3I should state from the outset that my sympathies are aligned with the first group of reviews. Roma’s politics, I shall argue in this article, is a result of its dialectics, oscillating as it does between the epic and the domestic, spectacle and everydayness, the monumental and the minute, the larger-than-life and the life-sized, foreground and background, and macro and micro histories. Of particular importance for my analysis will be the concept of the event, since, underpinning these polarities is the idea of eventlessness as encapsulated in the small, the mundane and the quotidian on the one hand; and the eventfulness of the extraordinary, the spectacular and the historical on the other. At the same time, the concept of the event provides an adequate lens with which to account for the heated debates Roma sparked as a Netflix film with restricted theatrical release. Here, “intimate epic” assumes another meaning related to the fact that, while the film was conceived for the largest possible screen, it was predominantly relegated to domestic viewing. To properly account for questions of scale in Roma, aesthetics and representation must thus be examined alongside modes of reception and spectatorship.

4Much of the discussion surrounding Roma attributed its politics, or lack thereof, to aspects of narrative representation and character psychology. I want to suggest that Roma’s politics is to be found, instead, in its aesthetic interplay of scale and the monumental. Commonly understood in terms of large-scale objects designed to memorialise and glorify regimes, nation-states and doctrines, monumentality is often deemed aligned with a conservative and authoritarian project. But what happens when the monumental is applied to objects and subjects existing outside the remit of its scale? What are, in other words, the implication of Roma’s adoption of a “panoramic scale often reserved for war stories,” to cite one reviewer of the film (Dargis), to open up the world of a maid living on the social fringes of 1970s Mexico? What should we make of the fact that this housemaid takes centre stage in the film while history figures as the backdrop of her story?

5Drawing upon Jacques Rancière’s concept of “dissensus,” I argue that Roma’s politics stems from an aesthetic reframing that rescales the distribution of the visible and unsettles commonly held correlations between form and content, subject and genre, and aesthetics and representation, especially as relates to the categories of the everyday and historical narratives. This leads me to examine the film in terms of an interpenetration of ordinary and extraordinary events that can be, in turn, productively mapped on to the categories of micro and macro histories. I then move on to consider Roma as a “historical” film in its own right owing to the debates it precipitated concerning old and new modes of film viewing. Featuring compositional and staging choices that hark back to a 1950s widescreen cinema, when horizontal film formats and screens emerged, among other things, as a response to the popularity of television, Roma is, paradoxically, intimately bound up with the rise of streaming platforms and the intensification of domestic viewing. The historical memories the film preserves thus also include the histories of cinema understood as a mutable screen medium.

Cinematic Monuments

7As regards time, the idea of films as “monuments” can be likened to both extremely long running times and extremely long production times. Frequently, though not exclusively, such elongated times are conflated, lengthy filmic duration figuring as the formal constitution and logical result of a mode of production that lasted for a long time, sometimes a lifetime. Claude Lanzmann’s Shoah (1985), for example, a documentary about the Holocaust that lasts 9 hours and 23 minutes (edited out of 350 hours of footage), was completed in the space of 11 years (1974-1985). Gregory Markopolous’s Enianios (1947-1992), for its part, comprises footage from the director’s nearly 100 films spread over 80 hours of projection. Placing it alongside Jacques Rivette’s Balzac-inspired, 13-hour Out 1 (1971) and Jean-Luc Godard’s 4h30min Histoire(s) du cinéma (1998), Richard Suchenski notes that these films “use duration to resist the industrial structures cinema is normally dependent upon” (5). They thus must be viewed as the cinematic equivalent of “cathedrals,” that is, as “monuments to the imagination that promise transformations of vision, selfhood and experience” (6). In this iteration, cinematic monumentality would seem to depart from its architectural counterparts, normally understood, according to James Osborne, “as directly correlated with power and with the social and political control of commoners by elites” (Osborne 5). Existing outside the confines of mainstream cinema, the films mentioned above instead radicalise conventional running time as a subversive act of defiance.

8Yet mainstream cinema has also produced its monumentalisms, often in the guise of historical figurations. As Marcia Landy tells us in her edited book The Historical Film (2001):

Monumental history as purveyed in the cinema has certain defining characteristics. In its use of narrative it relies on a vision of the past during moments of crisis and heroic conflict, and it reveals a penchant for the actions of heroic figures, such as Napoleon, Elizabeth I, Rembrandt, and Louis Pasteur. These figures come to define an age, and their actions are considered as models to be emulated. (Landy 3)

9Across cinema history, monumental history finds a special place in the Hollywood epic film. Although, as Angelos Koutsourakis cautions us, epic cinema “is an exceptionally expansive umbrella term that covers many and diverse film practices” (Koutsourakis 51), its 1960s Hollywood incarnations are a fertile terrain for an exploration of monumentality in both its spatial and temporal expressions. Here, impressive purpose-built settings and panoramic landscapes shot on large-scale film formats are translated into long running times. As Vivian Sobchack summarises: the epic film “constitutes its historical field as literally and materially […] extended and expanded. An excess of temporality finds its form in, or ‘equals,’ extended duration: films far longer than the Hollywood norm. Correlatively, an excess of space finds its form in, or ‘equals,’ expanded space: Cinemascope, Cinerama, Superscope, 70mm” (Sobchack 37, emphasis in original).

10At first glance, these preliminary considerations would seem unrelated to Roma. Although its running time of 135 minutes and 108-day shoot are above the average, they are nowhere near the films discussed above. Roma’s elegant style, black-and-white cinematography and art-house credentials would equally set it apart from the Hollywood epic’s flamboyant extravagance. Nevertheless, Roma is undeniably monumental in at least two respects. Whereas its long-take aesthetic can be likened to “slow cinema” owing to its predilection for domestic chores and moments where “nothing happens,” it is also the site of a meticulously orchestrated mise-en-scène set in reconstructed streets and buildings emulating the Mexico of 1970s. This is not the ancient history of Roma, the city, but it is a painstaking, indeed spectacular historical rendition of a neighbourhood: Colonia Roma, in Mexico City. At the same time, Roma’s monumentality is the result of its large-scale format, one whose pronounced horizontality (aspect ratio 2.39:1) is elongated further through visual composition and camerawork, including a preponderance of circular pans and tracking shots.

what Brody wishes for is that an authentic, politically-engaged cinema would resolve the ills of the world in representation. If the figure of the servant has been treated in life (and in older representations) as a “non-person,” as Erving Goffman has written, then her personhood must be restored in representation in order to re-animate her. (Skvirsky 3, her emphasis)

13But as Skvirsky goes on to note, for some films, Roma included, “to ‘give voice’ in representation would be to falsify the nature of the institution of domestic service”; hence, “far from being a symptom of Cuarón’s blind spot, Cleo’s inscrutability, her silence, might be read as a calculated choice in the service of a different kind of intervention in politically fraught terrain” (4).

14I would add that informing Brody’s appraisal is a quite specific, normative modality of representation that can be related to what French philosopher Jacques Rancière terms the “representative regime.” Significantly, Rancière notes that one of the main postulates regulating this regime is a “dependency of the visible on speech.” He writes:

In it the essence of speech is to make seen, to order the visible by deploying a quasi-visibility wherein two operations are fused: an operation of substitution (which places ‘before our eyes’ what is removed in space or time) and an operation of exhibition (which makes what is intrinsically hidden from sight, the inner springs motivating characters and events, visible). (The Future of the Image 113)

15Roma’s sin, for Brody, is that it does not make visible the inner springs motivating the character of Cleo via speech, whether through voiceover, dialogue or flashbacks. In its emphasis on Aparicio’s sheer physical presence and demeanour, enhanced by the fact that the nonprofessional did not receive acting training, Roma is instead attuned to Rancière’s “aesthetic regime.” This, Rancière maintains, is “a regime of immediate identity between the absolute decision of thought and pure factuality” (The Future of the Image 122).

16But Roma also upsets the notion of event in a twofold manner. It does so, first, through a durational focus on activities that are not normally deemed as much from a narrational standpoint, including Cleo’s domestic chores, examined below. Second, the film refuses to “make visible” the springs motivating historical events depicted onscreen. Thus, Brody also laments that, as with Cleo’s “cavalierly vague,” voiceless portrait, the film’s depiction “of the public and historical events in which she becomes entangled, and which he [Cuarón] dramatizes, is similarly flattened and obscured.” Citing the 1971 Corpus Christi massacre depicted in the film (more of which shortly), he asks: “What are the students protesting? What are they advocating? Why do they seem to threaten the regime?” (Brody)

17I would ask instead: what does this film’s historical scope tell us about its politics? Does it matter that it equates the eventless rhythm of the everyday with the eventfulness of history? Whereas monumentality is a dominant trope of historical narratives, including their cinematic iterations as we saw earlier, these often centre on those who purportedly make history. Is it relevant, then, that at Roma’s centre is an Indigeneous maid to whom the figuration of history is subordinated and not the other way around? To explore these questions is to veer the film away from a narrow focus on the politics of representation so as to account for aesthetic choices that throw into disarray and modify the very parameters within which such a representation must be assessed. By defying “rules of appropriateness between a particular subject and a particular form” (Rancière, The Future of the Image 118), Roma unsettles expectations regarding both the depiction of marginalised subjects and the figuration of history. In doing so, it reorders the “frames, speeds and scales according to which we perceive the visible” (Rancière, Dissensus 141), a process that, as Rancière maintains, is the very definition of politics.

The Scales of History

18Roma’s conflation of the epic and the domestic, understood as two equidistant extremes of a historical spectrum centred on the notion of event, is striking in light of the fact that these two domains are visualised through recourse to the same technique, the long take, even if its application is different in each case. On one side of the spectrum, there are unbroken shots devoted to studying situations frequently deemed “non-events,” in line with what is now normally defined as “slow cinema.” A quick glance at Roma’s first 20 minutes, in which the protagonist’s routine is established mostly through slow-moving circular pans, substantiates this claim. We see Cleo cleaning the outside patio of the house. We follow her upstairs going into and out of rooms while fetching clothes for laundry. We then follow her downstairs, tray in hand, as she turns off all lights as the household goes to sleep (Figure 1). In the film’s second sequence, Cleo even leaves the frame and the camera remains waiting for her to return, with no human presence in sight. For Skvirsky, this shot illustrates the way Roma overturns the idea of “eventhood,” which in turn foregrounds the surveillant character of the narration, as we are made aware of the unblinking stare of the camera as a felt presence (Skvirsky 9).

19Skvirsky’s analysis is symptomatic of a surge of interest in the concept of the event, as formulated, for example, in the burgeoning scholarship on cinematic time and duration over the last two decades. In her The Emergence of Cinematic Time (2002), Mary Ann Doane draws attention to how cinema’s appearance generated anxiety insofar as its indiscriminate recording upended commonly held definitions of eventhood. For Doane, the excision of so-called “‘dead time’ – time which, by definition is outside of the event, ‘uneventful’” – in narrative cinema speaks to this foundational problematic (Doane 159, 160), one that has been resuscitated more recently thanks to slow cinema. To quote Song Hwee Lim:

Rather than construct a narrative in which many events happen, a cinema of slowness chooses to dwell instead on the interstices between events or moments within events during which nothing much happens. Indeed, it uses so-called dead time to create non-events as events through which a different temporality, meaning, and value can come into being, thereby questioning the notion of “event” or “happening” and unsettling the very foundation of what constitutes a film’s narrative. (Lim 30)

20Among the many “non-events” privileged by a cinema of slowness, domestic activities occupy a special place, starting with the “hyperrealist everyday” (Margulies) of Chantal Akerman’s groundbreaking Jeanne Dielman, 23 quai du Commerce, 1080 Bruxelles (1975). As a foundational work of slow cinema (even if not designated as such at the time), Jeanne Dielman proves that domestic service is a privileged site for durational stretching because, as one of the pillars of the everyday, it is otherwise rendered invisible, deemed a non-event, indeed often not even considered work. Granted, Jeanne Dielman and Roma differ significantly when it comes to their rendition of domesticity and everydayness, the former characterised by an imperturbably stationary camera and the latter by an endlessly gliding one, among other things. Yet, as the camera unhurriedly pans to follow Cleo going up and down the house, picking up dog’s shit and doing the laundry, the spectator, as in Jeanne Dielman, is graced with time to study the minutiae of small-scale gestures that get no time in conventional narrative fare.

21Sparseness and minimalism do not reign supreme in Roma, however. The long take is here also the stage for a densely layered mise-en-scène, meticulously choreographed before a constantly moving camera, that is instead hyperbolic and maximalist. In some cases, as in the reconstruction of the Corpus Christi massacre, these are well-known historical events. Although Roma refuses to explain its context in a didactic manner, there are allusions to the volatile political climate of the time. The authoritarian Institutional Revolutionary Party (IPR), in power since 1929, had started losing support and become the target of student protests which were, in turn, violently repressed, as in the deadly 1968 Tlatelolco massacre in the Plaza de las Tres Culturas. Thus, in an otherwise irrelevant scene, Paco mentions at breakfast table that he saw a soldier shooting at a student. Likewise, when we are first introduced to the character of Fermín, he tells Cleo that he “owes [his] life to the martial arts,” which turns out to be a reference to Los Halcones, whose military training we see in a later scene, when Cleo travels to the outskirts of the city to tell Fermín she is expecting his baby. Los Halcones was a paramilitary group supported by the Mexican state and backed by the US. It aimed to supress social unrest by enlisting unemployed and working-class men in need of income. On June 10, 1971, in what came to be known as El Haconazo, or the Corpus Christi massacre, Los Halcones brutally clashed with a student demonstration, leaving more than a dozen dead and hundreds injured.

23Unlike the domestic scenes, the sheer scope of this episode, which culminates in an open-air battle sequence involving hundreds of extras, recalls the “eventfulness” Sobchack identifies as typical of the epic film, its operatic visual excess “commonly perceived as an ‘extraordinary’ mode of filmic representation” (Sobchack 27). One could go so far as to say that, in many instances in Roma, it is an eventful mise-en-scène alone, allied with an obsessive eye to period detail, that grants the representation an extraordinary, and thus historical, importance. The scene where Cleo runs after Toño across many blocks on Avenida Insurgentes is a case in point. Cuarón had to overcome the problem that the avenue had in some parts changed beyond recognition. He then had the production designer Eugenio Caballero reconstruct in a six-block-long set part of its façade, especially where it meets the Baja California avenue, complete with the same bridal stores, clothing shops, restaurants and diners that dotted the avenue in the early 1970s. The sequence is shot through with a synchronicity between extras dressed in period attire, old cars driving along, multi-layered sound design and a camera that, via a tour-de-force lateral tracking shot, mirrors Cleo’s pace, gliding along with her and coming to a halt when she stops to cross the avenue (Figure 3). Here, as in other shots in this film, historical eventfulness is created via form rather than content: a virtuosic mise-en-scène that, by dint of its excessive textuality and meticulous attention to period detail, imputes events with historicity.

25As Ginzburg notes (26), this scalar reduction was pioneered by Kracauer in his posthumous History: The Last Things Before the Last (1969), where, unsurprisingly, the German writer recruits film vocabulary to account for the intersecting of micro and macro levels rippling through the “nonhomogeneous structure” of the “historical universe” (Kracauer 127). Such levels, Kracauer goes on, are “both illustrated and explained by the analogous phenomenon of the paradoxical relation between ‘close-ups’ and long shots (shots of ensemble) in the cinematic narrative” (126). Whereas “small-scale histories may be called ‘close-ups’ because of their resemblance to the film shots of this name which isolate and magnify some visual details,” macro histories can be deemed the equivalent of “high-magnitude” and “panoramic views” (111), as the historian must “step so far back from the given data that all the destinies of that people enter his field of vision” (105).

26Kracauer’s filmic historiography can be productively queried in relation to Roma. Whereas close-ups are not entirely absent from this film, they are certainly not abundant – and indeed, even in the intimate and solitary scenes, Cleo is often framed at a certain remove, in long shots. Yet, to the extent that the epic and domestic scenes entail “differences in distance” from the subjects depicted, Kracauer’s distinctions hold sway, given that the “panoramic views” utilised to frame Cleo in public events and situations are wide and mobile enough to allow different phenomena to enter the film’s field of vision.

27To think of Roma as “panoramic” is further relevant in light of its camerawork and aspect ratio. The circular pan and the tracking shot, as well as the widescreen format, are all qualified by this adjective in English. According to Cuarón (see Dillon), the initial plan was to shoot Roma in a square format. It was his long-term collaborator Emmanuel Lubezki, who was set to be the film’s director of photography (DoP) but eventually dropped out, that persuaded him to shoot on the 65mm digital camera Alexa 65, which translates into the horizontal 2.39:1 aspect ratio. Cuarón, who became the DoP himself, recalls that this decision radically modified the conception of the film: “as we did tests, I realized this movie is honouring real-time and space, and here we would have a larger scope in which the characters could flow. I wanted to shoot very wide, and balance foreground and background with each informing the other” (Dillon). In order to highlight the width of the film, the application of gliding camerawork consisting of circular pans and tracking shots was the next logical step, whereas the film’s crisp depth of field was obtained by bringing in extra light sources, many of them erased at post-production.

29© Participant Media / Esperanto Filmoj, 2018.

30There is something anachronistic, almost defamiliarising, in Roma’s utilisation of the widescreen format and depth of field. It calls to mind a long-take director such as Otto Preminger who, as John Belton notes, favoured “horizontal compositions and lateral character and camera movements to explore the horizontal dimension of the frame” (198), while inserting cues that draw the eye “into [the image’s] depth, pulling the spectator’s attention in two directions” (199). We will shortly see that this style raises questions regarding the size of the screen on which the film is watched. Let us note for now that a predisposition for boundlessness also informs Roma’s topographical organisation. Indeed, the film’s title carries within itself geographical connotations, not only in the unavoidable association with Rome, the Italian city (and the ancient, “epic” connotations appended to it), but because it designates a specific location, the neighbourhood of Colonia Roma, in Mexico City. In fact, the film’s geographical organisation was made visible in a Netflix promotional document emulating the Guía Roji, an old guide and map of Mexico City, with the Colonia Roma neighbourhood highlighted within it (Smith).

31Yet Roma does not stay tied to this location. As it unfolds, it travels in an ever-expanding trajectory that widens its horizons while never losing sight of Cleo as its nucleus. As Luiselli perceptively notes:

The film begins with an image of the ground, a patio, where Cleo, the protagonist and beating heart of the story, cleans and sweeps an archipelago of dog shit every day. The eye of the camera makes its way around a house, upstairs, downstairs, the dining table and the kitchen; then in and out of the house, the city sidewalks, its buzzing streets; in and out of hospitals, cinemas, and hotel rooms; then out into the periphery of the city, then farther out into the countryside, and then even farther out to the geographical limits of the country, to the western coastline of the Gulf of Mexico, its tides pulling and waters churning. (Luiselli)

32Roma is always moving and elongating, letting new geographical portions and historical facets enter its purview, its worldview. Its unwavering focus, however, is Cleo, meaning that history and geography in this film are subordinated to the story of this domestic servant, often as the background, quite literally so. To use Rancière’s words, Roma’s political redistribution of the sensible resides in the way it rescales the visibility of history by rupturing “given relations between things and meanings” and inventing “novel relationships between things and meanings that were previously unrelated” (Rancière, Dissensus 141). Another way of putting this would be to say that the micro level is not utilised here, as is often the case, as a way to explicate larger historical forces: the “wholesale situations, long-term developments, ideological trends” characterising a given period which Kracauer defines as the “proper” goal of historiography (Kracauer 118). Rather, these situations, developments and trends are contemplated in Roma inasmuch as they form the concentric ripples revolving around Cleo who, despite her laconic nature, emerges as the pivot of the film’s expansive thrust, an expansiveness that makes sense only when measured from her and for her. Cleo may not have a voice, but she certainly has a world.

Made for TV?

33In her study of cinematic scale, Mary Ann Doane reminds us that it “can be measured in two ways – as the scale of the shot itself (close-up, medium shot, or long shot) or as the scale of the image projected on the screen (height and width of the screen, aspect ratio of the image)” (Doane, “Scale” 72). The interaction between these two types of scales is often mutually reinforcing. To return to Preminger’s long-take style, we may say that its “predilection for monumentality and infinity,” to use Doane’s own words (73), found its perfect fit in the wide cinema screen, with its “illusion of limitless horizontal vision” (Belton 197). Roma did not have the same fate. Blatantly conceived for the “macro,” epic screen of the cinema theatre but distributed by a streaming platform that plays on the smaller screens of televisions, tablets and mobile phones, the film thus warrants a discussion of scale beyond its historical story-world, one that also includes the actual scale of screens and the history of cinema as a screen medium.

34In this discussion, the notion of the event assumes different resonances. From the outset, Roma became an “event” by catalysing the animosity between Netflix and the film industry. It turned into the focal point around which several interrelated debates – old and new modes of spectatorship, film studios vs. streaming platforms, cinema vs. television – agglutinated and multiplied. Among such debates, the collectivity of the theatrical experience as a public event emerged as a heated topic. This was so especially as the film was the target of a concerted ban on the part of the exhibition sector, which, at least in his home country, Cuarón was partly able to turn to his own advantage through initiatives intended to promote the film as an event attached to national sentiment. Just as important, as I hope to show, the publicness of the theatrical experience appears in Roma’s textual makeup, making it therefore an ideal case study for questions of form, scale and screen, and how these variously interrelate with each other.

36Roma became the first Netflix film nominated for the Best Picture category out of a total of 10 nominations. Although it did not take home the most important prize, Netflix’s extravagant promotion – allegedly in the vicinity of $30 million, twice the cost of the film and possibly the biggest campaign ever for a Best Picture contender (Welk) – was for all intents and purposes successful: the film scooped up the Best Director, Best Cinematography and Best Foreign Film prizes. However, the path to Oscar glory was not without its obstacles, as the film set ablaze the feud between Netflix and the exhibition sector. To appease the latter, Netflix uncharacteristically decided to launch Roma theatrically for a 3-week period before the film landed on the platform, rather than its usual, simultaneous theatrical-and-online release. Many exhibitors were still not happy and refused to show the film unless Netflix respected the traditional 90-day “window” between theatrical and other forms of release.

37Things became especially heated in the UK. Roma was awarded 4 major prizes at the BAFTA in February 2019, including Best Film. In an open letter, the CEO of the European cinema chain VUE, Timothy Richards, decried that “BAFTA has not lived up to its usual high standards this year in choosing to endorse and promote a ‘made for TV’ film that audiences were unable to see on a big screen” (Richards). We should pause briefly over this expression and ponder what it entails for Roma. Whereas “made for TV” is seemingly used here as a literal descriptor indicating the medium on which Roma was most often seen, it is no doubt also meant to invoke the small cultural cachet attached to the film genre that goes by the same name and which is often frowned upon because of unrefined aesthetic choices and low production values. And yet nothing could be further from a “made for TV” film than Roma, featuring as it does an intricate mise-en-scène and monumental long-take style. In fact, one could go so far as to say that Roma is the epitome of a “cinematic” aesthetics if by cinematic we mean films that were clearly conceived with the big screen in mind. Indeed, this was the very reason why Netflix promoted and supported this film so lavishly. Roma was to be taken as a milestone in the short history of the platform’s continued efforts to distance itself from traditional broadcast (see Shacklock) and attempts to make inroads into the film industry. Put simply, Netflix was less concerned about where the film was seen than the fact that it looked cinematic enough to garner prizes and awards at traditional film competitions and ceremonies.

39Sobchack’s thoughts on how the epic film’s “historical eventfulness […] exceeds its already excessive screen boundaries” (Sobchack 32), that is, how its historicity is formulated not only on a narrative level, but also at both production and reception stages (often in relation to new technologies), is again helpful. Singling out the 1962 release of How the West Was Won in Cinerama as a germane case in point, Sobchack comments on how its press book memorialises the film by equating “the ‘new era’ of entertainment ushered in by the ‘futuristic’ technology of Cinerama and the ‘promise of the future’ the West held for past Americans” (Sobchack 30). Leaving aside many unbridgeable aesthetic, institutional and generic differences, Roma’s magnitude was similarly augmented by its “historical” reception, which condensed and reflected major shifts in the film business as regards production, exhibition, distribution and consumption practices. At the same time, Netflix’s aggressive promotion campaign – which included a hefty coffee-table photography book whose weight of six pounds concretised the film’s monumental aspirations (Matthews) – had similar memorialising goals in mind. Not only was it designed to laud the “new era” of cinema ushered in by the technology of Netflix; it attempted to legitimise Netflix itself as cinema, i.e. not television.

40To consider the advent of widescreen technologies provides us with some further angles from which to view Roma’s large-scale format and its complicated rapport with the cinema screen. As Belton has argued, the popularisation of widescreen and wide-film processes in the 1950s was inseparable from the goal to make the theatrical viewing experience an “event” in the collective, even participatory sense of the word. Belton writes: “the ultimate source of widescreen cinema lies not so much in the introduction of wide-gauge film as in the development of ‘large-screen’ cinema” (Belton 34), which, in its turn, promised a new mode of spectatorship based on “the notion of ‘audience participation’” (187). Although Belton is careful to stress that the need to reemphasise the publicness of the cinema experience was due to several interrelated factors and not simply the appearance of television, there is no doubt that the enlarged widescreen size was designed to widen the gap between cinema and television.

42These developments have therefore meant that the discrepancy between public and private modes of viewing may not be as big as it once was, although the situation becomes more complex when we turn to the proliferation of smaller screen devices that now populate our homes and on which films are increasingly seen. In fact, when Netflix’s CEO Ted Sarandos was asked about the viewing of Roma, his medium of choice was tellingly not television but the mobile phone, his point being that Netflix and the cinema are not antithetical but a “complement to each other” (Donnelly). This may well be the case, yet I would maintain that Roma is a limit-case audiovisual artefact because of its anachronistic staging and stylistic choices, which struggle on the small screen. Shot after shot in the film, because of their dizzying depth, slow tempo, wide scope, horizontal camera movements and the sheer amount of details in the image, reminds the viewer that this is a film made for the cinema theatre.

43Rather than a perfect fit between the scale of shots and that of the screen, then, what we have here instead is a mismatch that seems to smuggle something of cinema into television and related screen devices. As we saw earlier, this was very much what Netflix was after. Yet I also want to suggest that, whether intentionally or not, Roma carries within itself, that is to say in its textual makeup, a self-reflexive awareness of this scalar incongruity. In this light, the film becomes a historical testimony of the increasing uncertainty surrounding the theatrical space and the collectivity of shared experience, as if a cultural memory of sorts both inhabits and haunts its audiovisual constitution. By way of concluding this article, I turn to two scenes of the film that substantiate this claim.

44The first scene takes place inside a cinema, when Cleo tells Fermín she is possibly pregnant. The whole scene is filmed with a static camera in an uninterrupted take lasting 3 minutes and 20 seconds. This is the Teatro Metropólitan, on Avenida Independencia, a venue still functioning today but no longer as a movie palace. Cleo and Fermín, in the foreground, are seated in one of the back rows, their backs to the camera and to the left of the image; to the right, the screen showing La Grande Vadrouille (1966), in the background, is wholly visible, flanked by four neoclassical columns; in between, the whole seated audience is in focus (Figure 6). This means that throughout the scene the viewer’s attention is pulled into competing directions, even if only the foreground of the image carries narrative importance. In fact, Cleo and Fermín hardly ever glance at the film-within-the-film. Cinema does not seem to be of interest to them. Clearly, they are in the theatre because of the intimacy it offers. Indeed, in an earlier scene, when we are first introduced to the couple outside the same Metropólitan, and Fermín and Cleo hesitate to go in, Adela tries to persuade Cleo that “it is better to play in the dark” (the couple decide to go to a motel instead).

45Yet cinema, understood as a space for ritualistic collective viewing, is of interest to Roma. In this scene, the camera is uncharacteristically stationary, positioned strategically at the back of the cinema as if occupying a seat in one of the rows just behind Cleo and Fermín. The static long take provides the viewer not only with a panoramic view of the theatrical space but also an embodied sense of “being there,” as if the camera were itself a member of the audience, somehow subverting, furtively, the couple’s expectation of complete privacy. As such, the film’s imagery acts as an invitation into what Roland Barthes once described as a “perverse” mode of spectatorship that is “ready to fetishize […] the texture of the sound, the hall, the darkness, the obscure mass of the other bodies, the rays of light, entering the theater, leaving the hall” (Barthes 349). Thanks to the shot’s impressive depth, we can see the rows of occupied seats; notice the patrons kissing, smoking or simply engrossed in the movie; and study the architecture of the theatre down to the columns framing the screen. For its part, the illuminated screen in the background reflexively reminds the viewer of the edges of the frame. Not coincidentally, the aspect ratio of La Grande Vadrouille, filmed in Panavision, is 2.39:1, exactly that of Roma. When Fermín leaves his seat, never to return again, the camera, waiting for Cleo, remains in its place for 1 minute and 20 seconds: the film-within-the-film ends, applauses are heard, the lights come back on, curtains are drawn up over the screen and patrons start leaving the theatre, walking past the camera. Cleo is now alone, even if surrounded by people; the cinema ritual has ended, but her life (and the film) must go on.

46The second time cinema viewing shows up in Roma is when the whole family, bar Sofia, goes to the movies. This scene is preceded by the long lateral shot, described earlier, trailing Cleo as she runs after Toño on Avenida Insurgentes. Narratively, the function of the scene is to show that Toño accidentally catches his father coming out of a cinema, brazenly embraced with a young woman, the episode in turn witnessed by Cleo as a bystander. This cinema, its façade reconstructed for the film, is the former Cine de las Américas, now Auditorio Blackberry, where, incidentally, a special screening of Roma took place in 2018. Cleo and Toño walk in to get tickets. After a cut, the images that follow are all from the film-within-the-film, the 1969 sci-fi Marooned (aspect ratio 2:35.1), a major inspiration for Cuarón’s own Gravity (2013). The images are not shown on a screen within the cinema, however. They take up the entirety of Roma’s aspect ratio and depict two astronauts, whether alone or together, against the immensity of the universe, the Earth sometimes in the background.

47Although at first these images may seem unrelated to Roma’s universe, they accrue significance on further reflection. In a way, outer space appears as the logical culmination of Roma’s quest for monumentality, its predilection for expanding vistas and adamant refusal to be constrained by the frame through compositional horizontality and camera mobility. One could even argue that the sublime imagery of an astronaut surrounded by infinite darkness is the celestial counterpart of the film’s much-commented-upon aquatic climax, when Cleo walks into the sea in order to save Paco and Sofi from drowning. For a moment, all we see in that crucial scene is Cleo engulfed by threatening waves, the fluidity of the water and the lack of terrestrial signs implying an unbounded oceanic expanse stretching beyond the frame. In both cases, space and ocean epitomise and exacerbate the film’s affinities with boundlessness and infinity, suggesting that Cleo’s universe is the universe.

48At the same time, the images from Marooned are never contextualised, given the complete absence of shots locating the film within the space of the auditorium and on the screen. Perhaps this was because the Cine de las Américas has been refurbished to an unrecognisable degree, unlike the Metropólitan (even if all chairs in the latter were replaced for the film). Perhaps this was because another scene in a film theatre was logistically unviable. Whatever the case, the comparison between the two scenes is striking. Whereas in the first we are allowed to study a packed, sumptuous auditorium obsessively recorded by an imperturbable camera, in the second cinema-viewing is gone and only images remain, images which Roma usurps and almost presents as its own. I highlight “almost” because, upon closer attention, we can subtly hear on the audio track sounds and noises such as shushing, coughing, giggling and some indistinctive murmuring in Spanish, all of which, we presume, emanate from the auditorium in which these images are being projected within the diegesis. Yet, because we are never given the sight of the auditorium or the screen, the effect of this aural track (overlaid with that of Marooned) is jarring. With no screen or setting to anchor them, these floating images, filling Roma’s aspect ratio as a citation, a film-within-a-film, appear as a fitting commentary on the current elasticity and malleability of images – their detachment from a stable, fixed, recognisable space and the way they now traverse through a myriad of mediums, screens and devices. On the other hand, the faint murmur of bodily human noise, itself noticeable especially when the film is watched with headphones at home, infuses the visual track with something of a spectral quality, as if the film was haunted from within by the historical memory of a collective audience that remains trapped in the constitution of these images, which, as a result, function as an irresistible metonym for Roma as a whole.