Ozu Yasujirô: the master of time

I like to quote Wim Wenders’ statement at the beginning of Tokyo-Ga:

“If in our century, something sacred still existed, if there were something like a sacred treasure of the cinema, then for me that would have to be the work of the Japanese director Yasujirô Ozu… For me never before and never again since has the cinema been so close to its essence and its purpose: to present an image of man in our century, a usable, true and valid image in which he not only recognises himself, but from which, above all, he may learn about himself.”

What can I add? What seems like hyperbole is actually true. Perhaps I can supply the arguments Wenders didn’t feel were necessary.

What everybody notices first in Ozu is the visual form. He apparently decided at the very beginning of his filmmaking career to adopt his own cinematic language, an idiolect that is both conservative and radical.

It is conservative because the choices within his system are severely limited and because in some respects it is just a purification of the standard continuity system. Each scene follows the standard pattern: in, out, repeat if necessary. The scene begins with a long shot that establishes the characters, then moves into medium close-ups. If it is a lengthy scene, it will cut to the long establishing shot again and then back to the close-ups. At the end, it will return to the long shot.

The average shot length in his films adheres closely to the norms that prevailed in Japan and Hollywood, and Ozu keeps the duration of the shots within a film remarkably consistent: there are no long takes and very few noticeably quick shots. Most of the cuts are ‘return cuts’, to borrow Klaus Wyborny’s term – that is, they return to a shot already shown.

It has been written that Ozu pared down this system further by gradually eliminating camera movement, fades, and dissolves, but these figures appeared only exceptionally even in his first films. From this description an Ozu film might seem like a highly conventionalised TV series, such as Dragnet.

But he does everything wrong; he breaks every rule of conventional cinematic grammar. He always puts the camera too low, but he doesn’t angle it up, so the subject of the shot always occupies the top of the frame. The eye-line matches are always wrong.

A fundamental rule of standard continuity requires that the camera always stay on one side of an axis created by the actors’ gazes. Thus the camera may not be moved 180 degrees from one set-up to another; it must always stay within a semi-circle on one side of the axis.

Ozu doesn’t simply violate this rule, he overturns it: every cut crosses the axis of the gaze. Every cut is a multiple of 45 degrees, most often 180 degrees (especially when he cuts on an action match) or 90 degrees. The standard continuity system was developed to make cuts invisible, to the conscious mind at least. Ozu denaturalises the cuts, making them as noticeable as possible.

Then there are the shots of ‘empty spaces’: still lifes, unpeopled interiors, building facades and landscapes. They are Ozu’s trademark, the one part of his system that has been adapted by modern European and Asian filmmakers, and they have given his interpreters a great deal of trouble when they try to assign them a meaning.

In his essential book on Ozu, David Bordwell calls these empty spaces “intermediate” because these shots generally occur between scenes (although sometimes as cutaways within scenes). But they are not establishing shots, although some shots in a series may serve that function. They have an autonomy that led Noël Burch to call them extradiegetic, that is “on another plane of reality”, although they exist in the same space as the characters. Perhaps it suffices to define them simply by the absence of the characters and the suspension of the narrative.

In his book on cinema, Gilles Deleuze claims these shots are direct images of time, “the unchanging form of that which changes.” They do give Ozu’s films a sense of durée (a word we need in English since it is not synonymous with ‘duration’), that is, lived time, “concrete time” (André Bazin), “dialectical time” (Jean-Paul Sartre).

This may seem paradoxical: ever since Bazin, everyone has assumed that montage destroys the sense of temps durée, by “cutting the world up into little fragments”, as he put it. But Bazin never saw an Ozu film. By showing what endures, if only for a few seconds, Ozu suggested another possibility: the cinema is closest to its own natural movement when it doesn’t feel obligated to record external movement.

Ozu’s system is remarkably coherent, as complete as the standard continuity system it replaces, which was the collective work of many filmmakers. I want to call it ‘serialist’ because of its precise, nonhierarchical organisation. Bordwell prefers to call it ‘parametric’.

The film director Masahiro Shinoda, who worked as an assistant to Ozu in the 1950s, suggested the purpose of this system in talking about one of its aspects: “The reason [Ozu] placed the camera so low was to prevent it from having a human viewpoint.” Shinoda is precisely right. If the cinema as we know it is a continuation of the novel by other means, that is, if it projects a world already endowed with consciousness and meaning, a world seen from a subjective perspective, then Ozu refused this cinema and managed to create an alternative.

For this purpose, Ozu also had to create a dramaturgy to complement his visual system, to avoid the human viewpoint. Although critics have neglected it, his dramaturgical system is even more original and remarkable than his visual system. It’s like Brecht’s epic theatre, but better: more rigorous, more radical.

Ozu never focalises a film through a single character. It seems to go without saying that he never used flashbacks (except in his first, now lost film, a period movie titled Sword of Penitence) or voiceover narration, the means by which the cinema gives us a privileged access to the consciousness of a character. There are no subjective images of any kind, no images that purport to show what a character is imagining or dreaming or remembering or anticipating. In his early films, there are occasional images that show what a character is seeing, but he soon rejected these point-of-view shots.

Of course, there are major characters and minor characters, but he avoids the usual sense of subordination and creates a kind of equality. In Floating Weeds (1959), the minor characters dominate the first reel; two of the four major characters appear for the first time only in the sixth scene. In any case, we don’t see the minor characters from the viewpoint of the protagonist.

In his late films, co-written with Kogo Noda, there is a serialist patterning of encounters, most evident in Tokyo Story with its symmetrical structure. In the first part of the film, he works through the set of possible encounters among the characters, organising them like a tone row in serial music, and inverts them in the second half with variations to register the illness and death of Tomi.

Wenders says the body of Ozu’s work “depicts the transformation of life in Japan. [They] deal with the slow deterioration of the Japanese family and thereby with the deterioration of the national identity.” I would say they depict the development of capitalism in twentieth-century Japan or, more precisely, the disappearance of the old petty bourgeoisie and its gradual proletarianisation under the pressures of what we now call globalisation. These are big words, but no others will do.

This theme is evident enough in Ozu’s films of the 1930s. His first films dramatise the plight of university graduates unable to find a job (at the time, two-thirds of them could not find work). Those who have a job must choose humiliation or dismissal. In Tokyo Chorus (1931), Shinji gets fired when he protests the unjust firing of an older worker. His next job will be passing out handbills for his ex-teacher’s curry rice shop. Women may choose a literal prostitution, like Yasue in Walk Cheerfully (1930), Mitsuko in A Mother Should Be Loved (1934), or Chikako in Woman of Tokyo (1933), who gets in trouble with the police only because she is unlicensed and thus evading taxes.

With Passing Fancy (1933) and An Inn in Tokyo (1935), Ozu turned his attention to those already proletarianised, those for whom unemployment brings destitution, but in his working-class and lumpen-proletarian films, there is a sense of solidarity and community lacking in his films about unemployed college graduates. In both films, the illness of a child brings everyone together, and they are willing to sacrifice anything to pay the hospital bill.

And so is Ryosuke in The Only Son (1936), the first working-class college graduate in Ozu’s films. Like the characters of Passing Fancy and An Inn in Tokyo, he lives on the semi-rural industrial margins of Tokyo, near the city’s garbage incinerator, and he ekes out a living as a night-school mathematics instructor.

His sense of disappointment is particularly bitter because his mother has sacrificed everything for his education – her home, her land, her chance to retire from work in the silk mill (Ozu’s shots of the work process are a model of economy and clarity). Both he and his mother must confront his situation when she comes to Tokyo for a visit. Could there be any conclusion any resolution? After she leaves, he vows to try harder, and she goes back to scrubbing floors in the mill.

Another trip to Tokyo provides the basis for Tokyo Story, Ozu’s most savage film and his most elegant: all the elements of his system come together in a perfect equilibrium. A retired school teacher Shukichi Hirayama and his wife Tomi come to Tokyo from the provinces to visit their two grown children, Shige and Koichi, and their daughter-in-law Noriko, whose husband is still missing from the war.

They discover that their children aren’t doing as well as they had thought: Koichi runs a small medical clinic in an industrial area, and Shige owns a small beauty salon in a poor neighbourhood. For the children, their visit is a nuisance. Koichi is on call to visit patients day and night, and Shige can’t escape her salon. Only the childless, widowed Noriko, employed in an office, can take time to host them properly.

Thus many viewers have come to regard Noriko as representing good and Shige and Koichi as bad. But they misread the film. Ozu’s films are not psychological nor are they judgmental. For Ozu, what counts is the objective situation of his characters, the material conditions in which they make their lives. Their precarious economic situation does not allow Shige and Koichi to serve their parents as they should.

In one scene, Shukichi goes out drinking and ends up passed out in a chair in Shige’s salon. There is a parallel scene in Ozu’s last film, An Autumn Afternoon (1962). Haruko Sugimora, who plays Shige in Tokyo Story, again plays the daughter of a retired middle-school teacher. Like many retired teachers in Ozu films, he must run a small business to survive. He and his daughter run a noodle shop in a desolate, industrialised corner of Tokyo.

Some ex-students throw a party for him, and two of them bring him back to the noodle shop after he has fallen into a drunken stupor. The daughter reacts stoically at first, but after her father passes out, she breaks into tears. She has given up the chance to get married and have her own life in order to serve a broken, hopelessly alcoholic father. She has become a “withered spinster”, as one of the ex-students puts it. Perhaps Ozu included this subplot to write an ending to Tokyo Story. He shows the fate Shige has tried to avoid.

A new middle class does emerge in some of his 1950s films, the ‘salarymen’ who work for large unnamed corporations, free from the immediate economic worries that plagued their predecessors in the 1930s. But there is never enough money. In I Was Born, But… (1932), the two boys go on a hunger strike to protest their father’s subservience to his boss; in Good Morning (1959), they strike to demand a TV set. The young couple Koichi and Akiko in An Autumn Afternoon lack the money to buy a refrigerator, or even a set of golf clubs.

Ozu’s critical treatment of capitalism has led some viewers and critics to think he upholds feudal values. They have him mixed up with a Kurosawa or a Capra, who do propose that a strong feudal leader can overcome the evils of unbridled capitalism when he can be persuaded to make the effort. Ozu simply shows the change and doesn’t try to imagine the cure. He is as far from feudalism as he is from communism (although I am intrigued by the portrait of Karl Marx that appears on the wall of the protagonist’s apartment in The Lady and the Beard (1931) – I’ve only seen the still – and by the claim of film historian Tadao Sato that in the original script for Woman of Tokyo, Chikako gave some of her earnings as a prostitute to the Communist Party as well as to her brother to pay for his education).

I agree with Pedro Costa who said, “for me, the true Japanese documentaries are by Ozu.” In the same talk, at the Film School of Tokyo, he said, “the primary function of cinema is to make us feel that something isn’t right.” That is, cinema can and should be didactic. It is by becoming didactic that it shows us something we know but makes it strange and new. It is in this sense that the images created by Yasujirô Ozu are not only true and valid, but useful. Is this why I can watch them over and over with undiminished pleasure?