The new Criterion Collection 2-disc set of Yasujiro Ozu's
A Story of Floating Weeds (1934) and
Floating Weeds (1959) provides a fascinating opportunity to observe how a director treats the same story at opposite ends of his career. Both films bear the distinct stylistic signature and basic thematic concerns of Ozu, who cultivated what is surely among the most immediately recognizable bodies of work in the history of cinema, but there are subtle differences between them that demonstrate how his basic approach to filmmaking evolved over time.
Today Yasujiro Ozu (1903-1963) is commonly regarded as one of the greatest of all Japanese directors, but his films were not widely seen in the West until the 1970s. This was perhaps because, as has often been said, the Japanese regarded him as the most purely "Japanese" director and thus as someone whose work wouldn't translate well to a Western audience, or because the contemporary domestic dramas in which he specialized lacked the exotic glamour of period costume films by directors such as Akira Kurosawa and Kenji Mizoguchi. Regardless, Ozu's keen observations on everyday life have proven to hold universal human appeal. His best-known work is unquestionably
Tokyo Story (1953), which recently earned the number 5 spot on the
2002 Sight and Sound Critic's Poll for the top ten films of all time; Ozu himself was counted among the top ten directors in the same poll. His films are almost all set in the present and tend to focus on family interactions, or as Donald Richie succinctly puts it, "the dissolution of the family." His directing style is marked by generally low camera placement (the camera is often just above floor level), lack of camera movements such as pans or tracking shots, and the use of straight cuts rather than wipes or dissolves for punctuation between scenes. Each shot in an Ozu film is precisely composed to the extent that, as Roger Ebert points out in his audio commentary for
Floating Weeds, the composition of the shot may even take precedence over continuity of object placement between shots. Given the stylistic consistency of his work across much of his career, Ozu characterized himself with typically self-deprecating wit as a "tofu-maker." Yet within the alleged sameness of his films we find rich visual beauty, acute character study and, ultimately, great depth of feeling.
The revelation of the set is the 1934 film,
A Story of Floating Weeds, which is the first of Ozu's silents to be released on DVD. (Another important silent,
I Was Born, But... (1932) was previously released on VHS.) The Japanese, it should be pointed out, did not start making sound films until 1935 and Ozu's first sound feature was not released until 1936.
A Story of Floating Weeds, however, feels quite modern in its sensibility thanks to its natural performances and sophisticated editing style.
The plot is simple: an itinerant acting troupe arrives in a small town. The head of the troupe, Kihachi, visited the town years ago and fathered a child out of wedlock. The mother, Otsume, has told her son Shinchiki that his father passed away years ago and that Kihachi is his uncle. Otaka, Kihachi's current mistress, becomes jealous when she finds him spending time at Otsume's house. To exact revenge she pays Otoki, a younger actress in the troupe, to seduce Shinchiki. However, Shinchiki and Otoki fall in love. The troupe also runs into financial difficulty when constant rain ruins attendance at their show. Faced with the bankruptcy of the troupe and increasing personal conflict, Kihachi is forced to tell the truth to his son and to confront his own shortcomings.
As Donald Ritchie points out in the liner notes accompanying the set, the Japanese word "ukigusa" means "duckweed" and suggests drifting down river of life. The title represents both the aimlessness of the itinerant actor's life and the transitory nature of existence in general. This theme is also reflected in many concrete images within the film, from the train that opens and closes the film to the seemingly incidental detail of a dragonfly resting on a piece of laundry. At 86 minutes, the film is a marvel of narrative construction: the drama develops in an unforced manner out of the characters' relationships with each other. We never feel, as we do with some directors, that the characters are mouthpieces for particular ideas or that they are supposed to "represent" something. From a dramatic standpoint, Ozu's use of ellipsis is particularly effective; examples include the use of Shinchiki's bicycle to indicate his absence and the offscreen staging of Kichachi slapping Otaka's face. Such indirections encourage us to engage actively with the film, to reflect on the implications of what we are seeing, and to fill in the pieces ourselves. As a result, even the smallest details can take on extraordinary expressive power as an Ozu film unfolds.
The 1959 film
Floating Weeds has long been one of Ozu's most widely seen works in the West. By this time he had already settled into his signature style in its most purified form. Whereas the first film contained several tracking shots, this version has none at all, unless one includes the shot from a moving boat near the beginning of the film. The acting is more restrained and the pictorial beauty of each composition takes an even more prominent role than before. Indeed, the gorgeous color cinematography by the great Kazuo Miyagawa makes this Ozu's most beautifully designed film. At half an hour longer it is also more leisurely paced, though once you settle in to Ozu's rhythm the story becomes quite engaging. One interesting difference between the two films is that because of relaxed censorship standards, the newer film deals more openly with topics such as prostitution and the son's liaison with the young actress. For fans of Japanese cinema, the role of the mistress is played by Machiko Kyo, best known for her roles in Akira Kurosawa's
Rashomon (1951) and Kenji Mizoguchi's
Ugetsu Monogatari (1953); Chishu Ryu, Ozu's favorite actor, plays a small role as the theater manager.
The question of which version to prefer is a matter of which aspects of Ozu's direction you choose to emphasize. David Bordwell, author of an authoritative study on Ozu's films entitled
Ozu and the Poetics of Cinema (1988), argues that the chief value of Ozu's work comes from its "interplay of rigor and playfulness" in terms of film style. (In other words, the way Ozu establishes norms through his strict formal system then playfully undercuts them.) In that respect Bordwell prefers the 1959 version, which contains more of this characteristic interplay; the style of 1934 version is supposedly weighed down by the film's rural subject matter and its evocation of traditional Japanese values. The limitation of such an approach is that it neglects the emotional impact of storytelling and acting, which is, after all, the primary reason why most people go to see movies. In that respect, a major liability of the 1959 film is the stiff performance of Hiroshi Kawaguchi as the son. While most of the film works well in spite of him, the climactic confrontation between father and son simply does not have the resonance of the earlier version. Moreover, while Ganjiro Nakamura is fine as the father in the 1959 version, his performance lacks the charm and distinctiveness of Takeshi Sakamoto. And the young boy in the 1959 version is hardly a match for Tokkan Kozo, whose antics provide the 1934 version with marvelous comic relief to balance out the somber aspects of the story. The bed-wetting jokes, for instance, would seem overly cute if not for Tokkan Kozo's perfectly timed facial expressions. Finally, I also prefer the 1934 version's more compact narrative structure. Together with
I Was Born, But... (1932) and
Passing Fancy (1933),
A Story of Floating Weeds represents the peak of Ozu's achievement in the silent era and deserves to be counted among his best works. The 1959
Floating Weeds is still an exquisite film, but it doesn't have quite the emotional immediacy of Ozu's best silents or, for that matter, mature masterpieces such as
Late Spring (1947) and
Tokyo Story.
The transfer for the 1934 version looks exceptionally beautiful for a silent Japanese film. The film elements are not without damage - one scene wobbles for an extended period of time and one night scene has persistent scratches for a few minutes - but the image is sharp and has a beautiful range of gray tones. This is one of the better-looking silent films I've seen on DVD. The 1959 version has had several incarnations on video, from grimy transfers off faded 16mm prints to a very nice Criterion laserdisc. The new high-definition transfer easily tops them all, with its rich color and sharply delineated detail. This time Ozu's beloved red hues (it seems practically every shot has some kind of red object somewhere in the composition!) really pop out on the screen.
Disc 1, which contains the 1934 film, has an audio commentary track by Donald Richie, a leading English-language expert on Japanese cinema. A lifelong resident of Tokyo, Richie met Ozu and even observed one of his shoots; he provides invaluable insights on Ozu as a director and on Japanese culture in general. Richie also contributed the liner notes and the new subtitle translations. Donald Sosin's piano score, the style of which was intended to recall Schumann (one of Ozu's favorite composers), provides a delicate counterpoint to the emotions onscreen. The performance is noteworthy as one that was created digitally; Sosin generated a MIDI sequence via a keyboard and sent the file to Japan, where it was "interpreted" on a Yamaha Disklavier piano. If the result may not quite have the brio of a live performance, it's musically effective and I doubt that the casual listener would even notice that no human hand ever touched the piano keys! Disc 2, which contains the 1959 film, has an audio commentary track by Roger Ebert. While Ebert cannot pretend to be an expert in Japanese cinema like Richie, he does have a sharp eye for compositional and structural details, though some of his points are overly repetitious. Still, his commentary would be useful for anyone who wants to learn more about basic film analysis.
On the whole, Criterion has done an outstanding job with this set. The films themselves, with their many visual subtleties and affecting characterizations, reward multiple viewings. This is a highly recommended purchase, even for those who are not yet familiar with Ozu's work. The next release in Criterion's Ozu series is
Early Summer (1951); let's hope they release more of Ozu's unjustly neglected silent films such as
I Was Born, But...,
Passing Fancy and
An Inn in Tokyo (1935).
For more information about
Floating Weeds, visit
Criterion Collection. To order
Floating Weeds, go to
TCM Shopping.
by James Steffen
