… I’m coming from something to that, and from that going to something. So I always did everything as an arc. I never did this without hanging here and groping there. Any good director that has any quality or any competency at all does not work on the one setup. He’s coming from where he was and groping to where he’s going, in order for that to work right. Then there are directors who do specifically that, nothing else, next, nothing else, next … I don’t understand how that works, but they do it. They get it done; the movie’s made, you know … But it annoys me to think that it doesn’t take that much more to do it right.
– Jerry Lewis, 2003 [1]

The vast area of film theory written by filmmakers is a largely unexplored terrain, except for the pioneering (and necessarily partial) attempt at an initial synthesis attempted by Jacques Aumont.[2] Of all the filmmakers who, over the past century or so, devoted a segment of their energy to serious theorising, I believe the least appreciated is Raúl Ruiz (1941-2011). This is especially so in relation to the 2003 essay that is, in my opinion, the centre and pinnacle of his reflections in this field: “The Six Functions of the Shot”.[3]

Ruiz’s playful meditations in his Poetics of Cinema book series (two volumes, with a posthumous third promised)[4] are, on the whole, internationally better known and circulated than the “Six Functions” essay (as I shall henceforth abbreviate it), which was composed apart from this series. Even here, however, very little of the real substance of the Poetics books has been mined by commentators, however sympathetic they may be to Ruiz’s work.

His critique of the mainstream, Hollywood narrative principle of ‘central conflict theory’ is often cited – indeed, Ruiz’s entire œuvre as an artist may be, in the cinephile imagination, associated with this concept, above all, today. But what of the speculations in his Poetics about lighting, sound, genre – and of the governing notion, generative where the critique of central conflict is essentially negative, that “the images that together make up a film determine what type of narration will structure the film and not the contrary”?[5]

In the use of Ruiz’s theoretical ideas, we find, in fact, the same aporia as in much of the analytical commentary on his films: it stays at the level of generality and evocation, an auteurist genuflection before all that is somehow vaguely ‘Ruizian’. Or – a far worse fate – Ruiz is politely excused from the table of contemporary film-philosophy on the assumption that his writings, however entertaining, are fundamentally idiosyncratic, eccentric, applicable only or mainly to his own films. This seems to me very far from the truth, and I hope to show why in what follows. Ruiz’s meditation on the shot and its functions is a major contribution to film theory.

Ruiz’s writing on cinema poses an inventive theory, not a merely descriptive one. It often proceeds via what philosophy calls the thought experiment: imagine if X were Y … As such, Ruiz’s reflections relate to a large, similarly uncharted area of writing on film that I call creative criticism; in this field, we also find the very different figures of Manny Farber, Guillermo Cabrera Infante, Helmut Färber, David Thomson, Lesley Stern, Raymond Durgnat, and many others.[6] Creative film criticism finds its genealogical roots (sometimes without consciously knowing it) in a diverse range of philosophical, literary and artistic endeavours: Walter Benjamin’s or Siegfried Kracauer’s cultural commentaries, Surrealist games of irrational enlargement, Roland Barthes’ poststructuralist period, Gregory Ulmer’s teletheory, the visionary aphorisms of Malcom de Chazal, and Michael Taussig’s radical departures from anthropology and ethnography, just to name a few. Wherever and however it happens today, creative criticism begins from the felt need to re-energise or re-enchant (a key word in creative criticism) our living relation to films. And so we are obliged to invent forms of writing, or multimedia expression, that do not presume to substitute for the experience of cinema, but rather accompany it in a vivid, expansive, suggestive way.

Cinema, for Ruiz as for many creative critics, has precious little to do with the norm of the commonsense viewing or perception of cinema (such as, for instance, cognitive theory seeks to establish): the assumption that films are, first and last, basically stories about human beings who feel things, struggle against and overcome obstacles … Cinema should be, rather, a machine that allows us to attack all commonsense and commonplace assumptions, inside a movie theatre as well as outside of it.

Ruiz’s “Six Functions”, however, as creative as it undoubtedly is, returns us to the basic thrust of his books on cinema: it methodically proposes a poetics, which he defines not so much as set of basic principles (as in Aristotle) but a working method, a way of doing or making things, a means of generating images, sounds, and the connections between them. Ruiz always used, in his writing, the rhetorical form of the logical demonstration – with its assertions and syllogistic proofs – even as he spins it into surreal conclusions. Ruiz’s essay therefore has an eminently practical, and indeed pedagogical application; indeed, Ruiz used it within his teaching of filmmaking. On this level, it can be compared with the (very different) system proposed by a more classical filmmaker who Ruiz greatly admired, Alexander Mackendrick.[7] Unlike Mackendrick, however, Ruiz’s ultimate aim is to imagine an ideal, even liberated cinema – as he avows, his goal is to “awaken the imagination” of both viewers and makers of film, and to establish (through careful, practical steps and exercises) a whole new way of creating. The process of poetics was a means to achieve what he would often refer to as ‘a genuine moment of poetry’ – something much harder to have and to hold than a run-of-the-mill poiesis, with its cheap, routinesed effects of lyricism, beauty or shock.

Is there an Other, an enemy lurking in the background of this picture painted by Ruiz? We must take him at his explicit word here: the enemy is narrative. This can strike many devotees of Ruiz as an odd, unlikely, paradoxical or contradictory position for him to take: is not the joy of pure fiction more evident in Ruiz than in any other contemporary filmmaker?[8] Was not Ruiz a born storyteller, a spinner of tales – no matter how devious or Baroque – in the tradition of Luis Buñuel? Ruiz himself always publicly played down such a characterisation of his project. He was not interested in telling, for each of his films, a story, singular and particular. Above all, he did not wish to achieve the imaginary coherence or plenitude of a story. Narrative is just a pretext for him. A pretext for what? As he often declared, he is interested, above all, in the passage between diverse worlds (real or imaginary), or between different narrative levels; it is those bridges, those suspension points, and the difficult moments of connection or disconnection between levels that he seeks to understand, explore and work with.

“Six Functions” is Ruiz’s most brilliant and sustained meditation on cinema because it offers, precisely, a theory of connection and disconnection in the filmic medium. And it does so by grounding itself in a particular syntactical unit of cinema: the shot.

A caveat, before we move deeper into Ruiz’s theory, as to the scope and limits of my investigation here. I will be placing Ruiz’s essay essentially within the context of the developments (and arguments) of Anglo-Euro film theory since the 1960s. This is not necessarily, or solely, Ruiz’s own intellectual context: as we know from Poetics of Cinema, he reserved the right to freely generate his concepts from the world over and from the depths of time, drawing in (and recombining) ideas from the Latin American Baroque, Arabic culture, ancient Sophist philosophy, Chinese aesthetics, advances in contemporary neuroscience and mathematics, and so on – as well as a number of the contemporary currents (including Lacanian psychoanalysis, Third World Marxism, Baudrillardian postmodernism, and Jean Louis Schefer’s aesthetics). To draw the map of all that is a gargantuan task for another time – if, indeed, the secret key to these borrowings ever belonged to anyone but Ruiz himself. A secret cinema …

Nonetheless, we can relate Ruiz’s choice of the shot as the fundamental cinematic unit to many early film theories, from Sergei Eisenstein in the 1920s right up to the linguistic/structuralist period of Christian Metz’s pioneering cine-semiology in the 1960s and early ‘70s. Indeed, Ruiz’s affinity is more towards Eisenstein than the later theorists of film language: for, as Eisenstein first proposed (and as his commentators from Aumont and Gilles Deleuze to Marie-Claire Ropars have adopted), there are visible shots in cinema but, equally, the less visible intervals between shots, or the fleeting interstices between them – not only, technically, the literal black spaces between film frames on celluloid (so much a part of cinema’s peculiar metapsychological effect for the spectator)[9] , but also conceptually. This expandable interval between shots, both material and immaterial, is a crucial factor for Ruiz also.

More obviously poststructuralist in orientation are the two essential properties that Ruiz ascribes to any shot, all shots. Firstly, it is discontinuous in relation to all its neighbours in a given film; it is therefore an autonomous unit – an autonomy that Ruiz takes to the extreme in his now famous declaration: “When we see a film of 500 shots, we also see 500 films”.[10] Secondly, the shot must be conceptualised as a force-field, a complex (and even mysterious) interplay of energies and intensities. Here, Ruiz has some affinity (whether he knew it or not) with Jean-François Lyotard’s mid ‘70s notions of ‘the unconscious as mise en scène’ and of theatre (taken as the grand metaphor for all representational forms) as an energetic dispositif,[11] and also with Stephen Heath’s elaboration of Barthesian semiotics within the field of the textual analysis of film.[12]

In “Six Functions”, Ruiz proposes a novel definition of what is contained within any shot – although, at first, it may seem a quite conventional definition. A shot is comprised, he suggests, of events (things staged or found within the pro-filmic field) and a point-of-view that is imposed by the camera’s position and perspective; these events are settled (determined, shaped) by a change in the camera’s viewpoint (such as a reframing due to camera movement). A shot is, furthermore, a set of objects (including humans as objects) linked by actions.

Thus far, we have nothing that could not be accommodated within the most classical definition of filmic style, form or mise en scène – compare it, for example, with the proposition of Positif critic Alain Masson (who has written well on Ruiz) that “form results from the changing relations between places, gestures and camera viewpoints”, within “a mode of representation where movement constitutes the principal authority”.[13]

The novelty of Ruiz’s idea becomes evident when we reach the following provisos: firstly, that each shot, in its autonomy, also contains provisional openings and closings – which is a more radical notion than that a shot may modulate itself, or have successive stages – and secondly, that every shot possesses microfictions that are “struggling to draw attention to themselves”, the busy sum of which constitute the “unconscious of the shot”.[14] (Here a parallel can be drawn with Thierry Kuntzel’s influential notion of the ‘other film’, polymorphous and unbound, hidden within every seemingly normal film.)[15]

What are microfictions in film? Take a look at the opening street shot, pregnant with enigma and suggestiveness, at the outset of Hypothesis of the Stolen Painting (1978): it is up there on screen long enough for us to wonder about its dark impasse, the open windows that face each other, all the tiny objects and events that flutter in the breeze, the random street sounds … not to mention (in retrospect) the puzzling relation of this apparent establishing shot with any visualised space that follows it. Or recall Ruiz’s impromptu proposal for his ideal art installation (offered from the stage of the Museum of Contemporary Art in Australia): a film set after everyone has left, a dim light streaming in from somewhere outside … In such dispositifs, all the elements present in a pro-filmic situation are left free to interact with each other in imaginary scenarios.

Let us proceed to a discussion of the functions involved in Ruiz’s categorisation of shots. In his system, what is a function for, and what action does it perform? These functions are twofold: they mark both the potentialities in the shot – the latent or dormant energies or capacities that can be made manifest, sparked into action – and the possible ways it can be linked to the surrounding shots. Therefore, while holding to the principle of each shot’s autonomy, Ruiz is also seeking to establish the ensemble that a group of successive shots can form, and the (extremely variable) logic, rules or poetics of these ensembles.

We should note a curious limit of Ruiz’s meditation on this particular point: the constraint of shot succession, a linear or synchronic unfolding (even when the ensemble-relation is anything but linear in its sense or logic) – something that, in other contexts, when he speaks of the diverse entry points that a spectator takes into a film, he jettisons. We shall return to this complementary aspect of the entire Ruizian œuvre as constructed in its filmed, written and spoken dimensions – spoken, because Ruiz can certainly be classed among those directors (including Abraham Polonsky and Jean-Luc Godard) who eagerly and fulsomely use the interlocutory situation of the filmmaker-interview as an opportunity to extemporise and theorise; it is a principal outlet for his creative criticism).

The function of a shot, in short, is what determines its various openings and closings – and also the destiny of its numerous microfictions. “The microfictions told by the set are hierarchised with the appearance of a point-of-view”.[16] Any hierarchy, for Ruiz, performs a dual action: on the one hand, it closes down the field of possibilities (by brutally deciding on some over others) and, on the other hand, by this very act of elimination, it implies and even outlines the presence of all those other options – and therefore inadvertently sets them loose, sets them to work like Manny Farber’s termites throughout the body of the shot or the film.

The conceptual and aesthetic logic – whereby elimination is also a kind of perverse procreation – was expressed as early as 1981 in Ruiz’s writing, in his piece “Three Thrusts at Excalibur” (which prefigures many themes of “Six Functions”), where it is asserted that “every story, to the extent that it narrates, also at the same time excludes – and thus makes obvious – not the rest of the world, but its counterpart … its occult alternative”.[17]

Generally, however – within the context of the dominant, mainstream trend in filmmaking practice – the hierarchical nature of cinema is, for Ruiz, an impoverishment that he subsumes under the banner of a “subordination to the strategic model”. Strategic, here, is the opposite of a key term for Ruiz, recursive (to which we shall soon return); and our author’s dismay on this point is not so far from Jerry Lewis’ quoted at the head of this essay, when he pours scorn on those directors who attend to only the ‘piece of plot’ before them in the script at any given moment of the shooting schedule, who “do specifically that, nothing else, next, nothing else, next”, and in doing so they “get it done; the movie’s made” – but made only as a strictly functioning, streamlined narrative machine.

So here are the six functions as outlined at various times and in various ways by Ruiz, both in the essay and in the useful summary included in the interview given as a bonus on several Gemini Films DVD editions of his films:

1. A shot is a paradigm and an allegory for the whole film – as when it provides a ‘poster image’, or what John Ellis precisely calls a film’s narrative image.[18]
2. It is a self-reflective or critical image. This is what Ruiz calls the recursive function (opposite to the strategic function of ‘always going forward’), the function of doubt that makes a viewer want to immediately ‘rewind’ or re-see an image in order to verify the enigma of its contents.
3. It is centripetal (independent, autonomous).
4. It is centrifugal (it ‘reaches out’ to connect with the following and preceding shots).
5. It is holistic – it can ‘replay’ the entire movement or deep logic (not the superficial plot) of a film in microcosm.
6. It has combinatory potential – it suggests ways in which it can be ‘remixed’ into an alternative sequence of the given shots, resulting in different semantic and syntactic associations.

When we try to categorise the shots of a film using Ruiz’s schema, we must remain faithful to his underlying assumption that any shot potentially contains, or performs, all six functions. Indeed, his utopian goal as a filmmaker – rarely achieved in practice, as he hastened to point out, perhaps impossible to achieve in its totality – is to enable all the functions to figure in every shot for a whole film, the entire time. Therefore what, in less ideal aesthetic circumstances, determines a shot’s dominant or characterising functions? Precisely the connection it makes – or does not make – with its neighbours.

I will dare to propose a hierarchy of these functions, from the top downwards. Ruiz might have disagreed with me on this but, of the six functions, two of them seem to me to be the major categories: centrifugal and centripetal. The critical or recursive function is the next most significant. The paradigmatic and the holistic functions offer variations on each other; it is easy to confuse them conceptually. And the combinatory opens a set of possibilities that are not fully explored within the “Six Functions” essay itself, since they fall somewhat outside the field delimited by the centrifugal/centripetal opposition – but these possibilities frequently formed the substance of Ruiz’s remark about his own filmmaking practice, since they call up the non-successive, non-syntagmatic, diachronic dimensions of the work. (They are also evident in the 2008 short film In the Blink of an Eye by his Aberdeen students Cameron McEwan and Alison Teller, viewable on YouTube and Vimeo.)

The centrifugal/centripetal duo – I can testify to this from my own teaching practice – completely opens up the realms of cinema theory, analysis and criticism; it is this pair on which I shall concentrate here. In terms of Ruiz’s dynamic force-field approach, he describes the difference between these two tendencies thus: where the energy of a centrifugal shot seeks to exhaust itself, to die, so as to lead on or hook into the successive shot – a basic principle of classical narrative economy and efficacy – the energy of a centripetal shot seeks its central core, so as to assert and demonstrate its autonomy. As Ruiz has noted in interviews, the elaborate long takes of contemporary cinema (Angelopoulos, Tarr, Tsai, etc) are clear examples of centripetal autonomy in action – shots that are, in a palpable sense, films in themselves. But it is crucial to add that even a temporally brief shot can be centripetal in its function and effect. And this is also not an opposition, as is sometimes mistaken, between static shots (whose literal force-field centre may be quite visible, as in a film by Hou or Jia) and in-motion ones; for Ruiz, a shot that opens and closes many times over, constantly settling, unsettling and resettling itself via elaborate camera movement, can also be a model of centripetality – as is frequently the case in his own films.

The key feature or element, as I have already suggested, is the shot’s will to connect – or its refusal to do so in conventional terms. The traditional syntactic connection between successive shots can be resumed, within psychoanalytically-oriented film theory, under Jean-Pierre Oudart’s influential idea of cinematic suture: the editing join (such as shot/reverse-shot, graphic match, or raccord on movement) that stitches or repairs the always potentially gaping interval or discontinuity between shots, creating the illusion of a single, seamless, fictive space. But Jerry Lewis’ account may be just as apt, and indeed closer to Ruiz’s total view of the ensemble of a shot’s relations: the centripetal urge is all about ‘hanging here’ – sticking to the time and action of a shot, exploring it, opening out all its dimensions – while the centrifugal drive is about  ‘groping there’, restlessly setting up and leading on to the next plot-point, the next event. And richness may result from the co-existence and intermingling of these functions – not the polemical privileging (whether for classicist or modernist ends) of one over the other.

Let us look into a filmic example: eight shots from the first sequence of Time Regained (1999). Not that our example must come, or could only come, from Ruiz’s work itself; on the contrary, the subtlety and depth of “The Six Functions of the Shot” as an analytical tool reveals itself just as well when we use it on Orson Welles, John Cassavetes, Ritwik Ghatak, Roberto Rossellini, Max Ophuls or Bruno Dumont. But Time Regained certainly offers a remarkably concise instance of an ever-expanding ensemble of shot connections and disconnections.

During what is surely among the finest opening lines in all cinema – “Then, one day, everything changed” – we watch an optically distorted shot of written pages, ending in the glimpse of the hand that is inscribing them. The dying Marcel Proust (André Engel) is dictating to his assistant, Céleste (Mathilde Seigner). But this much is not yet clear in our first shot, which seems to ‘run on’, in metaphoric terms, from the previous, shifting image of running water. Shot 2 ‘places’ the voice of the speaker, lying in his sick bed, but completely jolts us away from the previous shot: in this very baroque image, Céleste’s head, seen for the first time, is jammed down into the bottom right-hand corner of the frame – the suture-effect is weak, not strong. Indeed, in each shot transition here (and all throughout the film) the raccord is always tentative (the spectator often needs to hunt out the tiny element of pictorial overlap), even in the case of shot/reverse-shot configuration. How can this be so?

Ruiz’s centripetal drive is ceaselessly pursued through multiple means: each shot folds into itself, seeks its own energy-centre, and breaks its usual stable connections with the shot to come, through elaborate camera movements (sometimes involving mirror reflections, as in shot 7), certain adjustments in the sound (music or not, Marcel abruptly passing from dictation to speech as at the start of shot 3), and especially the tendency for all furniture in the décor to cease being fixtures and instead become animated, moving across, up or down – creating many momentary distortions of visual perspective (shots 2 to 4). In terms of even the simplest plot information, Ruiz often plays a game of hide-and-seek: in the lengthy shot 7, Céleste carries out Marcel’s command to get him something from a drawer, but we do not see what it is until the following shot 8, which stands as its own autonomous scene: a procession of photos seen through a magnifying glass.

There is much more that could be said about “The Six Functions of the Shot”: its concept of “specifically cinematic emotion”; its views on the systematic organisation of sound (which constitute another kind of ‘viewpoint’ on a scenic ensemble); its adaptation of pictorial principles from Chinese painting that allow “ways into the shot, strolls inside the frame”[19] ; its idea that the ‘weak’ elements in successive shots can build strong chains, like we see in the comedies of Ernst Lubitsch (musically graded door chimes) or Stephen Chow. But Ruiz’s theory of the centripetal has some special uses for film critics and theorists. Not only does it serve as a good weapon against all those folk (cognitivists, genre buffs, screenwriting advisors) who preach the gospel of classical narrative – often explicitly premised on the purely centrifugal assumption that the question ‘what will happen next?’ is always more significant that what is going on in the here and now[20] – but it also helps us complicate and overcome one of the more lasting fixtures in the landscape of contemporary film theory.

I am referring here to the theoretical opposition – which has taken many forms since the 1970s – between narrative and spectacle. In a widely used and accepted account (derived, for instance, from Laura Mulvey’s famous feminist manifesto “Visual Pleasure and Narrative Cinema”), narrative is what forms the linear, forward drive of a film, while spectacle (or what Tom Gunning calls, in relation to silent cinema, its propensity for attractions) is what freezes, opposes, contests that movement.[21] We are no longer dealing with apologists for classical narrative; on the contrary, this is the space for people who wish (as Ruiz does) to undo the malignant hold that a certain dominant, institutionalised form of storytelling has over its audiences worldwide. But the clear-cut opposition between the fearsome narrative regime, always on the march, and the static interruptions to it (whether digressive monologues in Godard or songs in a musical) is, ultimately, a blunt analytical instrument – and it has led to many coarse applications in the realm of independent filmmaking (for instance in the work of Todd Haynes or Gregg Araki).

Let us take a recent version of the narrative/spectacle opposition that is pitched somewhere between liberating the collective imagination and policing an industry norm; it appears in Robert B. Ray’s book (itself the result of a pedagogical practice) The ABCs of Classic Hollywood. Ray is in the midst of discussing the difficult place of Vincente Minnelli in the American system – as an ex-designer who (like Mitchell Leisen) was dismissed by Billy Wilder (centrifugal screenwriter-director par excellence!) as “just a decorator”:

Filmmaking proceeds along two axes, a horizontal movement concerned with sequences and a vertical effort attending to individual scenes, a division of labour resembling music’s melody and harmony. While producers, editors and directors remain responsible for a movie’s horizontal progress, art directors, set designers and costumers focus on the cinema’s more static elements, the mise en scène. Art directors harmonise a movie’s melody, inflecting it as decisively as a shift from major to minor chord can transform a melody. But they can neglect a film’s movement from scene to scene.[22]

Vertical and horizontal here correspond to centripetal and centrifugal. But with what strange, amputating, exclusive consequences: the centripetal realm, far from being a force-field, is assimilated to the static, to mise en scène, and to musical harmony, all in the hands of the fussy art directors, costumers and designers; while the centrifugal is associated with ‘progress’, the proper care for ‘movement from scene to scene’, and melody – all of which are the province of ‘producers, editors and directors’! Minnelli, in this stark drama, is ‘out of place’ precisely as a recalcitrant, recidivist ‘designer at heart’ who never entirely mastered the horizontal obligations of cinematic storytelling. One shudders to imagine how the vertical/horizontal or spectacle/narrative distinction would deform the work of an Almodóvar, a Carmelo Bene, a Derek Jarman, a Powell/Pressburger … or, indeed, a Ruiz.

I do not wish to suggest that all film theory since the ‘70s has gotten itself locked into this rigid opposition. In more recent times, for example, Mulvey has developed the notions of delay and deferral in film (with respect to, for example, the work of Kiarostami) as other, finer forms of resistance to the death-driven machine of conventional plotting – which, like each one of its shots, aims to exhaust itself, empty itself out completely, by the final frame or event.[23] Gilles Deleuze’s idea (developed in his books of the ‘80s) of the time-crystal in cinema – like Jonathan Rosenbaum’s evocation of the nonnarrative layering of past, present and future in films by Alain Resnais or Carl Dreyer – have also pointed in this much-needed direction.[24]

But Ruiz’s demonstration of the centripetal and centrifugal in film is the most material and precise tool that we so far have to investigate these possibilities. Rather than the stand-off of narrative and spectacle, we have a new attentiveness to two, simultaneous kinds of involvement inherent in cinematic spectatorship: the sequential (centrifugal) and the immersive (centripetal). As cinema moves and mutates in the 21^st century, we will – as critics or as practitioners – need to be increasingly adept at both levels of engagement.

This essay is part of work funded by the Australian Research Council through Monash University for 2010-2012, on the topic of “Between Film and Art: An International Study of Intermedial Cinema”.