I’ve met a few people who think that the films of Andrei Tarkovsky are boring. I understand. These films are slow. Most don’t have much of a story. There’s not much sex or violence or snappy dialogue. I was bored myself the first few times I watched Tarkovsky films. I loved the beautiful, enigmatic imagery, the evocation of dream logic, but it was an effort to pay attention. Then something happened.
I think it was like this. There’s a point about 42 minutes into the film Stalker. We have discovered that the “stalker” of the film’s title is an illegal guide who, for a fee, leads people into a mysterious area of land called the Zone, prohibited to ordinary citizens. A poor, awkward man, he has taken two clients by devious and daring routes past the military defences that surround the Zone, and they have just made a long, silent journey to its interior on a stolen railway cart. Up to now, the film has been in high-contrast black and white. Now it switches into colour. The landscape is hilly, misty, overgrown meadowland, but not forested. The sky is lightly overcast, but the pink of dawn shows through, saturating the green of the lush vegetation. Stalker wanders off to be by himself. Out of sight of the others, he sinks down into the undergrowth and lies flat out on the ground. He turns over and we see him in medium close-up:
Then – did it? – the light seems to change colour slightly, as if one of the layers of cloud in front of the sun has moved away. Then the colour shifts back.
I first saw this film on video, and my first thought was that some subtle tracking error, or some artefact of the telecine or dubbing processes, was responsible. But it didn’t look like anything I’d seen before. I began to look for similarly subtle features – for things that might easily pass by unnoticed, that were not clearly intentional. And the film was full of them. Of course, as these things were not clearly intentional, they might not have been intentional at all. I might just have been finding whatever I had decided to look for, in a film that was vague and slow-paced enough to allow for reverie on the part of the viewer. Whichever is true, boredom stopped being an issue for the rest of the film – almost two hours. I was still thinking, still aware of myself watching the film, but I was also living in the present moment more fully than I had for a long time. That’s what I want to write about here.
* * * * * * * *
Bored, bored, bored
My friend Shirley’s dad used to say that if you find a thing boring, it’s your problem. Nothing is inherently boring.
The sociologist Michael Flaherty has investigated how people experience time. In cases of waiting and boredom – as in situations of violence and danger – time seems to slow down. Awareness of the world around is heightened, and so is awareness of ourselves. This happens, typically, when people are denied their usual activities, when there is lack of variety or of freedom to act – imprisonment gives the clearest example of this – or when people are clock-watching – wishing a future event or stimulus would happen soon. Boredom, suggested Flaherty, occurs when for want of other stimulation people end up concentrating on the passage of time itself, willing the minutes to pass more quickly. The proverb that supplies the title of his book, A Watched Pot, is an example of this. Flaherty summed up:
Obviously, a “boring” situation can present one with potentially engrossing circumstances. Perhaps some degree of agency figures in the phenomenon: Does one elect to be the kind of person who takes up that kind of challenge? 
Put it like that and boredom begins to sound quite exciting. So if we wanted to make a boring film, in order to open the viewer’s mind and root him or her in the present in this way, how would we do it? Taking a lead from Michael Flaherty, we could make it in a way that would frustrate the viewer’s expectations. To do that, we have to know what they are.
* * * * * * * *
Cutting on action and making you think
Most films aim to keep things snappy.
In his classic 1950s book on film editing, the director Karel Reisz begins with the principle that film editing can recreate the way we see the world around us: run your eye along a bookshelf – general view; a bright red spine catches your eye – cut to close-up; having read the title, you look back at your desk – cut to another wide shot. But conventional narrative film-making does not just show one point of view: dialogue scenes, most obviously, usually intercut shots that are meant to show each character’s point of view in turn. So how does a director know where to point the camera?
The director’s aim is to give an ideal picture of the scene, in each case placing his camera in such a position that it records most effectively the particular piece of action or detail which is dramatically significant.
In classic narrative, what is significant involves novelty, change or extremity of one kind or another. The mid-twentieth-century Hollywood film director Edward Dmytryk had no doubt about the matter:
In all good films it is essential that the characters grow or, to put it more accurately, develop, and such development is most effectively shown through their reactions either to physical crises or to verbal stimuli. These are the “moments of transition” which every actor and director looks for in the script’s scenes…
Karel Reisz’s advice to several generations of film-makers was more dogmatic still: “The spectator,” he declared, “[is] only interested in the sequence of significant events.”
These are not merely prejudices. The “Constructivist” film theorist David Bordwell summarises the results of scientific investigations into “how people comprehend and recall stories” like this:
[In Western culture] people tacitly assume that a story is composed of discriminable events performed by certain agents and linked by particular principles. People also share a sense of what is secondary to the story’s point and what is essential to it. Thirdly…people perform operations on a story. When information is missing, perceivers infer it or makes guesses about it. When events are arranged out of temporal order, perceivers try to put those events in sequence. And people seek causal connections between events, both in anticipation and in retrospect. […] One researcher found that comprehension and memory are best when the story conformed to the drive-to-a-goal pattern. […] The contemporary Western perceiver does typically expect expository material at the outset, a state of affairs disturbed by a complication, and some character ready to function as a goal-oriented protagonist.
In short, stories tend to emphasise causal connections and identifiable goals. When Oscar Wilde’s anti-hero Dorian Gray, intent on seeing life through the spectacles of art, discovers that his lover Sibyl Vane has drowned, he rejects the possibility that her death was “a vulgar accident”. To Dorian, only events with reasons, motives and meanings are aesthetically acceptable: “Of course she killed herself.”
Classic American and European narrative film-making – Hollywood films of the 1930s and 1940s provide the clearest examples – took up this principle with enthusiasm: “the form of [the] economy [of film] is narrative, the narrativization of film”. It skips over any event that is not dramatically significant, as far as is possible without damaging the integrity of the narrative too much. This selection of significant detail does not simply deliver efficient plot exposition. As David Bordwell points out, the activity of watching fiction films consists largely of reconstructing the plot and the motivations of the characters from the fragmentary clues presented: “Every fiction film… asks us to tune our sensory capacities to certain informational wavelengths and then translate given data into a story.”
So the test of a good editor, as described with wonderful subtlety by both Reisz and Dmytryk, is his or her skill in shaving off “dead time” – the seconds or fractions of seconds that halt the pace and rhythm of significant action, the space between thought and action, between action and reaction, keeping the experienced film viewer tuned in to Bordwell’s “informational wavelengths” and supplied with new stimuli before she or he has a chance to work out what’s coming next, or, God forbid, get bored.
In a conventional action-based narrative, such as all of us who were brought up near a television have been watching since before we can remember, the audience can and does expect that something is always about to happen. If nothing appears to be happening – a woman is wandering through an apparently deserted house, say – that’s because suspense is being built up in preparation for something so shocking as to repay us for our time – she’s about to discover a dead body, or be attacked, or find that her friends have organised a surprise party. David Bordwell again:
When we bet on a hypothesis [about plot or motivation in a film], especially under the pressure of time, confirmation can carry an emotional kick; the organism enjoys creating unity. When the narrative delays satisfying an expectation, the withholding of knowledge can arouse keener interest. When a hypothesis is disconfirmed, the setback can spur the viewer to new bursts of activity. The mixture of anticipation, fulfilment, and blocked or retarded or twisted consequences, can exercise great emotional power. The formal processes of perception and cognition – as Eisenstein knew well – can trigger affect.
More about Eisenstein in a moment. As an example of the emotional kick Bordwell was writing about, think of suspense. It’s the most extreme form of plot-induced emotion, but perhaps its workings reveal a feature of the whole of classic narrative style: what is made visible is merely a functional confirmation of what we have been led to expect. Once we’ve seen it, it’s dispensed with as soon as possible so as to take us on to the next new thing.
Soviet cinema had a different notion of classic style, one that perhaps demanded more work of its audience. In the 1920s and 1930s, the films and writings of V I Pudovkin and Sergey Eisenstein had provided an elaborate and heterogeneous body of film theory. It was Eisenstein’s theories of “montage” that became the orthodoxy, however:
Eisenstein set up his…theory of montage…as collision [between shots] rather than [narrative] linkage. […] For Eisenstein, montage has as its aim the creation of ideas, of a new reality, rather than the support of narrative, the old reality of experience.
The danger of Eisenstein’s approach was that it could easily cause a film to degenerate into allegory, a string of symbolic shots to evoke a precise sequence of ideas. As a film student in Soviet state film school of the late 1950s, Andrey Tarkovsky was not impressed by the Eisensteinian theory he was taught. He wrote an essay criticising a sequence in a then recent historical-drama film that used a shot of a breaking wave for allegorical effect. The film writer Ian Christie has commented: “While seeming to offer the audience something “artistic”, the intercut shot of a breaking wave actually both distracts and insults their intelligence.”
I would argue that some Western film-makers famous for making “artistic” films for the mainstream cinema fall into the same trap. Michelangelo Antonioni filled his films with settings and characters and techniques that scream their significance: the arid rocky island where the idle rich reveal their spiritual and emotional aridity in The Adventure; the foggy town of The Red Desert, in a landscape unsure whether it is land or sea, home to the (literally and metaphorically) out-of-focus, foggy, unsure heroine (the title itself gives away the “message” of the film, there being no desert in sight except, presumably, a spiritual one). And in Terrence Malick’s The Thin Red Line, the punctuating shots of parrots and sunlight shining through jungle leaves become pointless once we’ve understood their function in the film – to inform us that the beauty and cycles of nature continue despite the insane violence of war that provides the film’s plot – but, film being a time-based medium, we have to sit through them anyway.
Perhaps it sounds as if I don’t like either classic narrative films or the work of Eisenstein or Malick or Antonioni. Not at all: there’s plenty of good things about them. But I want to argue that Tarkovsky is doing something different, something subtler, enabling something richer in the viewer’s experience.
* * * * * * * *
Tarkovsky devoted much of his book on cinema, Sculpting in Time, to attacking Eisenstein’s approach. This passage sums up why he thought Eisenstein had got it all wrong:
Eisenstein makes thought into a despot: it leaves no ‘air’, nothing of that unspoken elusiveness which is perhaps the most captivating quality of all art, and which makes it possible for an individual to relate to a film.
Eisenstein’s work is usually contrasted with that of V I Pudovkin, who “felt that [the] purpose [of montage] was to support narrative rather than alter it”. But Tarkovsky was almost equally dismissive of the role of narrative in film-making:
…I am more at home with [the logic of poetry in cinema] than with traditional theatrical writing which links images through the linear, rigidly logical development of the plot. That sort of fussily correct way of linking events usually involves arbitrarily forcing them into sequence in obedience to some abstract notion of order. And even when this is not so, even when the plot is governed by the characters, one finds that the links which hold it together rest on a facile interpretation of life’s complexities. […] The usual logic, that of linear sequentiality, is uncomfortably like the proof of a geometry theorem.
And he was no fan of structuralist interpretations either:
I classify cinema and music among the immediate art forms since they need no mediating language. […] I resist structuralist attempts to look at a frame as a sign of something else, the meaning of which is summed up in the shot. […] [Like music,] one cinema frame is always a particle of reality, bearing no idea; only the film as a whole could be said to carry, in a definite sense, an ideological version of reality.
Perhaps he was just refusing to acknowledge the inevitable semantic content of images, rather as his denial of allegory is a little suspect (more later about this). My point here is that he refused to give house room to any explicit theory of film-making. Instead, he argued that an artist had to commune with his or her own sensibilities, both aesthetic and moral:
…in his work the artist breaks down reality in the prism of his perception and uses a foreshortening technique of his own to show different sides of reality. […] Masterpieces are born of the artist’s struggle to express his ethical ideals. […] If he loves life, has an overwhelming need to know it, change it, try to make it better – in short, if he aims to cooperate in enhancing the value of life, then there is no danger in the fact that the picture of reality will have passed through a filter of his subjective concepts, through his states of mind.
As well as rejecting theoretical supports, he also withheld spectacle and refused to acknowledge the expectations created by genre. For instance, science-fiction films from Metropolis onward have treated their audience to spectacular visions of grandeur – think of the vertiginous cityscapes of Blade Runner, the vast spaceships of the Star Trek series – and provided stimulating speculation about the forms human societies might evolve into. But Tarkovsky had no intention of satisfying these expectations when he made Solaris, his first science-fiction film:
As far as…Solaris is concerned, my decision to film it does not denote any affection for the science-fiction genre. For me, the important thing is that Solaris poses a problem that means a lot to me: the problem of striving and achieving through your convictions, of moral transformation in the struggle in one man’s life.
He had additional objection to adding spectacle or futurological speculation to the film:
For some reason, in all the science-fiction films I have ever seen the audience is forced into a detailed, close-up examination of what the future will look like. Indeed, often (like Stanley Kubrick) they call their films “visions of the future”… I would like to film Solaris in such a way that the audiences are not faced with something technologically outlandish.
The Future, as imagined by Stanley Kubrick…
If, for instance, we were to film passengers getting into a tram as something never before seen or even heard of, then it would look like Kubrick’s moon-landing sequence. But if we film a moon-landing the way they film a tram-stop in an ordinary film, then everything will be as we would wish it.
…and by Andrei Tarkovsky
In spite of which, he later wished that he’d dispensed with the “rockets and space stations” altogether. He was more resolute when it came to Stalker. Apart from the prologue text, you would hardly know this was a science-fiction story.
Not that Tarkovsky abandoned narrative altogether. His films all have more or less coherent plots, and, to use Bordwell’s terminology, his characters in their own ways conform to the drive-to-a-goal stereotype. From the doubting monk Andrei Rublyev to the doubting poet Gonchalov in Nostalgia via a whole troupe of men in Stalker who aren’t quite sure what they want, Tarkovsky’s heroes are nevertheless searching, for spiritual, ethical or emotional certainty or stability.
Plot mechanics, though, are removed as far as possible. This is most obvious in Andrei Rublyev and Mirror, which are structured as a series of episodes, but linkage between scenes is flimsy even in the more linear and unified plots. This leaves Tarkovsky free to concentrate exclusively on what he finds potent and evocative, which may have nothing to do with the narrative backbone of a film. For instance, his disdain for the “technologically outlandish” led to him depicting Chris Kelvin’s interplanetary journey in Solaris with a static shot of (rather unconvincing) stars, a close-up of Kelvin’s face, and some short blurred shots of girders racing by. In contrast, a little earlier in the film, a minor character’s suspiciously 1970s-looking car journey into an equally unfuturistic city is an intensely worked prolonged montage of sound and picture, leaving me with the impression that this is somehow the real journey into space.
Tarkovsky’s treatment of dream sequences gives them equal value to more literal reality, as he refuses to announce their unworldly status by such tricks as slow motion or soft focus. And his frequent shifts between black and white and colour photography defy analysis in either narrative or expressionistic or allegorical terms: there is no clear “meaning” to be had from them.
Andrei Rublyev: the first image; tilt down to…
Take, for instance, the first sequence in Andrei Rublyev. It begins, bizarrely, with a shot of what appears to be a medieval hot-air balloon being prepared for take-off next to a white stone tower. After weaving through feverish preparations for launch, we see the historically dubious aeronaut’s point of view during his brief flight.
A horse, steaming in the chilly air, rolls on its back in what may or may not be slow motion. The balloon incident and characters are never referred to again in the film.
Or the first sequence after the titles in Mirror, which shows a woman sitting on a rustic fence, her back to the camera, smoking and looking out over a wide field of tall wheat:
through which an anonymous man wearing a dark suit and carrying a briefcase is approaching. The voiceover implies that it might be “father”. But it is a stranger:
He asks for directions, and proceeds to press a conversation that is by turns flirtatious:
intimidating, philosophical, perhaps even generous; she is wary, non-committal;
he sits on her fence beside her and it breaks under their weight:
as he walks away, a improbably violent gust of wind rolls across the field; he turns back for a while, then continues on his way:
He is not seen again in the film. There are no repercussions of the incident in the plot. Not only that, I am not even sure what is supposed to have just happened. The scene has the pregnancy of a Raymond Carver short story. But frustrating if you are trying to work out its relevance to the plot.
The films do not play the narrative game properly. The characters’ struggles are mostly internal and not made explicit. Nor do they reveal their meanings through clear allegory or symbolism – even if Tarkovsky was not quite the stranger to allegory he said he was. The images that unfold on the screen often defy attempts to fit them into a coherent plot structure and fail to deliver the ceaseless flow of significant information and action that is the founding principle of classic narrative style. What I hope I have shown so far is why many people might find his films hard to follow – indeed boring. Returning to Michael Flaherty’s research on experiences of time, we can now see that Tarkovsky created two of the key conditions for boredom: in refusing to play by the rules, he both removed the familiar and thwarted the audience’s conditioned expectations. So what was he up to?
* * * * * * * *
What is not plot
While cinema was perfecting its cut-to-the-quick narrative style, writers like Joseph Conrad, Marcel Proust, Virginia Woolf, Samuel Beckett and James Joyce were exploring the very thing that classic cinema storytelling seeks to eliminate – what happens when nothing much is happening. Take, for instance, Beckett’s description of his creation Molloy’s recreation of sucking the collection of pebbles he keeps in his pockets, and his system for ensuring that he sucks each pebble in strict rotation; or the section “Time Passes” in Woolf’s novel To The Lighthouse, a description of the years passing in an empty house, into which human beings scarcely intrude, and the deaths of several significant characters, not to mention the entire First World War, are mentioned in three-line paragraphs, enclosed within square brackets. In the first half of the twentieth century, these writers were presenting human life and action as a pointless transit from birth to death, a parenthesis in an inscrutable, overwhelming space-time. Meanwhile, the philosopher of phenomenology Maurice Merleau-Ponty was speculating along similar lines: “The aspect of the world would be transformed if we succeeded in seeing as things the intervals between things.”
But back in the world of conventional narrative film-making, the decisive, autonomous human actor still dominated all – and still does. This has given rise to a “rule” of editing so widely observed in the West that Edward Dmytryk wasted few words on it when describing the principles of classic narrative style: when cutting together a shot that shows an actor walking out of frame with a shot of the same actor entering the frame, “good editing practice rules that the cut away from the first scene should occur at the point where the actor’s eyes exit the frame”.
This rule says a lot about the prejudices of conventional film-makers – or rather, as Dmytryk would point out, audiences. Once the actor has left the frame – not just that, when the eyes, the things we look at to ensure the identity of the person we’re looking at, to check that they’re paying attention to us, the “windows of the soul”, have left the frame – the editor believes that the audience will lose interest in the shot. When filming a scene between two people at different depths in the frame, convention demands that the camera “pulls focus” to bring the person speaking – more precisely, the person’s eyes – into sharpest focus. Where a person enters frame at a different depth to the background or other preceding object of interest, we expect focus to be pulled immediately onto the actor. Not only are human beings the most important thing in a conventional narrative film, human beings in action – speaking, for instance – are more important than those in repose. If we are shown a shot of a person who is not doing anything in particular, it will be a “reaction shot” in which the character is responding to some other action.
I think that such features of traditional film technique reflect what Jean-François Lyotard called the “pre-modern” individual: “characterised by the self-knowing, punctual, subject of humanism, ‘expressing’ itself, and/or its world (a world simply there, as ‘reality’) via a transparent language.”
Not that cinema was ever restricted entirely to conventional storytellers. Famous examples such as Mechanical Ballet, Earth, Man With a Movie Camera and Un chien andalou would have showed film-makers from the 1930s onwards that cinema need not be the slave of smooth exposition. But the various styles of these films all denied them the unique power of narrative: the ability to encourage an audience to engage emotionally with fictional characters.
Mainstream fiction film-making, meanwhile, established a narrative style strong enough to be considered an uncontroversial norm for the first edition of Karel Reisz’s book in 1953. But by the time the British director Gavin Millar came to expand the book for a new edition in 1968, mainstream European cinema had a new dimension. In Millar’s analysis of a sequence from Michelangelo Antonioni’s The Adventure, released in 1960 and set partly on a rocky volcanic island, he wrote:
It is noticeable that the most impressive “character” in this extract, if we judge by the weight given it, is the landscape. By conventional standards indeed the sequence is “badly cut.” The shots go on absurdly too long – long after the apparent action is over. But…the action is elsewhere.
The island is a major participant in the action…
The objectives of such a film are evidently not plot exposition or entertainment. Millar suggests an alternative organising principle:
…it is in the moments when we are not looking at each other that we reveal most, in the dead time…between deliberate gestures, action or speech that the most significant emotional decisions are taken. It is in these moments that Antonioni is particularly interested.
Here it is in commercial Western cinema, the dead time, the time between events, and the extra-human world, now achieving the respectability of industry-standard production values, international release, critical approval. Antonioni’s camera follows his characters attentively and intimately enough to allow the audience to feel for them, to be interested in what will happen to them. But the film is a cool, distant examination of human behaviour. It requires a different sort of viewing to a more conventional narrative. It’s not a question of brute length: The Lord of the Rings: The Fellowship of the Ring is only two minutes shy of three hours long, well over the running time of The Adventure or almost any Tarkovsky film, but builds up its action-movie pace so effectively that it leaves its audience no time to wonder what the time is.
In Tarkovsky’s films, as in the work of other “modernist” film-makers, the space and time around action is put into the foreground. This is clearest in sequences that concern waiting and boredom itself: peasants and monks in a barn, sheltering silently from the storm in the second episode of Andrei Rublyev – a sequence which is given as much weight in the scene as a jester’s high-energy song-and-dance routine earlier – Burton’s car journey in Solaris, mentioned earlier, the long track-cart ride into the Zone in Stalker. The hallucinatory sound and hypnotic images of the Solaris sequence and the slow dawning of the unnaturalness of the sound in the Stalker interlude make these some of the most memorable moments of the films, suggesting that “dead time” may be in fact when we are most alive: in a heightened state of awareness, as Michael Flaherty said, and free to think our most important thoughts. Not that Tarkovsky was crass enough to tell us what they should be.
Time was the central concept in his analysis of cinema. His raw material was not story, or words, or music, or photography, but time. Discussing the Lumière brothers’ early film The Arrival of a Train in La Ciotat Station – which shows nothing more or less than its title suggests – and the famous story of how the audience ran out of the theatre in panic at the image of an approaching train, he wrote:
For the first time in the history of the arts, in the history of culture, man found the means to take an impression of time. And simultaneously the possibility of reproducing that time on screen as often as he wanted, to repeat it and go back to it. He acquired a matrix for actual time. […] …the one precious potential of the cinema [is] the possibility of printing on celluloid the actuality of time. […] I think that what a person normally goes to the cinema for is time: for time lost or spent or not yet had. He goes there for living experience…
And in his comments on structuralism, he returned again and again to this point:
A film is an emotional reality, and that is how the audience receives it – as a second reality. …cinema and music…need no mediating language. […] I want to emphasise yet again that, with music, cinema is an art which operates with reality.
But this seems to me too simplistic a notion. As Michael Flaherty’s book pointed out, we do not perceive time as clocks do. The motion picture camera records twenty-four images per second and the film projector delivers those images to the screen at precisely the same rate. But the experience of watching those events through the mediation of film is not “actual time” as we would have experienced it had we been bodily witnesses to the real events, whether the events that were staged for filming or the events as they would have been had they occurred spontaneously. Practical evidence of this appears when working with sound: if the sound of a click is added to a silent video at precisely the frame where a switch can be seen being pressed, the sound will seem to come too late upon playback. To make the sound believable, it must be placed about three frames earlier that its fictional visual cause. Other sound effects with clear visual clues – door slams, gunshots – require the same paradoxical treatment. We knit together our experience of unmediated reality differently to the way we perceive moving images on a screen or sounds coming from a loudspeaker.
It seems that Ian Christie could see Tarkovsky’s “reality” in the earlier films, and so felt disappointed when he became aware of evident artifice:
[His insight into time] may also have led him down the dangerous path of ‘aestheticizing’ time by unduly prolonging it in the last films. …Tarkovsky tended to elaborate ever longer shots containing a multiplicity of time traces. The long candle ritual in Nostalgia and the final tableau shot in The Sacrifice are cases in point, where, especially in the latter, narrative and logistical manipulation (the audience has plenty of time to consider both) may have defused any ‘real’ time experience.
I would argue the “real” time experience was inevitably only part of a more complex set of responses to the films, about which more later. The film theorist André Bazin had a more accurate view: time is preserved in film, but not kept alive:
…the cinema is objectivity in time. …Now, for the first time, the image of things is likewise the image of their duration, mummified as it were.”
If we want to give an audience the experience of unmediated, lived perception, we would perhaps do better to return to those, from V I Pudovkin to Karel Reisz, who have claimed this aspiration to virtual reality as the foundation of classic narrative style.
* * * * * * * *
Never a dull moment
For Tarkovsky, there was no such thing as dead time. He filled it to overflowing:
How does time makes itself felt in a shot? It becomes tangible when you sense something significant, truthful, going on beyond the events on the screen; when you realise, quite consciously, that what you see in the frame is not limited to its visual depiction, but is a pointer to something stretching out beyond the frame and to infinity; a pointer to life. 
The balloonist’s point of view in the first episode of Andrei Rublyev
The landscape of the Zone in Stalker
Chris Kelvin’s last day on Earth in Solaris
Despite his reputation for lack of incident, I have never seen a shot in a Tarkovsky film in which nothing happens. At first, however, the films may seem rather mundane. His preferred landscapes appear to be mud and puddles, as in Ivan’s Childhood and Andrei Rublyev, and a kind of in-between scrubland, bumpy rather than hilly, alternating meadows with woods – the landscape of Chris’s father’s home in Solaris, of Mirror, and most notably of Stalker. Like the muted colour palette of his films, and his ordinary-looking actors, he uses nondescript landscapes as a ground for the quiet bravura of other elements. And the drama and beauty he finds in these settings – the mists and changing colours of daylight in Stalker and Nostalgia, the rain among trees in Mirror – thereby appears accidental, as if we have discovered it for ourselves – or so I felt when I first experienced this in watching Stalker. One obvious technique, though, is his particular way of turning any piece of water, however tiny or filthy, into an expanse of limpid gold or silver, likely to harbour abandoned icons (as in Stalker and the submerged church in Nostalgia).
And despite his protestations of ascetic rigour, there are plenty of other tricks at play. For instance, he is not above shock effects:
[In Solaris] in the scene where Hari, locked in the cabin by Chris, bursts through the metal door [rather like Jack Nicholson and his axe in The Shining] and falls senseless to the ground; when she tries to commit suicide by drinking liquid oxygen and then comes painfully and relentlessly back to life [echoes of Frankenstein movies]; and in the atmosphere of desolation and terror on the station – in all these he showed himself in perfect command of the techniques of the horror film.
In the subtler other-worldliness he sought to evoke in Stalker, a supernatural voice from nowhere scares the over-confident Writer into believing Stalker’s warnings about the dangers of the Zone; an ultimately uneventful passage through an underground tunnel is laden with all the suspense a thriller could arouse; a slow track towards the heap of debris in a room reveals it to be an ancient human skeleton.
Tarkovsky regarded an infinitesimal dislocation of the everyday as more threatening and frightening than anything a cunning property-master or special-effects team could ever produce. Suddenly and incongruously, for instance, an ancient telephone rings in the empty, long-deserted ruins. For Tarkovsky, the techniques of terror (for creating a truly chilling Future) are those of estrangement, and ambiguity – not the inexplicable, but the unexplained – sliding into the unsteady sphere of suppositions. The black dog that attaches itself to the Stalker in this dead zone where the only signs of life are the distant voice of a cuckoo or the squeak of some other bird is not, of course, a Faustian poodle, but an ordinary mongrel. Even so, in its silent appearance there is a hint of warning, like a distant echo of some half-forgotten legend.
This is the device that the Russian critic Viktor Shklovsky called ostranemie or “making strange”, renewing the novelty and potency of familiar things.
Tarkovsky also presented things that are evidently supernatural and so already strange. This allowed him to use straightforward depiction to disconcerting effect: as I wrote earlier, the viewer is disoriented at finding no distinction made between images with different degrees of reality. The reappearence of Theophanes the Greek as a ghost, or dream, in Andrei Rublyev; mid-air floating – even when logically justified by zero gravity, it seems more to do with magic than science – in Solaris and The Sacrifice, the ghostly grandmother in Mirror, her vanishing proved for a moment by an evaporating coffee-cup ring on a table-top; a psychokinetic twist at the end of Stalker; all bring the supernatural into the room next door. And Tarkovsky could give an aura to the most mundane object:
Someone remarked at the Cannes Festival that Hari’s knitted shawl in Solaris seemed possessed of an almost magical power as if it had a life of its own. It is true; the shawl has its own plot-line in the film … Indeed, its white and brown wool and its hand-knitted texture really do seem to acquire their own independent existence, partly linked to that of Hari, but not completely so.
Tarkovsky also makes use of virtuoso techniques that draw attention to themselves, thereby breaking another of the rules of classic narrative style. One repeated trick is to have a character exit frame on one side only to re-enter on the other, not only disorienting the viewer still further but also inducing her or him to think about the dash behind the camera that the actor must have taken to achieve the effect. Examples of this can be found in Solaris, when Chris Kelvin first encounters Snaut in his quarters and in the multiple-Hari dream sequence, and in Nostalgia as Gorchakov explores Domenico’s derelict home.
Another favourite self-referential trick is for a travelling shot to return on itself to reveal that something has changed while it was offscreen: in the tracking shot that looks down on Stalker’s family in their shared bed, for instance, the camera begins by observing a glass on the bedside table moving from the vibrations of a passing train, then passes over the sleeping wife and daughter to Stalker awake, then back over the sleeping child and wife – whose eyes are now open a crack – to the bedside table again.
Tarkovsky claimed that there was no allegory in his films. About Stalker, for instance:
People have often asked me what the Zone is, and what it symbolises, and have put forward wild conjectures on the subject. I’m reduced to a state of fury and despair by such questions. The Zone doesn’t symbolise anything, any more than anything else does in my films: the zone is a zone, it’s life, and as he makes his way across it a man may break down or he may come through.
But this seems to give itself away: “The Zone doesn’t symbolise anything…it’s life.” I cannot help but understand Stalker as a modern version of the mythic quest for the Holy Grail. And can the bell-casting episode that is the climax of Andrei Rublyev be anything other than an allegory of art-making, and especially film-making? The boy bell-maker is just a younger, crueller and dirtier version of Marcello Mastroianni’s film director in 81/2: both are heading vast and expensive creative enterprises, knowing that they don’t know what they’re doing, their fear of failure the only thing that keeps them going. Why would we empathise with the relationship between Chris and the non-human Hari in Solaris if it were not a metaphor for the need to accept the essential strangeness that lies between any pair of people? What are we to see in her reverie before a reproduction of a snowy people-filled landscape painting by Pieter Brueghel, on a space station’s library wall, in orbit above an alien planet, but a symbolic proof of her empathy with human history and society, and with Earth itself? And should we not notice that the opening voiceover of the Mirror sequence detailed above implies that “father” might be a stranger who comes to the mother seeking a direction, to flirt, to hint at assault, to break her fence with his weight, hard to trust?
Though Tarkovsky denied it, I think it does no harm to be aware that the stories have a value as allegory or metaphor or myth alongside their own autonomous life. Tarkovsky’s unemphatic, evenly weighted presentation of his material prevents the allegory from becoming too definite or overbearing. And the fundamental mundanity of his imagery, together with the sheer amount of screen time devoted to his characters’ mute suffering, prevents the films from turning into eye candy, keeps us engaged in the people we are watching. Compare this with the films of Peter Greenaway, for instance, in which the rich beauty of composition, colour, lighting and camera movement are often in evident contrast with the flatness of his actors’ performances, and so risk becoming self-conscious, alienating the audience.
Keeping the audience aware, thinking, also appears to be the motivation behind Tarkovsky’s characteristic colour shifts:
The picturesque character of a shot, due often enough simply to the quality of the film, is one more artificial element loaded onto the image, and something has to be done to counteract it if you mind about being faithful to life. […] Perhaps the effect of colour should be neutralised by alternating colour and monochrome sequences, so that the impression made by the complete spectrum is spaced out, toned down.
Tarkovsky’s matter-of-fact treatment of dreams – Ivan’s dream of the horse and girl and apples on the beach in Ivan’s Childhood, Andrey’s fantasy of himself as a northern Christ in Andrei Rublyev, Chris’s dream of the multiple Haris in Solaris – also keep the viewer on their toes. Any shot, any element within a shot, may be readable as dream, magic, fantasy or everyday reality; and perhaps we’ll never know which. Everything has equal status. Everything matters.
* * * * * * * *
Is that all there is?
Little of the above has anything to do with plot. It is the non-narrative stuff that David Bordwell calls “excess”, musing on it for a few paragraphs before declaring it beyond the concerns of his 370-page large-format book on fiction films. Into this bag go all Tarkovsky’s favourite images, the wandering horses and dogs, sudden rainstorms, peeling plaster, fires, puddles shining like a revelation from God, derelict churches, carefully selected paintings, poetry and flowing water: the constant flow of pregnant imagery. This is what Tarkovsky used to replace the constant flow of plot in more conventional films. But then how, with all his evident artifice and trickery, could he claim to be a servant of realism? Perhaps for him, realism meant awakening the viewer to the texture and light and drama of the reality that he or she would otherwise overlook. I’m very glad that he was able to do this. But is that enough? Will that do?
Not for Tarkovsky himself:
Cinema is a high art…It seems to me that the purpose of art is to prepare the human soul for the perception of good. The soul opens up under the influence of an artistic image…
I’m not sure that’s true. I agree with Ian Christie on this one:
To continue basing our aesthetics on ethics since the onset of modernism is also to deny the epistemological, semiotic and indeed ludic aspects of art.
Which leaves me wondering if there is something being avoided in the non-rational approach of Tarkovsky’s films, in the refusal to acknowledge allegory or the fundamental unreality of the cinema image, in characters like Chris Kelvin and Stalker who find their truth by accepting their own irrational foibles, in Tarkovsky’s insistence on the artist’s autonomous intuition:
The director’s personality defines the pattern of his relationship with the world and limits his connections with it; and his choice of those connections only makes the world he perceives the more subjective. …To seek one’s own truth (and there can be no other, no ‘common’ truth) is to search for one’s own language, the system of expression destined to give form to one’s own ideas.
It is the danger of mystical answers, gnostic truths, that by definition are insulated from external challenge. The altered experience of time induced by Tarkovsky films, the openness of perception they renew, the engagement with the present moment, are certainly characteristic of meditative, even mystical experiences, not to mention drugged ones, as Michael Flaherty showed. But isn’t there something shallowly opiate about a vision that turns every muddy puddle into a magic pool? A retreat into beauty, a hiding in the refuge Tarkovsky shows us how to carve out of the present moment, an acceptance of one’s personal crown of thorns, an astrology of one’s personal constellation of doubt? What would a Tarkovsky of the city look like, a Tarkovsky of men who do not leave their families – as the film-maker himself did, and his father before him, and every one of his male protagonists – a Tarkovsky of people engaged in the culture and media and society of their own time, still holding on to their integrity and their poetry? That’s what I’d like to see.
(in order of appearance; full details of films in the bibliography)
Andrei Tarkovsky during the filming of Stalker (from Andrey Tarkovsky, Sculpting in Time, The Bodley Head, London, 1986.
Aleksandr Kaidanovsky as Stalker in the film of that name, communing with the ground of the Zone (still taken from Connoisseur Video VHS release of Stalker).
Keir Dullea as David Bowman in 2001: A Space Odyssey, about to disconnect HAL’s brain (adapted from 2001-background2.jpg from http://www.the-labs.com/Backgrounds/ 2001-background2.jpg, last checked on 4 February 2003).
Donatas Banionas as Chris Kelvin, just arrived to find the Solaris station in bad shape (still taken from Connoisseur Video VHS release of Solaris).
Beginning and end framings of the opening shot of Andrei Rublyev (still taken from VHS recording of BBC2 transmission, I’m afraid I can’t remember when).
The slow-rolling horse that ends the first episode of Andrei Rublyev (source as above).
Margarita Terekhova as the mother in the first shot after the main titles of Mirror (still taken from Artificial Eye VHS release of Mirror).
Anatoly Solonitsyn as the stranger and Margarita Terekhova in the first sequence in Mirror proper (source as above).
The balloonist’s point of view in the first episode of Andrei Rublyev (source as above).
The landscape of the Zone in Stalker (source as above).
Chris Kelvin’s last day on Earth in Solaris (source as above)
Andrey Tarkovsky’s feature films
(full details in the bibliography)
Ivan’s Childhood, 1962.
Andrei Rublyev, 1966.
The Sacrifice, 1986.
The Adventure (L’Avventura), Michelangelo Antonioni (director), 145 minutes, Italy, 1960.
Andrei Rublyev, Andrei Tarkovsky (director), different release versions had lengths of 185 and 146 minutes, USSR, 1966.
The Arrival of a Train in La Ciotat Station (L’Arrivée d’un train en Gare de La Ciotat), Louis Lumière (director), France, 1895.
André Bazin, “The Ontology of the Photographic Image” in What Is Cinema?, edited and translated from French by Hugh Gray, University of California Press, 1967; cited in Maya Turovskaya, Tarkovsky, page xx.
Samuel Beckett, Molloy, translated from the French by Samuel Beckett and Patrick Bowles, Calder Jupiter, Calder and Boyars, London, 1966; first published in French by Editions de Minuit, Paris, 1950.
Blade Runner, Ridley Scott (director), 117 minutes, US, 1982.
David Bordwell, Narration in the Fiction Film, reprinted Routledge, London, 1988; University Paperback, 1986; first published in Great Britain by Methuen, 1985.
Pieter Brueghel, The Hunters in the Snow, painted 1565, panel, 117 x 162 cm, Kunsthistorisches Museum, Vienna.
Victor Burgin, The End of Art Theory: Criticism and Postmodernity, Macmillan Education, Houndmills and London, 1986.
Un chien andalou (An Andalusian Dog), Luis Buñuel and Salvador Dalí (directors), 16 minutes, France, 1929.
Ian Christie, “Journey Into Inner Space”, Stalker, sleeve notes for VHS release, Connoisseur Video, no date.
Edward Dmytryk, On Film Editing, Focal Press, Boston & London, 1984.
Earth (Zemlya), Alexander Dovzhenko (director), 54 minutes, USSR, 1930.
81/2 (Otte e mezzo), Federico Fellini (director), 138 minutes, Italy, 1963.
Sergey Eisenstein, Film Sense, Harcourt, Brace and World, New York, 1947), Film Form, Harcourt, Brace and World, New York, 1949; cited in James Monaco, How to Read a Film.
Michael Flaherty, A Watched Pot: How We Experience Time, New York University Press, 1999.
The Great Warrior Skandberg, Sergey Yutkevich (director), USSR/Albania, 1954; cited in Maya Turovskaya, Tarkovsky.
Stephen Heath, “Narrative Space”, in Questions of Cinema, Macmillan Press, London and Basingstoke, 1981.
Ivan’s Childhood (Ivanovo detstvo), Andrei Tarkovsky (director), 95 minutes, USSR, 1962.
The Lord of the Rings: The Fellowship of the Ring, Peter Jackson (director), 178 minutes, US/New Zealand, 2001.
Jean-François Lyotard, La condition postmoderne, Les Editions de Minuit, 1979; quoted in Victor Burgin, The End of Art Theory.
Man With a Movie Camera (Chelovek e kinoapparaton), Dziga Vertov (director), 60 minutes, USSR, 1929.
Mechanical Ballet (Ballet méchanique), Fernand Léger (director), 15 minutes, France, 1924–1925.
Maurice Merleau-Ponty, “Cinema and the New Psychology”, (“Le cinéma et la nouvelle psychologie”), in Sens et non-sens, Gallimard, Paris, 1948; quoted in Stephen Heath, “Narrative Space”, page 62.
Metropolis, Fritz Lang (director), Germany, 147 minutes (reconstructed version), 1926.
Mirror (Zerkalo), Andrei Tarkovsky (director), 106 minutes, USSR, 1974.
James Monaco, How to Read a Film: The World of Movies, Media, and Multimedia, third edition, Oxford University Press, 2000.
Nostalgia (Nostalghia), Andrei Tarkovsky (director), 126 minutes, Italy, 1983.
V I Pudovkin, Film Technique and Film Acting, one volume, Grove, New York, 1970; first published separately in Russian, Soviet Union, 1926 or 1929 (different sources give different dates) and 1937 respectively; quoted in James Monaco, How to Read a Film.
The Red Desert (Il Deserto Rosso), Michelangelo Antonioni (director), 116 minutes, Italy/France, 1964.
Karel Reisz and Gavin Millar, The Technique of Film Editing, Focal Press, London, 1968; first edition 1953.
The Sacrifice (Offret), Andrei Tarkovsky (director), 149 minutes, Sweden/France, 1986.
The Shining, Stanley Kubrick (director), 146 minutes, GB, 1980.
Solaris, Andrei Tarkovsky (director), 165 minutes, USSR, 1972.
Stalker, Andrey Tarkovsky (director), 161 minutes, USSR, 1979.
The Star Trek films, various directors, US, 1979–2002.
Arkady and Boris Strugatsky, A Roadside Picnic, translated from the Russian by Antonina Bouis), Penguin, Harmondsworth, 1979.
Andrei Tarkovsky, with, he acknowledges, considerable help from Olga Surkova, Sculpting in Time: Reflections on the Cinema, translated from the Russian by Kitty Hunter-Blair, The Bodley Head, London, 1986; original Russian title Sapetschatljonnoje Wremja; first published in German as Die Versiegelte Zeit.
The Thin Red Line, Terrence Malick (director), 166 minutes, US, 1998.
Maya Turovskaya, Tarkovsky: Cinema as Poetry, translated by Natalie Ward, edited and with an introduction by Ian Christie, Faber & Faber, London and Boston, 1989.
2001: A Space Odyssey, Stanley Kubrick (director), 141 minutes, GB, 1968.
Virginia Woolf, To The Lighthouse, Grafton Books, London, 1977; first published by The Hogarth Press, London, 1927.
Zabriskie Point, Michelangelo Antonioni (director), 110 minutes, US, 1969.