The View From the Train

by Patrick Keiller

  I would like to suggest that film space can offer an implicit critique of actual space, so that looking at and researching films can constitute a kind of architectural criticism. I would also suggest that one can make films (and I suppose I would claim to have done so) that set out to criticise architectural space rather than simply depict it (which, given the marked differences between film space and actual architecture, is much more difficult). Lastly, I would suggest that film2 from the past that depicts urban and other architectural space of its time can offer an implicit critique of similar spaces of the present, and can inform our understanding of the ways in which urban and other landscapes change in time.

  A few years ago I embarked on a project to explore urban space as it appears in films made before the mid 1900s.3 The following paragraphs set out the context for this exploration, and identify coincident periods of transition in the histories of architecture, urban space and film.

  Carrington Street, Nottingham in 2003, with inset from Tram Rides Through Nottingham, Carrington Street, Mitchell and Kenyon (1902)

  Until the mid 1900s, most films were between one and three minutes long, and consisted of one or very few unedited takes. The Lumière company’s films, for example, are typically from 48 to 52 feet long, and last about a minute. They were made by exposing a complete roll of film, often without stopping. Most early films were actualities, not fiction, and many were street scenes or views of other topographical subjects, some of them photographed from moving vehicles and boats. Cinematographers would sometimes pause if there was a lull in the ambient action, or if the view was blocked, but other kinds of editing are unusual. The reconstruction of time and space by joining individual shots together was an aspect of film-making that began to dominate only after about 1907.

  Tom Gunning has called this early cinema ‘the cinema of attractions’,4 a reference to Eisenstein’s ‘montage of attractions’, conceived as a new model for theatre. Eisenstein took the term from the fairground, where his favourite attraction was the roller coaster, the Russian for which translates as ‘the American Mountains’. There is an early Biograph film A Ride on a Switchback (1900, or possibly 1898), which was made by mounting a camera not on a roller coaster, as early films sometimes were, but on a railway engine. A switchback was a railway engineers’ device for negotiating steep gradients with a siding and a set of points, entering by one branch and backing out into the other, so as to avoid the construction of a hairpin bend. Biograph’s film was photographed in mountains near Fort Lee, New Jersey, which one might imagine were the (or at least some) American Mountains. Films photographed from the front of railway engines were known as ‘phantom rides’, presumably because of the sensation of disembodied consciousness they offer. Views from other moving vehicles – trams and, later, cars – are sometimes called phantom rides, but the term seems to have been most specific to the view from the front of a locomotive, which was then seldom encountered in ordinary experience, even by an engine driver.

  As Gunning writes, after 1907 ‘the cinema of attractions does not disappear with the dominance of narrative, but rather goes underground, both into certain avant-garde practices and as a component of narrative films’.5 There is a story that Andy Warhol’s Kiss (1963) was prompted by an archive viewing of Edison’s Kiss of May Irvin and John C. Rice (1896), and whether or not it was, the formal evolution of Warhol’s films – from the hundred-foot rolls of Sleep (1963) and Kiss (1963) to the 1,200-foot rolls of the two-screen The Chelsea Girls (1966) – strikingly resembles that of early film.6 In narrative cinema, phantom rides appear in films noirs, often at the beginning of a film or in title sequences, as in Fritz Lang’s Human Desire (1954), Mike Hodges’s Get Carter (1971), and the car shots in Jacques Tourneur’s Out of the Past (1947), Edgar G. Ulmer’s Detour (1945) – looking backwards – and Robert Aldrich’s Kiss Me Deadly (1955). Since the 1960s the cinema of attractions has emerged from underground, in films and installations by a wide variety of artists and other film-makers, most of them outside the mainstream of Western cinema. Whether in the gallery or in what used to be called art cinema, there is a tendency towards some of the forms of early film.

  Both early films and these more recent examples differ from what became the dominant form in the way they represent space on a screen. In films constructed as montage, space is assembled in time, as an implied continuity of fragments. In most early films, space is represented within a single frame, either static or moving. Early films are also less likely to direct the viewer’s attention to a single subject in the frame: one’s eye can more easily wander in their spaces, and because of this they invite (or even require) repeated viewing. Moving-camera films often create a striking illusion of three-dimensionality, which early film-makers sometimes referred to explicitly as ‘the stereoscopic effect’.

  Between the mid 1900s and the outbreak of the First World War, the spaces and spatial experiences characteristic of industrialised economies appear to have undergone significant transformation. These transitions have been described in a variety of ways: for example, in his afterword to the English translation of Henri Lefebvre’s definitive La Production de l’espace, first published in 1974, but in English only in 1991, the geographer David Harvey quoted a passage in Lefebvre’s opening chapter:

  Lister Gate, Nottingham in 2003, with inset from Tram Rides Through Nottingham, Liser Gate, Mitchell and Kenyon (1902)

  The fact is that around 1910 a certain space was shattered. It was the space of common sense, of knowledge (savoir), of social practice, of political power, a space thitherto enshrined in everyday discourse, just as in abstract thought, as the environment of and channel for communications; the space, too, of classical perspective and geometry, developed from the Renaissance onwards on the basis of the Greek tradition (Euclid, logic) and bodied forth in Western art and philosophy, as in the form of the city and the town … Euclidean and perspectivist space have disappeared as systems of reference, along with other former ‘commonplaces’ such as the town, history, paternity, the tonal system in music, traditional morality, and so forth. This was truly a crucial moment.7

  Harvey had already quoted from this passage in his The Condition of Postmodernity (1990), following mention of ‘the incredible confusions and oppositions across a spectrum of possible reactions to the growing sense of crisis in the experience of time and space, that had been gathering since 1848 and seemed to come to a head just before the First World War’, and noted ‘that 1910–14 is roughly the period that many historians of modernism (beginning with Virginia Woolf and D. H. Lawrence) point to as crucial in the evolution of modernist thinking’.8

  For Harvey, the crisis was one ‘of technological innovation, of capitalist dynamics across space [and] cultural production’. He notes the slightly different emphasis of Stephen Kern who, in The Culture of Time and Space 1880–1918 (1983), offered ‘generalizations about the essential cultural developments of the period’.9 Other writers have dealt with these in detail: for John Berger, ‘The Moment of Cubism’ was the period between 1907 and 1914, and during the period 1900–14 ‘the developments which converged at the beginning of the twentieth century in Europe changed the meaning of time and space’.10 Berger listed these as:

  An interlocking world system of imperialism; opposed to it, a socialist international; the founding of modern physics, physiology and sociology; the increasing use of electricity, the invention of radio and the cinema; the beginnings of mass production; the publishing of mass-circulation newspapers; the new structural possibilities offered by the availability of steel and aluminium; the rapid development of the chemical industries and the production of synthetic materials; the appearance of the motor car and the aeroplane.11

  More recent writing (including that of Kern and Harvey) has stressed the role of telecommunications; others mention emigration, both within and away from Europe.12 Some of these developments suggest comparisons with the present.

  At about the same time, architectural theorists bega
n to develop new concepts of architectural space. For Reyner Banham, in Theory and Design in the First Machine Age (1960), ‘a series of revolutionary gestures around 1910, largely connected with the Cubist and Futurist movements, were the main point of departure for the development of Modern architecture’.13 Banham’s narrative is that of evolving concepts of space, specifically ‘the change-over from the Lippsian idea of space, as felt volume [my emphasis] … to the later concept of space as a three-dimensional continuum, capable of metrical subdivision, without sacrifice of its continuity’.14 The idea of space as volume enclosed by solid surfaces (characteristic of early modern architects such as Voysey or Berlage, and of Cerdà’s Barcelona) began to give way to concepts in which the solidity of matter was less certain, just as the early modernist city, with its bicycles and electric trams, would give way to the city of the motor car. By 1929 László Moholy-Nagy was able to formulate the minimum definition – ‘space is the relation between the position of bodies’15 – which for Banham confirmed ‘the whole revolution in architectural theory that had been going on since 1908’.16 One of Moholy-Nagy’s earlier spatial expositions was his 1921–22 proposal for a film, Dynamic of the Metropolis, which somewhat anticipates Dziga Vertov’s 1929 Man with a Movie Camera. Dynamic of the Metropolis was never realised, but by 1929 Moholy-Nagy had made Berliner Stilleben (1926) and perhaps also Marseille, Vieux-Port (1929), so that the ‘minimum definition’ of modernist space was put forward by a theorist who was also an experienced film-maker.

  Banham saw the distinction between Theodor Lipps’s and Moholy-Nagy’s spatial concepts as sequential, but the idea of space as ‘felt volume’ only slightly pre-dated the subsequent, more abstract formulation – it appears that the word ‘space’ (raum) was not used in Lipps’s (or any other architectural) sense before about 190017 – and Lipps’s concept never really went away. The distinction between the two spatial concepts is very like that between Gunning’s two kinds of cinema, and the spatiality of the early films – their depiction of architectural space within a single frame, their uninterrupted, lengthy spatio-temporal continuities (the tram rides especially) and the ‘stereoscopic effect’ – is easy to identify with Lipps’s formulation. Banham’s Theory and Design was published in 1960, before the revival of urbanism in architectural theory in the mid 1970s, since when architects and others have attempted to revive this early modernist space, just as film-makers have revived some of the forms of early cinema. Both Lipps’s space and the ‘cinema of attractions’ might be seen as early modernist forms which were eclipsed in the late 1900s, as part of a wider cultural transformation, but have since re-emerged, usually in opposition to the mainstream architectures and cinemas of Western and other capitalist cultures.

  In The Condition of Postmodernity, Harvey also quoted the famous passage from Walter Benjamin’s ‘The Work of Art in the Age of Mechanical Reproduction’ (1936):

  Our taverns and our metropolitan streets, our offices and furnished rooms, our railroad stations and our factories appeared to have us locked up hopelessly. Then came the film and burst this prison-world asunder by the dynamite of the tenth of a second, so that now, in the midst of its far-flung ruins and debris, we calmly and adventurously go travelling.18

  Benjamin’s ‘now’ refers to film as it had evolved after the mid 1900s – his essay, published in 1936, mentions nothing earlier than the films of Abel Gance, Vertov and Joris Ivens – but it is not entirely clear at what date ‘came the film and burst this prison-world asunder’. If the development of cinema was a significant factor in the transformation of urban and other space during the 1900s, one wonders whether this was the development of cinema per se, or the development of cinema with editing, narrative and close-up as it was undertaken after the middle of the decade. The fragmentation Benjamin describes can be identified in post-1910 experience as a breaking up of space into individual shots, in which case ‘the dynamite of the tenth of a second’ is the interval between the end of one shot and the beginning of the next, rather than the medium’s primary fragmentation of continuous duration into the discontinuous individual frames of a single shot. But the essay also famously stresses ‘the incomparable significance of Atget, who, around 1900, took photographs of deserted Paris streets’,19 and it might seem to us (over seventy years later) that in some ways one could ‘calmly and adventurously go travelling’ (even) more easily in the early 1900s than in the period of Gance and Vertov.

  Whatever the date of the films that Benjamin had in mind, something happened to the medium in the mid 1900s. The change that Gunning identified seems to have followed a distinct lull in output, during or soon after which many of the pioneers ceased production. After the mid 1900s, films are generally longer but with shorter shots, close-ups and, increasingly, fiction and studio sets; few of them show very much of ordinary landscapes. When they do, the shots are usually so short as to permit relatively little exploration, even when examined frame-by-frame. In contrast, the brief, continuous or near-continuous films of spatial subjects made by the Lumière and Biograph companies and their contemporaries before about 1903 accumulate an extensive document of ordinary, everyday spaces of their period: the spaces that Lefebvre and others suggest were radically transformed soon afterwards. In enabling us to see so much of this landscape, these early films are truly extraordinary, as they offer the most extensive views of the landscape of another time at or just before the moment of that landscape’s transformation – a transformation brought about (at least in part) by the development of the very medium in which the opportunity to explore these long-lost spaces was constructed.

  What do these films mean for us? On looking at them, what struck me first was a contrast between their often familiar-looking landscapes and the unfamiliarity of the society glimpsed in them. In the last hundred years, the material and other circumstances of the UK’s population have altered enormously,20 but much of the urban fabric of the 1900s survives, often – like so much of the built environment – in a surprisingly dilapidated condition. In terms of life expectancy, physical health, income, mobility and so on, we are far better off than our predecessors of a hundred years ago: developed economies experience unprecedented levels of consumption and GDP per head, but in other respects – especially when measured in terms of social, cultural and environmental assets – wealth has not increased anything like as much. In some ways, in some places, it has probably decreased.21 In emphasising this, the films might be thought subversive.

  Walking in the streets of UK towns and cities today, the decline of what Lefebvre described as ‘the environment of and channel for communications … in the form of the city and the town’ is easily recognised. One often detects a sense of absence, even in the centre of London. The spatial qualities suggested by many early films are very like some of those that attract tourists to less advanced or (some) socialist economies – to places where artisanal production or its past products survive, where domesticity is still found in city centres, and where there are fewer cars, or at least less traffic engineering. In advanced economies, such environmental qualities are typically achieved or retained through socialist (as in, say, Barcelona) or social-democratic (as in the Netherlands) politics.

  In this context, Lefebvre’s shattered ‘space of common sense’ suggests both the spatial concepts of Lipps and the urban design of Camillo Sitte. In 1903, Lipps ‘argued that our bodies unconsciously empathised with architectural form’,22 and Sitte,

  rooted in the craftworker tradition of late nineteenth-century Vienna … sought to construct spaces that would make the city’s people ‘secure and happy’ … He therefore set out to create … spaces – plazas and squares – that would promote the preservation and even re-creation of a sense of community.23

  Such ideas re-emerged in the postmodern urbanism of the 1970s, for which early films might initially seem to offer some support, with their depiction of what to us appear ‘traditional’ urban spaces in which we might imagine we could be ‘secure and happy�
�, spaces which ‘would promote the preservation and even re-creation of a sense of community’. Sitte’s polemic, however, was not in favour of the actually-existing spaces of the 1900s – the spaces that appear in the films – but against them, ‘abhorring the narrow and technical functionalism that seemed to attach to the lust for commercial profit’, and seeking ‘to overcome fragmentation and provide a “community life-outlook” ’,24 rather as we might today. Also, though Sitte is popular with present-day urban designers, his desire for spaces that he believed would promote ‘community’ was not unproblematic. As Harvey writes:

  many of the Viennese artisans whom Sitte championed … were later to mass in the squares, piazzas and living spaces that Sitte wanted to create, in order to express their virulent opposition to internationalism, turning to anti-semitism … and the place-specific myths of Nazism.25

  The spaces of UK cities in the early 1900s were subject to transformations at least as sudden as any we experience today,26 but there were few cars until later in the decade.27 Early films depict a space in which there are electricity and telecommunications, but not much oil, so that the transformations of circa 1910 can be seen in terms of the coming of the oil economy and the motor car, which since the mid 1970s has been so widely cast in opposition to conventional formulations of dwelling, and to certain kinds of urban space and architecture. Here again, pre-1907 films might seem to offer a polemic – for streets without cars, for architecture, for public transport, and for a less centralised, less dematerialised economy. They might even resemble science fiction: a future in which the costs of distant labour, and of energy, and hence transport, have increased, so that production becomes more local. At the same time, we can assume that, as images, the films bestow an illusory coherence on their subjects. The spaces that appear in the films were dynamic, subject to tensions as unsettling as (and sometimes surprisingly similar to) any we experience today.