Energy and eventhood: the infrastructural set piece

ABSTRACT “Energy and Eventhood” considers the relationship between infrastructure and the cinematic set piece. It proposes a working definition of the set piece as a distinctly (and perhaps excessively) effortful passage or sequence, and reflects on the tendency for narrative films to incorporate infrastructural sites and structures—such as bridges, dams and pipelines—into such set pieces. A number of writers on infrastructure, as both a cultural phenomenon and a representational object, have noted the aesthetic challenge of rendering it as visible and locatable; this article examines how that difficulty becomes manifest in the infrastructural set piece. It takes as its case studies Unstoppable (2010) and Night Moves (2013), two films noted for their distinctive rhythmic expressiveness, and each one deploying the convention of the set piece in ways that exemplify and reflect on the resistance of infrastructure to narrative containment.

In a very astute skit on HBO's Last Week Tonight, comedian John Oliver defines infrastructure as "basically anything that can be destroyed in an action movie," and shows us a string of clips from Hollywood films demonstrating exactly that. He then riffs on the "unsexiness" of infrastructure, on its resistance to drama and spectacle, and introduces a parodic film trailer for a hypothetical movie about heroic routine maintenance of pipelines, dams, and bridges. The joke rings true, but leaves aside another, less cataclysmic version of infrastructure in popular cinema, namely that which invokes a kind of conspiratorial energy. Harry Lime's attempted escape through the Viennese sewer in The Third Man (Carol Reed, 1949) is perhaps the most famous rendition of this tendency, in which films trade on the opacity of infrastructure to make palpable a sense of doubt and confusion regarding power and responsibility. 1 Important meetings in The Parallax View (Alan J. Pakula, 1974), The Insider (Michael Mann, 1999) and Inherent Vice (Paul Thomas Anderson, 2014), for example, are conspicuously backed by visible indicators of powerful networks, even if those networks seem to have no discernible connection to the conspiracy stories in play. Infrastructure-and this is not an uncommon observation-carries into the aesthetic realm a sense of both the spectacular and the inscrutable.
In this essay, I will explore another (related) tendency in filmic explorations and representations of infrastructure: its appearance by way of set pieces. The discussion is about infrastructure in films; not, however, primarily as a theme, or as an iconographical element, or even as a location category, but rather as a subject that is signalled in particularly concentrated passages, in the energetic and punctuative events we call set pieces. The films on which I focus, Unstoppable (Tony Scott, 2010) and Night Moves (Kelly Reichardt, 2013), both offer distinctive and telling variations on the cinematic set piece. Unstoppable punctuates its almost relentless momentum with intermittent events, while the most narratively salient event in Night Moves is not even seen by the protagonists or the audience (though it is heard by all). What, I ask, do these characteristics have to do with the films' infrastructural subject matter, and their deployment of images, locations and scenarios of energy and communication networks?
Recent years have seen a growing recognition of infrastructure's fundamental relevance to questions that are firmly rooted in some long-standing concerns of humanists-questions of (for example) sublimity, of our experience of modernity, of power relations, of scale, and of material culture. 2 Given this vast scope and reach of infrastructure, it is not surprising that writers have tended to explore what we might call its macro implications for societies and ecologies; furthermore, film and media scholars have also had to urgently consider the infrastructural and environmental contexts of production, distribution, and storage in an era when it is proving all too easy to forget the material preconditions and traces and of moving-image culture. Broadly put, these are some of the key concerns and insights brought together in the important and ambitious collection Signal Traffic, whose contributors explore media infrastructures as crucial (and crucially overlooked) political, environmental and cultural subjects. In their introduction to Signal Traffic, editors Lisa Parks and Nicole Starosielski rhetorically ask "what can media and communication studies gain by adopting an infrastructural disposition?" (2015,5) and suggest that one of its most important tasks as a methodology is to serve and develop technological literacy, ultimately in the name of social justice. If my own discussion can be said to take up an infrastructural disposition, its particular contribution is to the interpretation of filmed infrastructure, rather than to our understanding of material sites or technologies themselves. This is not to say that knowledge and power are irrelevant, but rather that I study them as part of a range of associations and potential effects bound up with infrastructure-to be brought into some kind of meaningful life by filmmakers.
John Oliver's joke on Last Week Tonight relies on the fact that a great many films have capitalized on these potential effects-what Brian Jacobson (2016) terms cinema's "infrastructural affinity"-in familiar and exhilarating ways, invariably through set pieces. Unstoppable and Night Moves warrant particularly close attention, I argue, because they demonstrate a kind of deft reflexivity about the logistical logic of narratives and aesthetics in popular cinema. Recent work on infrastructure as a system into which all media is necessarily bound has been an important development, but there is perhaps more to be said about how this plays out in the meaningful expressiveness of particular works; what, in other words, does this infrastructural condition look and feel like at the level of moments and sequences? Set pieces offer a productive starting point for thinking this through, especially those set pieces which foreground what we might call systemic locales and technologies. This essay considers the inevitable tension that arises when films play off the diffuse, circulatory nature of infrastructure against locally concentrated events.
The cinematic set piece is too vague and diverse to constitute a convention, appearing in myriad shapes and forms across many genres and modes. But the term's popularity and resilience in film culture should be acknowledged as evidence of some kind of usefulness. Set pieces are closely related to what Higgins (2008) calls "situational dramaturgy," in which classical causality and goal orientation are temporarily suspended. Higgins's focus, though, is on the action film, while set pieces in some shape or form permeate almost all cinema. An arbitrary list of examples to initially consider might include: Roger O Thornhill's fleeing on foot from a crop-dusting plane in North By Northwest (Alfred Hitchcock, 1959); the suicide of Majid in Caché/Hidden (Michael Haneke, 2005); Anna's ice-floe rescue in Way Down East (D.W. Griffith, 1920); Mabel's "Swan Lake" dance in A Woman Under the Influence (John Cassavetes, 1974); Chantel Akerman's boatmounted camera gazing back at a receding Manhattan in News From Home (1977); the jousting tournament in Lancelot du lac/Lancelot of the Lake (Robert Bresson, 1974); Jack Terry's desperate sprint through a fireworklit public parade to rescue Sally in Blow Out (Brian De Palma, 1981); enormous car jams in Weekend (Jean-Luc Godard, 1967) and Nashville (Robert Altman, 1975). Each of these can be said to develop some level of internal coherence and identity through a bracketing off of production methods, style and/or address, however subtle. Set pieces seem to want to be received (and remembered) as operating within slightly adjusted parameters; their relative independence punctuates a film's rhythm. 3 That set pieces tend to attract a disproportionate amount of critical commentary is neither mere coincidence nor simply a self-fulfilling prophecy. It is a response to those sequences' apartness, which is not reducible to their narrative significance.
A fully developed aesthetic theory of set piecesbeyond the scope of the current essay-would, I believe, need to interrogate their experiential character, and may even require something akin to the "naturalized aesthetics of film" recently set out by Murray Smith in Film, Art and the Third Culture (2017). Smith argues for a "triangulation" of phenomenological, psychological and neural considerations in our accounts of particular films and sequences, and describes a "fully fledged aesthetic experience" as "one that we savour rather than simply have" (7)-a turn of phrase which likewise speaks to our heightened investment in (and self-consciousness about) set pieces in film, and our relishing of their expressive organization and execution. This essay, meanwhile, concerns not so much the theoretical parameters for an extensive study of the set piece, but rather a conjunction between its defining qualities and the subject of infrastructure.
My list of provisional, indicative examples of setpiece sequences deliberately veers across very dissimilar films and scenes, so what common qualities are we left with that "set piece" might still be sensible and useful as a critical term, and one which could help us better understand Unstoppable, Night Moves and their infrastructural subjects? I propose the following two: • A set piece is a sequence in a film when we are invited to appreciate (if not consciously consider) the logistical efforts of the filmmaking process, including design, performance and recording. • Although set pieces may well be deliberately positioned at crucial narrative junctures, they nevertheless mark a shift towards a less narratively economical mode of storytelling. It should already be clear that these are not separate from one another, and that a set piece's reflexive and excessive effects are mutually enriching. To put this another way, the savouring of technological or organizational achievement in a given sequence must surely be related to a suspension or disruption of dramatic engagement. In this dynamic, not all film subjects are equal. Consider, for example, the difference between a "still life" table-top composition and an intricately choreographed ensemble fight sequence located in a busy public space. These sequences not only contain different information, but they require (or seem to require, to the best of our general knowledge) different degrees of profilmic effort, labour and/or capital. One is more likely than the other to achieve a "set-piece effect".
Infrastructural phenomena (materials, locations, processes) exist at a scale and level of complexity that already promise something of this effect; their presence in a film could be said to provide set-piece conditions, which may or may not be brought to bear. John Oliver's clip show of Hollywood disaster scenes shows us the most generically familiar case of this in action. Night Moves, as I will discuss, works against this tradition or familiar dynamic. But I begin my analysis with Unstoppable, and its infrastructural "spin" on the subject of filmed railways.
The deep and long-standing relationship between trains and moving images is well known, and there is a strong case to be made for situating Unstoppable in the canon of films which imagine the sensory pleasures and cultural dynamics of rail technology-from L'arrivée d'un train en gare de La Ciotat (Auguste and Louis Lumière, 1896) and The General (Clyde Bruckman, Buster Keaton, 1926) through to La Bête humaine (Jean Renoir, 1938), Pather Panchali (Satyajit Ray, 1955) Café Lumière (Hou Hsiao-hsien, 2003) and Scott's own The Taking of Pelham 123 (2009). 4 I wish, however, to suggest that Unstoppable is alive to a slightly different "structure of feeling," a nuance that is most apparent in the central presence of freight trains as opposed to passenger trains. In this film, the notion of transport has less purchase and relevance than that of infrastructure (a distinction which in turn places Unstoppable in the company of the nearcontemporary RR). 5 As is perhaps to be expected in a work directed by Tony Scott, and one based on the true story of a runaway train, the film features many episodes in which spectacle, profilmic extravagance and suspenseful intensity are brought together; as I will describe in more detail, its context and its characterizing elements make Unstoppable a film that cannot but stage a series of set pieces. However, I will also note the ways in which these "showstoppers" are also compromised by the film's narrative logic and energies.
At a train yard in Northern Pennsylvania, two distracted and complacent engineers oversee the accidental release of an unmanned engine (and train) onto a local mainline. The air brakes are disabled, and the vehicle is carrying hazardous chemicals and a considerable amount of diesel fuel. Meanwhile, in the fictional town of Stanton in Southern Pennsylvania, an experienced engineer (Frank, played by Denzel Washington) and a young conductor (Will, played by Chris Pine) embark on their first journey together and soon learn that they are on collision course with the runaway train. Within a little over 10 minutes, Unstoppable establishes two threads of inevitability: first, that the escaped train will cause extraordinary damage unless it is halted by human ingenuity; and second, that Frank and Will are inescapably obliged to contend with this threat. (The film's first third moves between sites in the north and south, the train and the heroes, and is what might be called a textbook example of crosscutting as an expository narrative convention.) The increasing and relentless power of the train, coupled with a trajectory which is entirely predictable-to us, to Frank and Will, and to the various groups of onlookers and active players in the film-establish extremely narrow geographical and dramatic parameters for Unstoppable. Unless the film abruptly disowns this main narrative thread in favour of a fuller engagement with its loosely sketched subplots (Will is temporarily estranged from his wife, Frank has failed to properly mark his daughter's birthday), we know what we will get: a series of dramatic but inconclusive events precipitated by the train's onward rush, followed by a climactic disaster or heroic intervention.
Of the many instructive and inevitable comparisons to be made between Unstoppable and Speed (Jan de Bont, 1994)-including what Richard Dyer describes in the earlier film as "the exultation of mastery of a machine" (2000,19)-the most significant is surely the simple fact that the rogue bus in Speed is populated with a range of sympathetic characters whose subjective experience of the events complements the story of the two main drivers. Unstoppable largely forgoes this mechanism for sympathetic investment; Frank and Will are alone together. However, there is an exception early on in the film, when a group of schoolchildren ride on a train as part of an educational trip about public safety on the railroads. When the rogue vehicle is let loose on the same line, the school-train engine driver is instructed to divert his carriages into a siding and is visibly terrified. The children, oblivious to the threat, are thrilled, their faces tight up against the window as they witness the near-miss collision, presumably interpreting this episode as part of the regular excitement of train travel. An ungenerous reading of this scene and its function in Unstoppable would be to claim that Scott is indulging in cheap emotional manipulation, thrusting innocent and playful youngsters in the way of an uncontrollable, deadly threat. A critical response to this accusation might be to ask why, if the purpose is to elicit intense affective engagement, Scott uses this strategy so early in the film (approximately 20 minutes in, following a "plant" during the opening credits) without returning to it, and includes no characters who are granted any individual agency or currency. I would argue that, by quickly introducing and then summarily resolving this familiar generic model of innocent-civilians-as-potential-victims, Unstoppable clears the way for the prospect of a different kind of film-in which the railroad claims our attention as something not reducible to a transporter of people.
Connie (Rosario Dawson), the yardmaster responsible for the reining in of the escaped train, advocates derailing it while it speeds through rural space, as the best chance of avoiding a fatal collision in the towns further along the line; she argues the case with Galvin (Kevin Dunn), who oversees the freight company's commercial interests and who refuses to countenance the sacrifice of so much freight and rail hardware. 6 Connie makes her case with compelling and evocative language: "We're not just talking about a train here, we're talking about a missile the size of the Chrysler Building." As if to corroborate this metaphor, the next scene we see is one in which the train, for the first time, violently collides with something on the track. This is unlikely to be a moment of genuine narrative intrigue, for the reasons I have already suggested; we can be almost entirely certain that the relentless onward motion of the train will not be threatened 30 minutes into the film. The qualities and effects of this set piece, then, are more rooted in its logistical choreography of the moment than its suspenseful currency.
The scene in question is the first time in the film that the news media is physically present to the train's escape. Indeed, its first shot is of an outside-broadcast van hurrying to the rail crossing of a small town, which the onscreen text identifies as Findley. A reporter argues with a State Trooper guarding the tracks, demanding to know why so many crossings in the region have been closed. Ned Oldham (Lew Temple), a welder from Connie's train yard sent by her to follow the train, is next to arrive on the scene, as an onlooker. The film then cuts to an unfamiliar man in a truck, driving a van at speed, and so distracted by his in-car stereo that he fails to heed the roadblock and crashes into a large white caravan, nudging it onto the tracks; the sound of neighing horses reveals the contents of the van. Now that all the setpiece components are "in place," the film cuts to an extremely long frontal shot of the oncoming train. In line with the terms I have argued are characteristic of a set piece, we appreciate the purposeful arrangement of the scene's conditions-the ideal location, the confluence of decisions and mistakes, the positioning of the vehicles, the presence of panicked horses-as the scene's subject, and not just as a means to an end. Andrew Klevan reminds us that a film viewer's identification is not necessarily with a character, but may also be with "the process of filmmaking" (2018, 101) and this is such a moment. It is something like reflexivity, but without a strongly individualized authorial voice or control.
A particularly striking shot, of which we see four variations, is taken from inside the caravan, looking outwards through its open hatch. The dimly lit white horses occupy the near and middle distance, while the glowing yellow headlights of the train move directly towards us in the far distance. The horses, though we almost certainly don't want them to suffer, are positioned by the film as cargo which has strayed onto the wrong circuit; a touch of happenstance, even a trace of the carnivalesque, crosses paths with the non-negotiability and merciless predictability of an onrushing freight train. The feat of production-for we surely know that this was a considerable on-set challenge-expresses the infrastructural conditions of the event.
The most technically complex set piece in Unstoppable, occurring a short time later, is understood as such diegetically, which is to say that characters in the story world acknowledge it as a bold, extravagant and largely impractical logistical manoeuvre. These include Connie and her control-room colleagues, media commentators, local onlookers, and the train company employees who gather to watch media coverage of the event. As an alternative to derailment, Galvin has designed a plan whereby an engine joins the track ahead of the runaway train, and brakes to retard its momentum, thus allowing an accompanying helicopter to lower a driver towards the empty cabin. Connie is incensed at the absurdity of the plan-'This is Galvin's strategy?!' she fumes-which does indeed fail; the young man lowered onto the train is struck unconscious, and the engine driver tasked with obstructing it (a veteran and friend of Frank) is killed. The film is of course not coy about displaying the immense assemblage of trains, helicopters and police cars, and some extraordinary telephoto cinematography squeezes onto a single plane these different vehicles travelling at different speeds. 7 What is surprising is that "Galvin's strategy" is treated by the film as excessive and indulgent, an outgrowth of corporate greed and capitalistic inhumanity. Scorned and resented by Connie and Frank, consummate professionals utterly fluent in railroad matters, the plan is a spectacular failure. On the strength of Tony Scott's reputation as a maximalist, we would not expect his film to provoke questions about the appropriateness of excess, but that seems to be the case in Unstoppable, whose actual climatic resolution is relatively straightforward (Frank applies manual brakes to the train carriages, and Will jumps into the engine's cabin).
Elevating infrastructure from a setting to a subject allows Unstoppable to present a surprisingly subtle and complex reflection on what it means to arrange spectacular action as concentrated events in and amongst municipal structures and energy networks. More so than many generically related films (of the kind sampled in Last Week Tonight), Unstoppable retains a sense of infrastructure's "bulky and boring" nature (Peters 2015, 30), its resistance to the rhythms of suspenseful and eventful narrative.
Night Moves, as I will explore, finds a different but similarly distinctive way to reconcile infrastructural diffusion with concentrated narrative action. In it, a large dam is targeted by environmental activists. But is there not something retrograde, some might argue, about fixating on these physical sites of largescale engineering and transport in our age of "liquid modernity"? In his book of that title, Zygmunt Bauman argues that the late twentieth century saw the eclipse of "bulk-obsessed modernity, 'the larger the better' kind of modernity, the 'size is power, the volume is success' sort. That was the hardware era, according to Bauman; the epoch of weighty and ever more cumbersome machines […] of ponderous rail engines and gigantic ocean liners" (2012, 113-114). Bauman's thesis, though, seriously underestimates environmental and ecological dimensions of social change, and this may be why he too swiftly dismisses the continuing vitality of "heavy modernity" in the contemporary world.
A more flexible and adaptable conception of infrastructure and history is offered in Rosalind Williams's influential "Cultural Origins and Environmental Implications of Large Technological Systems" (1993) in which she traces the value of connectivity through Enlightenment traditions and into European and American modernity. "In the spatial arrangements of the modern world," Williams writes, "there is constant tension between the political and economic organizations of space-between the tentacular lines of economic power connecting cores and peripheries, as opposed to blocks representing political power" (384). For Williams, infrastructural sites and networks are manifestations of our growing reliance (for better or worse) on "connective systems rather than on local sources of power" (395). If we are evermore networked, we are evermore obliged to observe and understand how these networks are physically sustained.
A number of recent films can be said to have taken up this challenge in some way or other, including Behemoth (Liang Zhao, 2015); Western (Valeska Grisebach, 2017); The Island of St. Matthews (Kevin Jerome Everson, 2013); and Cities of Sleep (Shaunak Sen, 2015). Such films compel us to see and look at infrastructure, as if cinema can effectively resolve (or respond to) the "play of surface and depth," of hypervisibility and invisibility, which Patricia Yaeger describes as its perpetual condition (2007,16). Akhil Gupta, meanwhile, writes of the arresting, anti-progressive energy of infrastructures, which "fix space and time because, once finished, they are hard to reverse" (2018, 63). The struggle against this stubborn permanence (or unstoppability) is a feature of both films addressed in this essay, even if their most salient contributions to our imaginative understanding of infrastructure have less to do with political responsibility, and more to do with cinema's channelling of networks into sequences of felt intensity.
We feel that intensity, I have suggested, partly because we register (I do not know how consciously) profilmic effort. What can film theory tell us about this effect? Although he does not directly address the question of set pieces, John David Rhodes (2012) in his article "Belaboured" works through some cognate questions concerning style, labour and effort. His starting point is a conundrum: is style that which is typical of a work, or exceptional within it? "Style might feel more palpably dense in one moment in the film," writes Rhodes, "but that density would depend-vitally-on the lighter, more transparent moments of the rest of the film" (48). He proposes labour and effort as a way out of this conundrum, or potential tautology: What if, instead of thinking about style as something that an artwork has, we thought of it as something that artist or artwork does? That is, what if we thought of style not as property, but as labour? We know that works of art get to be what they are because some labour has been expended in producing them, but what if we place the emphasis exactly here-on the work of the work, and on the making visible of that work as style? (48) Set pieces seem to be an ideal case study for exploring in practice what Rhodes proposes in theory. These are passages in which the effortfulness, if not the laboriousness, of filmmaking can be at its most readily apparent. And yet, while Rhodes looks to sketch out the problem of how style and labour engage more generally across film, set pieces perhaps mark the points at which such a general give-and-take slips out of balance-when effort becomes too visible and tangible, when it becomes excessive.
Cinematic excess is a concept that is often filtered through Kristin Thompson's article on the subject. In "The Concept of Cinematic Excess" Thompson (1977) enters into dialogue with Roland Barthes's essay on Eisenstein's Ivan the Terrible, Part 1 (Sergei Eisenstein, 1945), "The Third Meaning," published at the beginning of the decade in Cahiers du cinéma, and included in Stephen Heath's edited and translated collection of Barthes's essays, Image Music Text (1977). Both Thompson and Barthes are interested in how certain moments in film resist analysis or attribution to meaning. In Barthes's terms, an image may operate at an informational level, a symbolic level, and then-more unusually-at a third level, an "obtuse meaning." He writes rather beautifully about this: "I read, I receive (and probably even first and foremost) a third meaning-evident, erratic, obstinate. I do not know what its signified is, at least I am unable to give it a name" (53). Barthes's conception of the obtuse meaning as something that "declares its artifice but without in so doing abandoning the 'good faith' of its referent" (58) is particularly pertinent to my understanding of the set piece, and what I see as its curious position between narrative function and extra-narrative invocation.
Crucial to Kristin Thompson's engagement with Barthes's essay is her sense that his discussion plays fast and loose with the notion of meaning. For Thompson, it is misleading for Barthes to talk in terms of a third meaning when describing precisely those elements which do not participate in the creation of meaning. She quotes a passage in which Barthes suggests that excess need not compromise meaning: At one point, Barthes claims that excess does not weaken the meanings of the structures it accompanies: 'if the signification is exceeded by the obtuse meaning, it is not thereby denied or blurred.' This seems doubtful, however. Presumably the only way excess can fail to affect meaning is if the viewer does not notice it. (55) This a very telling passage. Barthes writes of a mode or effect in which a passage exceeds meaning without abandoning meaning, and Thompson complains that meaning cannot be left unaffected by excess. Drawing on this dialogue in the context of the set piece, I would argue that such sequences speak more to Barthes's imagination of an almost renegade expressiveness. Yes, set pieces bear excessive qualities which affect a film's meaning (the long duration of Thornhill's escape in North by Northwest, the heightened material realism in Way Down East, etc.), but not in the same manner as other meaningful aspects of form, such as particular edits, camera movements or costume details. A set piece is more diffuse, and its meaningfulness as a set piece is only narratively salient in quite general terms (though it may, of course, contain any number of pregnant details). In describing his reception of a third meaning in one moment from Ivan the Terrible, Barthes does in fact try to itemize its traits (or "signifying accidents")-details of physiognomy, costume and make-up-but concedes that this is a thankless task, as if such a mode of reading would draw us too quickly back to second-level symbolic interpretation. His writing gets us closer to the spirit and character of the set piece when he writes of the third meaning in holistic terms, as a confluence. It cannot be subdivided into technical elements; it is "expenditure"; it is "on the side of the carnival" (55).
A brief example will help to demonstrate how this can play out in film interpretation and analysis. In "Symbolic Blockage" Bellour (2000) undertakes an exhaustively detailed dissection of the aforementioned crop-duster set piece in North by Northwest. Through precise discrimination between different shot types and functionssupported by diagrams which help to demonstrate the importance of proximity and orientation throughout the sequence, and tables chronicling the "means of locomotion" at various points in the film-Bellour provides us with all the evidence necessary to appreciate the remarkable organizational and dramatic achievement of these 10 minutes. For the scene to be understood as a set piece, however, I propose that there would need to be a slightly different interpretive and analytical approach, one which considers more fully the singularity of this sequence in the film's broader scheme (though Bellour can hardly be criticized for integrating this scene analysis into a holistic reading of the film), and a critical reflection on its communication of effortfulness and orchestration. For example, Bellour works towards a very developed interpretation of the airplane as "the only irreducible element within a paradigm in which all the other elements refer to each other according to a strictly hierarchized network of correspondences," but his account does not register the (excessive?) profilmic effort required to incorporate an airplane into a dramatic sequence, and all the attendant challenges of design, direction, cinematography and editing (179), akin to those already explored in the case of Unstoppable. I believe that this logistical energy is part of the airplane's significance-alongside its undeniable symbolic potency. 8 Boldly denying us this kind of symbolic potency seems to be one of the major structuring principles of Night Moves, its own way of drawing our attention to the relative eventfulness of infrastructure. As an audience, we are primed to witness (and savour?) the detonation of a dam, but denied visual access at the crucial moment. Both the film and the characters it narrates reflect on not just on the arrangement of an infrastructural event, but on the relative merits of presenting one at all.
Before Night Moves, Kelly Reichardt made Meek's Cutoff (2010), a film which begins with a series of shots of a group of people (including their cattle and wagons) and water. At first, they're crossing a river; carefully, undaunted, without triumph or urgency. The film then begins a series of shots in which individualized members of that group make use of the water, collecting some in a bucket, delicately offering some to a pet bird, washing clothes. The arrangement of this opening leaves things tantalizingly vague about the interval between the crossing of the river and the harvesting of the water. Is the collection of water part of the same event, so to speak, as the crossing of the river, or have we leapt forward to a new point in time and space? This is not a trivial or pedantic question to ask. Seeing how and where water comes and goes is of acute narrative importance in Meek's Cutoff, and its opening sequence registers this as an intriguing formal problem, too.
The opening of Night Moves offers a meaningful variation on that sequence. It likewise begins with people at the water's edge, but now, rather than wade through it or collect it up, the characters take up a stance which is almost inscrutable. The first shot is of a discharge pipe at a dam; timed perfectly to let us see the pipe for a couple of seconds before it starts to eject water. The film then cuts to Josh (Jesse Eisenberg), leaning on the railings of some sort of municipal jetty, staring down, presumably at the pipe. A point of view shot confirms this. A fourth shot then reveals another character, Dena (Dakota Fanning). The two then walk to their car; that they clamber through a hole in a fence is at least some clue that they have been somewhere they shouldn't. What they were doing is unclear, and is only confirmed later on in a very general sense. The subject of infrastructure in Night Moves is an important factor in the distinction between these two openings, a distinction made richer and more telling by the fact that both films are about the same region of the USA (the Pacific Northwest). The ecological conditions have become less visibly and physically determining of the characters' actions, but not less important.
If the eventlessness of infrastructure poses problems to narrative filmmakers, Night Moves offers a curious response to that challenge. It is a film about a small, three-person gang of radical environmental activists who conspire to detonate a dam in Oregon. The film draws on a narrative blueprintthat of the heist film-which absolutely organizes itself around a central climactic event, and which in many films is realized as a set piece. But heist films also presume on the part of their characters an expertise in how things work normally: hierarchies, security systems, routines, spatial arrangements. Yes, the goal is to manipulate or circumvent those patterns in one triumphant "job," but considerable energy goes into mastering them to begin with. Night Moves builds on this aspect of the genre by attending to the circulatory networks within which the destructive event takes place. This includes the dam in the opening sequence, the fish in the reservoir about which Dena (played by Dakota Fanning) talks very knowledgeably, the region's golf courses which are referred to-angrily-as a criminal siphoning off of water supplies; Dena even works at some sort of luxurious physical-health retreat, complete with plunge pools, whose water source we may or may not choose to question. It is to vernacular, local and regional details such as this that Reichardt seems most attuned. 9 In what is perhaps the first clue to the film's reflexivity regarding environmental representation, Josh and Dena attend an amateur film screening of an activist documentary about ecological catastrophe. The film appears to have been put together from stock footage, and with little nuance, or aesthetic or political imagination; images of oil derricks and smokestacks are pitted against images peaceful street protests, and the film is "rounded off" with a whole-Earth icon. 10 Visibly unimpressed by the screening, Josh is barely more interested by the question-and-answer session with the director, Jackie Christianson, which begins straight after the film. Dena asks the inevitable question; what is to be done? More specifically, she asks whether the filmmaker has a "big plan" for averting environmental disaster. The answer is no. Christianson explains her preference for "small plans" over big ones, as Josh watches her talk with an almost surgical stare. First-time viewers of Night Moves may not yet realize that Josh and Dena have already committed themselves to a very big plan indeed, but they are unlikely to miss that the film is explicitly staging a debate about rhetoric, image and action-and that Josh, in particular, is coldly impatient towards calls for undramatic change. As will transpire over the next hour of the film, his commitment and dedication focus resolutely on a concentrated action, what the filmmaker Christianson describes a little dismissively as "one-big-plan thinking".
At least as significant as the contrast between these philosophies is the manner of their presentation and deployment by way of Reichardt's creative choices. Most crucially, the detonation of the dam is not, as I have mentioned, registered visually; it happens offscreen, during a long take of the saboteurs driving away in their van. The terms used by Katherine Fusco and Nicole Seymour to describe this have an interesting bearing on the question and absence of set pieces in Night Moves. "After the boring uneventfulness of Reichardt's previous films," they write, "it's a bit surprising when something explodes; when the trio's bomb goes off. But Night Moves pointedly frames this explosion as being both too late and ineffectual" (2017, 83). As Fusco and Seymour point out, no "specific conspiratorial network ultimately materializes," (73) and this is one of the film's main distinguishing features from the New Hollywood paranoia films with which it has clear affinities. Night Moves achieves its particular insights by "obviating the typical cathartic moment" (73) we find in movies such as Chinatown (Roman Polanski, 1974), The Parallax View and The Conversation (Francis Ford Coppola, 1974).
But it does more than deny us a set piece; characters in Night Moves themselves reflect on whether events can ever mean anything in the context of infrastructure. A few minutes after the off-screen explosion, there is a "morning after" sequence in which Josh sits on the periphery of a conversation between housemates (he lives on a sustainable-agriculture commune) as they react to online reports of the dam's destruction. He feigns ignorance of the attack, but is visibly shaken when he learns that a man is missing as a result. The scene's attention is split between Josh's anxiety (which is mostly silent and private) and an adjacent conversation between Sean (Kal Lennox), who oversees the farm, and Dylan (Logan Miller), a younger man. Sean and Dylan debate the efficacy of the attack, and their exchange is another a fascinating passage of ediotorializing on the part of Reichardt and screenwriter Jon Raymond (deliberately echoing the film-screening debate analysed above), bringing into question the relative validity of focusing our attention on concentrated events at the expense of a so-called bigger picture. In short, Sean sees the sabotage as a cheap and ineffective stunt, while Dylan sees it as a step in the right direction: "Someone's got to start somewhere, you know?" From a dramatic point of view, Josh's distress about the potential fatality is, I would suggest, "enough" for the scene. But the back and forth between Sean and Dylan alters its stakes. Given the fact that Night Moves is about the attack on the dam but also hesitant about its central role, we have good reason to interpret this conversation as a reflexive one about modes and priorities of representation. And indeed Sean finds himself resorting to aesthetic as well as political terms to question the attack: "One dam, who cares, that river has ten dams on it," he complains. "I'm not interested in statements, I'm interested in results." "Seriously, you don't call that results?" asks Dylan. "No," replies Sean. "I call that theatre." What does he mean by theatre? A contrived intensification of action and representation; something whose potential to cause and affect something else is limited, and secondary to its immediate appeal; an ostentatious production of effects. A set piece. Challenged to put forth his own solution, Sean says "look out the window-it's a lot slower but it makes a lot more sense" (a line which is significantly not accompanied by a cut to a shot offering this view). Night Moves thus sustains a productive tension between event and process and makes this tension apparent in its dialogue and its form. The film's main characters intervene in their infrastructural network by way of an actual set piece, so to speak, but Reichardt refrains from showing us this, fostering doubts about its validity, doubts which are articulated by Sean. Where, when and how do we look for infrastructure, intervene in it, and tell stories about it? This is not, of course, a challenge limited to film. The paintings and photographs of Charles Sheeler, for example, of railroad cars, smokestacks and grain silos, are characterized by what one writer calls a "bewildering directness of vision," (Maroney 1991, 32) coupled with an ability to see such objects as "vessels of energy." (37) In short, the challenge to which Sheeler responds is one of recognizing the material presence of infrastructural sites and structures while not allowing them to settle into a (misleading) stability or locality. This essay has discussed in detail two films' similarly deft approach to infrastructural subject matter, and has also gestured towards a number of other works of contemporary cinema which are chronicling energy cultures and structures in bold and imaginative ways. Unstoppable and Night Moves are particularly interesting case studies in this field because they veer so close to the tradition of infrastructure spectacle whilst ultimately maintaining a critical distance from it. Neither film completely denies the aesthetic pleasure of energy systems, but both acknowledge the problems that come from building stories and images around that pleasure.
What are the broader implications of this? In "The Politics and Poetics of Infrastructure," Brian Larkin importantly reminds us that governments and states might strategically deploy a "poetic mode" in which "form is loosened from technical function," (2013, 335) and pipes and bridges and dams (for example) are liable to be used to generate a false sense of efficiency and sociotechnological fluency, regardless of how substantial the provision may in fact be. This is why, argues Larkin, we should not accept too readily those definitions of infrastructure which assume its invisibility, and must instead be alert to the "range of visibilities that move from unseen to grand spectacles and everything in between" (336).
I believe that our alertness to this range of visible forms must be combined with a sensitivity to nonspectacular qualities, too. It is not sufficient to mark if and when we see infrastructure, and we must instead explore how the durational and locational specificity of the film-not to mention its general tendency towards human-scaled actions and interactions-makes something telling of the extraordinary vastness of power and energy networks. Infrastructural set pieces offer a revealing test case for such ideas, because they allow (and invite) us to feel the difficulty of containing a time, place and action within a narrative's economy-and make a virtue of such difficulty.
for example, their emergence in the French and Iranian new waves), the conditions of privacy and independently controlled journeys are fairly constant. The significant differences between urban subway trains, freight trains and intercontinental passenger trains, meanwhile, and the contrasting scenarios they afford, make it difficult to conceive of trains as a graspable cinematic motif or narrative convention. 6. Galvin is very much the villain of the film, even if we appreciate that his capacity for judgement and initiative is strongly curtailed by the pressures of shareholders. Throughout Unstoppable, the hierarchies in play have strong echoes of the Hollywood war film as it is interpreted by Steve Neale (1991). 7. The combination of different transport technologies in a single sequence, an effect which can only be achieved through considerable logistical effort, is a significantly common feature of set pieces, and one whose uncanniness warrants more study. The influential car/train chase in The French Connection (William Friedkin, US, 1971) is just one notable example. 8. Hitchcock is a filmmaker whose work registered infrastructural locations and architecture fleetingly but not insignificantly; the looming gas holder in the opening of Marnie (US, 1964) and Norman Bates's lament about the "moved highway" in Psycho (US, 1960) are particular highlights. Murray Pomerance (2013) is alert to such details. See, for example, his passage on the opening of Saboteur (US, 1942), and its array of subtle but telling environmental details (20-21). 9. For a more sustained discussion of film narrative and regionality, see Adam O'Brien (2013). 10. Another important quality of this film-within-thefilm, and something which sharply distinguishes it from Night Moves, is its geographical vagueness (at least as far as we can see), as exemplified by the closing image of a globe. It is reasonable to suggest that Dena, Josh and Night Moves would deem the local of local specificity a weakness in the work's political vision.

Notes on contributor
Adam O'Brien is Lecturer in Film at the University of Reading. His research sits at the intersection of film aesthetics and the environmental humanities, and he has published essays on themes such as scale, location, materiality and infrastructure. O'Brien is the author of two books: Film and the Natural Environment (2018) and Transactions With the World (2016).

Disclosure statement
No potential conflict of interest was reported by the author.