Yasujiro Ozu: The Syntax of His Films Author(s): Donald Richie Source: Film Quarterly, Vol. 17, No. 2 (Winter, 1963-1964), pp. 11-16 Published by: University of California Press Stable URL: http://www.jstor.org/stable/1210862 Accessed: 17-10-2017 15:09 UTC JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://about.jstor.org/terms University of California Press is collaborating with JSTOR to digitize, preserve and extend access to Film Quarterly This content downloaded from 147.251.99.131 on Tue, 17 Oct 2017 15:09:17 UTC All use subject to http://about.jstor.org/terms 11 DONALD RICHIE Yasujiro Ozu: The Syntax of His Films In the Autumn issue of Sight and Sound, Tom Milne surveys Ozu's work on the basis of a series of films recently seen in London. Five Ozu films will shortly tour the United States (Late Spring, Tokyo Story, Early Spring, Good Morning, and Late Autumn) and the following stylistic analysis explicates some of the methods we will soon be able to observe in these pictures. With Ozu, as with Antonioni or Resnais, the critic may speak of grammar, of vocabulary, of syntax-something which one cannot do with Mizoguchi, with Bergman, or even with Truffaut, intuitive directors all. Ozu is not an intuitive film artist, he is a master craftsman; for him, film is not expression but function. In an Ozu film, as in Japanese architecture, you can see all the supports, and each support is as necessary as any other. He uses neither paint nor wallpaper; he uses natural wood. He makes a film as a carpenter makes a house. The finished object one may measure, one may inspect, one may compare. But within this object, as within the house, lives the human, the immeasurable, the nonfunctional. It is this combination of the static and the living, of form and content, which makes the films of Ozu the compelling emotional experiences they are and, at the same time, the wonderfully hand-tooled containers which they also are. The opening funeral scene in AxxBronR (Setsuko Hara, Yoko Tsukasa, and Ryu Chishu) ~?s- -;;;I~ I ~wa~psli~e~s~n~ d ~ n: ai-?8' p ~1? This content downloaded from 147.251.99.131 on Tue, 17 Oct 2017 15:09:17 UTC All use subject to http://about.jstor.org/terms 12 I OZU GRAMMAR Ozu, like Antonioni, knows that plot is worthless because it is manipulated. It is life used and consequently untrue: life must at least appear to be gratuitous to appear true. Antonioni believes that "the episode is the only fit unit for film" and this Ozu too believes-with the difference that he believed it thirty years before Antonioni did. For this reason, though the chronicle of an Ozu picture is fairly straightforward, you cannot make a pr6cis. Everything Ozu-like evaporates if you merely tell the story, for the reason that story (or, more often, merely anecdote) is but a pretext for the film, the real reason for which is revelation of character. Ozu therefore restricts content (a plot is an indulgence-it is too easy) and, in the same way, he restricts his technique: hence his celebrated avoidance of these elements of film grammar which other directors find indispensable. Dissolves are "cheating"; fades are "merely attributes of the camera"; dollies, pans, etc., are "uninteresting." The only punctuation which Ozu allows himself is the simple cut; the only camera position, that of the person seated upon tatami, his eyes about three feet from floor level, the traditional attitude for talking, for watching, for listening. He allows himself three kinds of shots-the classical three of primitive cinema. (1) The long shot is used to show solitude, precisely because it isolates; or humor, for it isolates and makes apprehendable; or aesthetic beauty, because it gets us far enough from it to see it all. (2) The middle shot, the standard unit of the Ozu film, is the "business" unit during which most of the action occurs. (3) The close-up, used for heightened moments, either with or without dialogue, is used rarely and never allowed to enlarge itself into the "big" close-up. Each shot has its place within the sequence and the order of the sequence is usually 1 - 2 - 3 - 2 - 1. Musically, it is the a-b-a pattern, simple binary form, one of the most immediate and satisfying formal experiences possible, through reason (in films as in music) of its being firmly apprehendable, and perhaps for the more metaphysical reason of its being circular: a balanced, continuous geometrical form congenial to the human mind. The sequence in Ozu is the paragraph (the Ozu film has no "chapters") and within these paragraphs the shot becomes the "sentence." STRUCTURE Just as the sequence in Ozu is circular, so is the basic form of the entire picture. It would be difficult to find an Ozu film that did not end where it began-though such an atypical picture would be Soshun (Early Spring 1956). Often, indeed, this effect of form becomes "formal," even-in the best sensemannered. The neighbor lady appears twice in Tokyo Story (Tokyo Monogatari-1953), once in the first reel, and once in the last. In the first the old couple is preparing their trip and she comments upon it; in the last reel the wife is dead, the husband will remain where he is alone (the opposite certainly of travel) and this too she obliquely comments upon. Ohayo: (Good Morning-1959) like its ancestor, Umareta a Mita Keredo, (I was Born, But . .. (-1932) ends precisely where it began and the adventures of the little boys (very meaningful in the latter film; merely comic in the former) count for nothing other than the emotional experience which they give us. In most Ozu films the structure presumes this "return" and it is this which makes the final reels of these pictures so compelling. The idea of the "return" (like the idea of the circle) is something which all of us find emotionally compelling-a somewhat common, if not vulgar, example of its great filmic effect is in the two celebrated 180 pans (before and after Micheline Presle's death) in Le Diable au Corps. Musically, it is more instantly apprehendable. The master of the "return" is Mozart, because of the freshness, the surprise, the astonishing "newness" of the sound when he completes the return in a rondo. For one thing we are back in the home key, always a grateful feeling; and for another we return home (as in the finale of the Jupiter) doubly This content downloaded from 147.251.99.131 on Tue, 17 Oct 2017 15:09:17 UTC All use subject to http://about.jstor.org/terms OZU 13 enriched; we are surprised by the new beauty of the familiar. The formal parallel to Ozu is precise because the effect is never merely formal, the "return" is not contrived (as it is all too often in the pictures of, say, John Huston), nor is it for its own effect (as in Carn6), an aestheticism for its own sake. Rather, with true art, that art which hides art, Ozu triumphs in making this necessary formal device appear natural. Perhaps the main reason for this is that the structure of the Ozu film appears so logical-there is a definite reason for each shot. Even in Antonioni (who shortened L'Avventura himself) shots and sequences may be removed. Remove one shot from an Ozu picture and you damage it irreparably. The logic which controls the structure is responsible for this. Take, for example, the extremely logical way which Ozu will move us from one sequence to another. He feels it important that we keep our bearings, that we do not get lost. In Akibiyori (Late Autumn- 1960), as in most Ozu films, there is, at the beginning of a sequence, a "still" shot showing either the location (the bridge) or the place, the wall on which there is a picture of a bridge and over which are the ripples of reflected water from the river) or the particularity (an object in the room itself, a lamp or a single vase) - we always know where we are. This is true even if great distances are involved: the Kyoto sequence of Banshun (Late Spring-1949) begins with a "still" shot of typical and unmistakable Kyoto hills; in Tokyo Monogatari, the cities of Onomichi, Tokyo, Atami, and Osaka, are always introduced by, respectively, still shots of, the strait and the graveyard, factory chimneys, the wharf and beachfront, and Osaka Castle. Often these make a comment on the film (as does the Onomichi graveyard); thus in Soshun, in order to get us from the suburbs to downtown Tokyo, we are given three shots: the first shows the suburbs, a few men in white shirts (in Japan an office worker actually is a "white-collar" worker) walking to the station; the second shows, somewhere nearer the city, many more workers going into a larger station; and, the third, Tokyo Station itself, with literally thousands of men going to their offices. Thus we are bodily moved and at the same time are shown something relevant to the ethos of the film, which is concerned with the anonymous life of one of these workers. It is telling, too, that in these three shots Ozu deliberately chose for the second a location which is actually in a direct line from the first location. It would be possible to see this if you yourself actually took the train trip. Ozu never lies about geography, unlike any other film director now alive. The taxi-ride in the first reel of Ochazuke no Aji (The Flavor of Green Tea Over Rice-1952) is literal, the visual continuity outside the window is just as it really would be; the train-ride to Tokyo in the opening reels of Banshun is chronologically accurate; so is the hiking trip in Soshun -first the island of Enoshima in the background and then Ozu computes his time and for the next shots the walkers are passing the oddly-shaped rock which only a person familiar with this coast would know is precisely where he shows it to be. This logical concern for geography is worth commenting upon because it is a strong indication of the scrupulousness of Ozu-a quality which is most noticeable in his treatment of character, and in his editing. EDITING All of Ozu's technique has only one obiect -the revelation of character. His waiting, listening camera records, not the heights of emotion but those moments, those signs, which both precede and come after such momentsthese little tropisms through which true emotion is to be apprehended. The "portraits" which he thus achieves of character are typical of the richness of Ozu's world: the fathers alone at the end of Tokyo Monogatari, Hitori Musuko (The Only Son-1936), Tokyo Boshoku (Tokyo Twilight-1956), or Higanbana (Equinox Flower-1958); the pleasure of the mother listening to the radio in Higanbana, of the father in the train in the same film, of the This content downloaded from 147.251.99.131 on Tue, 17 Oct 2017 15:09:17 UTC All use subject to http://about.jstor.org/terms OZU ,,,, ~~: a s The married couple (Ryo Ikebe and Chikage Awashima) in SosHuN. father at the Noh drama in Banshun-one could continue such a list indefinitely. But one of the ways in which he controls this presentation is in his placement of the cut. Just as he refuses to use (for to use is to misuse) plot or story and so falsify his film, so he refuses to use dialogue, to use character study. Both serve their function but both appear to be gratuitous. Within the a-b-a pattern of the Ozu sequence, the cut can occur some time before the story point is made and, also, some time after it has been made. Most directors allow their characters only enough time to contribute one more link in the story-chain and will then efficiently but also ruthlessly clip off their film-life in the interests of such abstract considerations as story-continuity or what is called a tight film. By contrast Ozu's scrupulousness is astonishing, as is his tact. Necessary dialogue will be given but will occur only in the center (the "b" section) of the sequence. The 1959 Ukigusa (Floating Weed) is filled with marvelous examples of this but perhaps the most perfect is the very long Nohwatching in Banshun. There is a point to be made (silently: the daughter sees her father nod in greeting to the woman she suspects will become her stepmother) but it only occupies about a half-minute of this four-minute scene. The rest of the time we watch the father, his pleasure, his delight in the Noh. Both before and after the anecdotal point of the scene, the camera stays with the father whose supreme delight, and our delight in his, is the real, the emotional point of the shot. Ozu places his cut after the emotional point and not the anecdotal point of the scene is made-and he almost never cuts directly from either dialogue or action. Tactfully, he waits for his characters to fall into repose. Only in moments of great emotion (which he must show), weeping in Banshun and Tokyo Monogatari, the completely atypical "romantic" kiss in Soshun, does he cut from action, and then out of a sense of discretion-like that discreet moment when Antonioni's camera can no longer bear to watch the protagonists of La Notte and consequently turns away. Ozu refuses to "take advantage" of his characters in the way that most directors are only too delighted to do. His scrupulousness is extreme and the placing of the cut within the body of the picture shows this. Again, like Mozart, he knows precisely when to end. TEMPO Ozu's time is not clock-time, it is psychological time, and he will give any length so long as there is a point of character involved. But such revelation of character need not be direct. Ozu's films are almost as noteworthy for what they leave out as for what they include. His treatment of time as a continuum, for example: in Banshun the daughter visits her friend, then they go to another room but the camera remains behind and its only ostensible reason is to watch the clock (in the background) and to listen to it strike four o'clock; at the beginning of Akibiyori (the scene is in a temple) the characters leave, but the camera remains to listen for the striking of a gong (meaning that funeral services are about to start), then there is a cut to the temple where we see the second stroke of the gong; in Tokyo Monogatari just before the very touching scene between mother and daughter-inlaw, the camera places itself in the outside corridor (a geographically-placing shot) and listens to ten of the strokes of midnight, only entering the room with the characters in time for the final two strokes. Ozu, it will be apparent, observes chronological time only for its This content downloaded from 147.251.99.131 on Tue, 17 Oct 2017 15:09:17 UTC All use subject to http://about.jstor.org/terms OZU 15 psychological effect-the effect upon his characters and upon us. In the same way he will interrupt or punctuate clock-time. This is seen mostly in his placing shots. In Soshun there are five identical set-ups (the outside of the couple's house) and each is held for at least fifteen seconds. These perform the double function of "placing" us geographically and, at the same, indicating that "an amount" of time has passed. The "actual" time of scenes which are chronological is rarely interruptedthough it occasionally is, as in Soshun where, perhaps in order to induce an actual feeling of alienation which is one of the moral points of the scene, Ozu cuts from the middle of the mah-jong scene out into the empty corridor before returning us again to the (uninterrupted) game they are playing. There is another kind of tempo in Ozu's films, however, and this is within the sequence itself. The order is almost invariable and put into musical terms it would be: 1. the placing shot, often largo with no one in it; or, the establishing (long) shot, with characters in repose: 2. people; or, preliminary moderato assai action: 3. dialogue-the "anecdotal" allegro point: 4. the point made, the "aft- allegretto er" talk, often humorous: 5. the return to repose and/ poco a poco or the final and/or next larghetto placing shot: Just as the visual form is the binary a-b-a, so the temporal form is slow-fast-slow. Each sequence may follow this pattern precisely because the sequence pattern is not the story pattern. Though the length may vary, the balance of the sequence is usually the same. One of the reasons Ozu can do this is that he is not concerned with the past-only with the present. His characters, no less than Antonioni's, are living in the now, and they have no history (certainly Ozu's purposeful failure to mention the all-important dead mother in Banshun until the final reel is just as astonishing, and just as right, as Antonioni's apparent lack of curiosity as to what happened to Anna in L'Avventura.) And when a person dies in Ozu's world (which is often) they are merely and instantly gone. There are no ghosts in Ozu as there are in Resnais and Bergman. The past barely exists in Ozu. Tokyo Monogatari is about the natural advisability of forgetting the dead (daughter-in-law forgets dead husband; old man will forget dead wife) just as much as L'Avventura is about the horror which Claudia and Sandro feel at forgetting Anna. The difference is that Ozu's people accept this from the beginning and Monica Vitti (in both L'Avventura and L'Eclisse) must "learn" to accept-she does not know this truth. The length of her education is the length of the film. Too, Ozu would not be concerned with a tableau at the end which contrasted Sandro with a blank wall and Claudia with a dormant, or dead, volcano. SCENE The reason is that Ozu has also refused himself what has now become the most popular way through which directors may directly comment: the placement of people within a scene. When Ozu photographs his people from the back, or shows them facing opposite directions, it does not necessarily mean a direct comment, as it would in Bergman, in Fellini, in Ichikawa. People kneeling side by side mean nothing beyond the image itself-just as those who read symbols into Ozu's work (the dropped apple peel, the final shot of the sea in Banshun) are on the wrong track. Direct comment, symbolic scenes-these are alien to a director of Ozu's sensitivity precisely because they constitute an unfair comment on character-unfair because they are the kind of comment which attempts to sum up something as complicated as a character with something as simple as a symbol. Ozu prefers something more subtle-the still-life. The hanging lanterns in the later Ukigusa, the flowerarrangements of Higanbana, the single vase in This content downloaded from 147.251.99.131 on Tue, 17 Oct 2017 15:09:17 UTC All use subject to http://about.jstor.org/terms 16 OZU the darkened room in Banshun-what do these mean? They are apparently still-lives, objects, photographed for their own beautiful sakes. This is not true, however. Take the vase, for example. Father and daughter (soon to part since she will get married) are going to bed. They talk about what a nice day they had (and Ozu very carefully refrains from their saying that it is their last day togetheras it is) and then the daughter asks a question in the darkened room, lying next to her father, and gets no answer. From here on: a shot of the father asleep, a shot of the daughter looking at him; a shot of the vase in the alcove, and over it one hears the gentle sounds of the father's snoring; a shot of the daughter, halfsmiling; a long shot of the vase, almost ten seconds; a shot of the daughter, her mood entirely changed, near tears. The vase serves the function of pivot. It means nothing in itself (except perhaps repose, sleep); it is, rather, a pretext, a reason for an amount of elapsed time, an obiect, something to watch, during which the feelings of the daughter change. It is difficult to say why this is more satisfying than would be the ordinary way of doing it-our actually watching the change on the daughter's face. Perhaps the reason is that Ozu imposes a kind of impersonality, a kind of coolness, between the daughter and ourselves. Not by seeing her but by seeing what she sees (a vase, alone, solitary, beautiful) we can more completely, more fully comprehend and hence feel what she herself is feeling. Again and again in Ozu this technique occurs (in the Noh-watching scene from the same film, you watch a good portion of the Noh itself-even after it is over, the next shot, a placing shot with trees, carries under it the music of the Noh) and we feel by objectively sharing the object of the emotion and not by subjectively observing this mere emotion upon the face of the character. This is a kind of impressionism in that Ozu brings us both the impressions and the things which created them, combining them in such a way that the impression upon us is the same as that upon the characters. This is one of the reasons for the flavor of the Ozu film. He is not concerned with action. He is concerned with reaction. Ozu almost never makes a "story" point visually; he makes it verbally, in the dialogue. And he almost never makes a "character point" verbally; he makes it visually. Did the old people in Tokyo Monogatari have a good time on their trip? They say they did; their children say they did; everyone appears to believe that they did-and probably this is sincere, everyone (including themselves) agreed to think that they did. This is one of the points of the story. Yet, we know the truth. We have been shown it rather than told it, we have read it in their faces-we know it by having seen everything they have seen and then seen their faces. We have been seduced into comparison-and this is the function of the scene in Ozu's pictures. Ozu's method, like all poetic methods, is oblique. He does not confront emotion, he surprises it; he restricts himself precisely in order to achieve profundity, in order to transcend these limitations; his formality is that of poetry, a context which surprises and hence destroys habit and familiarity, returning to each word, to each image, a freshness which was its originally. In all this Ozu is very close to the sumi-e ink-drawing masters of Japan, to the haiku poets, and this the Japanese somewhat dimly understand, calling Ozu "the most Japanese of all Japanese directors." Ozu's cinematic syntax is as it is because he has found through trial and error (as have all master craftsmen) that it best creates the container, the structure, which must best contain, must best present, must best preserve this revelation of human character because it is human character, these observations which are moral because they are true. This content downloaded from 147.251.99.131 on Tue, 17 Oct 2017 15:09:17 UTC All use subject to http://about.jstor.org/terms