery of a formidable institution that requires careful planning and the skills of specialists. Kirsten Thompson’s lucid definition calls the heist ‘a cycle or sub-genre of crime films that feature an elaborately planned and executed robbery, and whose narratives emphasize the logistical and technical difficulties of a crime and its execution’ (2007: 43). And Kim Newman’s superbly concise entry on the ‘caper’ as a criminal ‘sub-genre’ in the cheekily titled BFI Companion to Crime points to how ‘professional crooks plan and execute a clever, daring but (the censors insisted) ultimately unsuccessful robbery’ (1997: 70). Newman’s definition reiterates the central narrative focus on crime, highlights the skill and ingenuity of the ‘professional’ criminals, and foregrounds the failure of the crime in most capers made before 2000. (At this moment the genre seems to turn away from its longstanding ‘fatal strategy’, as Telotte puts it [1996: 163], that is, its tragic ending or the failure of the criminal exploit; I’ll return to this point.) Newman and Thompson’s definitions echo Kaminsky’s: ‘the one essential element of big caper movies, the essential which defines them, is the plot concentration on the commission of a single crime of great monetary significance, at least on the surface’ (1974: 77). Kaminsky’s qualifier ‘big’ implies a unique subset of films differentiated from those in which a robbery functions merely as a backdrop. The job must involve a risky operation that will reap great rewards; the plot ‘concentrates’ on a singular crime that occupies a significant portion of the diegesis and the affective buildup of the story, ennobling the work, skilled labour and physical or mental effort of the crimes represented. The narrative focus of the heist film is wrapped up in the ‘execution’ of the crime as much as or more than criminal psychology.

Still, this textual account of the heist film falls short if it fails in the end to show how the genre appropriates its own function and expresses a particular message that corresponds to the changing desires or interests of its purveyors and audience. This inevitably requires us to consider the ‘social context in which the [genre] is used’ (Altman 1999: 97). This book sets out to provide a theory of one recurrent social function of the heist film as one among many (criminal) genres. Classifications, definitions and illuminating interpretations of heist films already exist. The task will be to emphasise underappreciated connections between existing typological definitions and histories, and relate them to possible social meanings. Is there a community of structure or intention between early, middle and late heist films? For the most part, from the mid-1950s onward the semantics and syntax of the heist film remain in place. Yet there may not be continuity across six decades in terms of the genre’s social message, although I do believe there is one in terms of two of the genre’s social functions.

My analysis will draw out another aspect of this socio-economic critique in the heist film to articulate narrower function of the genre. To get at that function requires us to pause on a key phrase in Kaminsky’s definition of the crime in the big caper: ‘a single crime of great monetary significance – a crime always directed at monetary gain, at least on the surface’ (1974: 77; emphasis added). What motivates the thieves appears to be straightforward: cash for a better life, the means to get out of financial trouble or stagnancy. But as critics and audience members alike know, while the leader’s motives may start there, they are inescapably more complicated. As Barry Keith Grant reminds, ‘genre movies take … social debates and tensions and cast them into formulaic narratives, condensing them into dramatic conflicts between individual characters and society or heroes and villains’ (2007: 16). If so, then a clean break from society may be a surrogate for something more ethereal, perhaps even a search for a kind of transcendence that makes the heist, well, rather poetic. In fact, that’s precisely how Steve Frizelli (Edward Norton) dismisses Charlie Croker’s plan to steal back gold bullion from him in The Italian Job: ‘I mean, that’s very poetic and all, but I just don’t see it.’ Charlie’s curt reply is that Frizelli ‘ain’t got no imagination’.

Heist films are also about imagination and poetry. They examine aesthetic activity by encoding the values of imagination and creative effort into criminal activity, and construct criminals as rule-breaking artist-geniuses whose labour (mental or physical) unfolds as a process of artistic creation and their efforts producing an artistic or poetic work. This is true from the genre’s origins. In Stanley Kubrick’s The Killing (1956), one of the first and most violent noir heists, one gang member tells another: ‘the gangster and the artist are the same in the eyes of the masses. They are admired and hero-worshipped, but there is always present an underlying wish to see them destroyed at the peak of their glory.’ The analogy between the artist and criminal in mass society dates back at least to nineteenth-century romantic heroes, probably even to Enlightenment theories of art (see Porter 1981, Senelick 1987 and Black 1991). This fact locates the heist film within a larger field of reflection on the nature and role of artists and art in modern times. These theories shaped the law-breaking criminal genius we see in many crime films – displaced from a moral or criminological plane onto an aesthetic one. Moreover, the exploding market economies of the modern era ushered in a new age of consumerism and economic valuation that impacted the wages and living conditions of a new industrial class of labourers, as well the products of the artist’s mind and hands. Market economies pressured the widespread nineteenth-century notion of aesthetic autonomy – the doctrine that art is not subject chiefly to the money-making interests – making it seem impossible to preserve art from the supposedly corrosive touch of commercialisation.

Any account of the heist film must take into account its changing critical conceptualisation, paying close attention to a fluctuating taxonomy and terminological differences proposed by film theorists, historians and industry players. I have already mentioned Kaminsky’s claim in the second edition of American Film Genres that the genre dwindled ‘as a significant generic form’ (1985: vii), a position seconded by Rick Altman’s sense that the genre hadn’t produced a ‘coherent’ or ‘stable syntax’ (1984: 16). Film historians and theorists generally take the big caper to be a criminal sub-genre or at best a weak genre, but ambivalence sways the day. Eugene Rosow is non-committal in his study of American gangster films: ‘heist or caper films … can be considered a sub-genre (or separate genre)’ (1978: 279). Ron Wilson claims ‘the caper or heist film was another crime sub-genre of the 1990s’ (2000: 154). For Steve Neale, heists are an ‘intermittent series’ of the gangster film stretching from the 1950s to the 1980s (2000: 81). The most relevant contribution comes from Altman, for whom the ‘big caper’ is fully a genre and not a sub-genre. Indeed, the big caper figures in his discussion of generic durability:

Terminology can be mildly elusive here, so some attention to the terms caper and heist is in order. ‘Caper’ likely derives from a Dutch noun for a corsair and its corresponding verb, kapen, which means to take or plunder. But a more suggestive possibility is related to ‘capriole’, a ‘fantastic’ and ‘frolicsome leap’ that applies to dance and horsemanship (http://www.oed.comaccessed7/24/2011). From this, ‘caper’ has assumed a connotation for any fashionable pursuit, even an elusive or tricky game. American usage includes the notion of a ‘capricious escapade’ and an ‘illegal or questionable act’ (http://www.merriam-webster.com/dictionary/caper accessed 7/14/2011). The term thus bridges ludic notions of performance, spectatorship and criminal activity, always governed by playful moves and unexpected outcomes. ‘Heist’, common to United States slang, derives from the German and is a variant of ‘hoist’, for lifting or raising. Historically it signified housebreaking or shoplifting, and in twentieth-century usage it refers to armed robbery. That aspect gives ‘heist’ a more ominous, violent association than the lighter, performative aspects of ‘caper’, and this divergence appears to have impacted critical usage. ‘Heist’ is more inclined to evoke the serious, dramatic, violent and perhaps professional dimensions of criminals carrying out ill-fated robberies, while ‘caper’ lends itself to the colourful and amateurish side of such crimes (almost literally, colour films as opposed to their black and white – or film noir – counterparts). The difference between the two terms generally identifies the heist as a dramatic form, whereas caper tends towards the comic. The distinction between the two is not absolute; however, it may be useful if we keep in mind two things. First, that a film audience roots for the ‘thuggish prole’ (Newman 1997: 71) in the heist just as much as for cool caper characters – sympathy is elicited for both, an aspect of the genre’s audience reception that surpasses semantic difference (thug versus cool hero). Second, that crime films (and fiction) do not parse the terms in the same way some critics do; they use caper and heist interchangeably. ‘Caper’, for example, is the actual term used for the robbery in the source novel High Sierra by W. R. Burnett and its film adaptation by Raoul Walsh (1941), as well as in Robert Siodmak’s crime films – all works that were gangsterish, noir, certainly not comic.

The relation between the two critical labels is subject to history too. ‘Caper’ gained currency in critical discourse well over a decade after the 1950s dramas first appeared, at a time when comedy had overtaken the genre. Newman explains that the caper film ‘was largely unknown in Europe until the late 60s’, at which time it was used ‘retrospectively’ to describe 1950s dramas about professional crooks trying, unsuccessfully, to pull off a daring robbery (1997: 71). When the term ‘caper’ came into more common usage in the 1960s, it was living up to its double meaning: it ‘had a much stronger element of romantic comedy and often a fantastic plot involving the theft of a priceless objet d’art’ (ibid.). With reference to The Thomas Crown Affair (1968 version), the Overlook Film Encyclopedia says essentially the same thing: ‘the film’s playfulness is an indication of how far the heist film had come from its origins (represented in America by The Asphalt Jungle, 1950) in the process of its transformation into the caper film (as exemplified earlier by Ocean’s Eleven, 1960, and Topkapi, 1964)’ (Hardy 1998: 273). The designation between ‘heist’ and ‘caper’ is thus not a static binary but a shifting language used by writers, the industry, audiences and critics to make sense of forms.

Needless to say, I proceed with some trepidation. At times I adopt the difference between (comic) caper and (noir) heist as used in critical language. Otherwise, in the following chapters I use the terms interchangeably, and overall my study subsumes both under the designation ‘heist’. Chapter one will trace the origins of the heist film, moving from precursors in the silent era through film noir, particularly Robert Siodmak’s The Killers (1946) and Criss Cross (1949), to John Huston’s groundbreaking film The Asphalt Jungle, and setting the heist film against the cultural anxieties of urbanism and the post-war economy. Chapter two reads a core set of mid-1950s films – noir heists by Jules Dassin (Rififi), Stanley Kubrick (The Killing), and Jean-Pierre Melville (Bob le flambeur), and comic heists by Charles Crichton (The Lavender Hill Mob, 1951) and Mario Monicelli (Big Deal on Madonna Street, 1958) – that crafted the heist into a reflection on the film artist’s work in an era pervaded by ideological and economic compromise. Chapter three considers how the heist film, having consolidated its conventions in the late 1950s, tosses off the fetters of fatalism in favour of more staid dramatic tones or flashy big production caper comedies that provide a ready-made form for pressing socio-economic concerns the heist addressed across the 1960s and early 1970s. Chapter four analyses the resurgence of the heist in the 2000s, extrapolating a general principle about the contemporary heist by noting the contradictory myths of originality conveyed in the numerous remakes that characterise the last decade.