House Logo

In a sense, Ozu doesn’t help his case when so many of his shots are visually artistic for the sake of being visually artistic. A perfect example in my mind would be his fascination with corridor shots. Corridors make for perfect Ozu landscapes, of course, because they allow him to quite naturally play with visual depth, layering and patterning, which are things that Ozu otherwise had a habit of forcing into his compositions (think of the way he arranges teacups and beer bottles in the foreground of so many of his otherwise bland one-shots). In another director’s work, say Pedro Almodóvar’s, such an intense fascination with corridors would almost have to be symbolic (the corridor as birth canal, or some such thing), but all I think it reveals in An Autumn Afternoon is Ozu’s fondness for those compositions. He ogles corridors the way Quentin Tarantino ogles women’s feet. There’s nothing “wrong” with that, per se, so long as we don’t force a meaning onto those shots that I don’t believe is there to be found.


An Autumn Afternoon

EH: I feel like we’re starting to run aground on the reef of subjectivity with this topic. Partly, that’s because Ozu’s cinema steadfastly refuses the kinds of concrete meanings that would allow for more solid, definitive interpretations. That’s why I’ve tried so hard to avoid suggesting that Ozu’s images are overtly symbolic; there’s no trace of that “corridor as birth canal” type of sensibility that finds symbols everywhere. Generally speaking, Ozu’s corridors are just corridors, and his bar signs are just bar signs, but they still fulfill multiple roles within his films. At the most basic level, they transition from place to place or character to character. They are also visual signals for recurring locations, which is why Ozu sometimes walks through similar sequences of shots several times to create an association between a location, its surroundings and the emotional undercurrents of that place. And in the tensions Ozu sets up between the different elements that he places so carefully within the frame, the tensions of the narrative are often worked out in miniature. When Ozu positions a small black-and-white TV set amidst the clutter of traditional tea cups and sake pitchers, it’s yet another small suggestion of modernity and traditionalism coexisting, though it’s also (and perhaps more obviously) simply a way of setting the scene at a small bar where businessmen are watching a baseball game after work.

In the past, I’ve compared Ozu’s exteriors to the Japanese artist Hokusai’s famed Thirty-six views of Mount Fuji, a set of woodblock prints depicting the mountain from various angles and in different lightings and seasons. Ozu’s methodical examination of exterior locales—a cluster of shots of baseball stadium lights, a progression down a street lined with bars, a rubble-strewn and rundown area with apartments overlooking its disarray—seems like a modern expression of the same impulse to limn a setting’s visual and poetic possibilities. That, and not the need to pin a single meaning or intent on each frame of the film, is what I’m talking about when I say that Ozu’s exteriors are “evocative” or “suggestive.”


An Autumn Afternoon

JB: Fair enough. Again, I agree that Ozu’s shots are evocative in the collective. That said, I think this was a worthwhile discussion to have, because I think we’ve just demonstrated how easy it is to exaggerate the evocativeness or symbolism of a specific shot in Ozu’s oeuvre, either by reading more into the composition than is actually there to be found or by not specifically tying a shot’s evocativeness to other shots that are crucial to giving that single shot meaning.

In my mind, David Bordwell makes at least one of those mistakes in his audio commentary for the Criterion edition of An Autumn Afternoon. As the author of the book Ozu and the Poetics of Cinema (which, for the record, I haven’t read), Bordwell might be the foremost expert on Ozu. So of course his commentary is packed with sharp analysis and interesting historical anecdotes. But as I listened to his commentary, I came away with the feeling that Bordwell knows Ozu and this film almost too well, because at this point he seems to see An Autumn Afternoon through Ozu’s eyes and with Ozu’s sensibility: Ozu was obsessed with color, particularly red, and so Bordwell fittingly calls attention to the presence of color in many shots. Ozu was also obsessed with the general compositions of his shots and with the duration of each shot, and so Bordwell calls attention to that, too. It’s all very educational. But it’s also a little like an algebra equation. Bordwell frequently combines Ozu’s cinematic instinct with Ozu’s cinematic execution of his intent and comes to a result that, in my opinion, is greater than the sum of those parts. In other words, Bordwell doesn’t solve for x.

For example: In a scene where Hirayama is drinking with his friends, Bordwell notes how a small flower in the background “provides a nice burst of red,” and then he refers to a pair of mirroring shots of two men drinking as a “visual flourish.” Later, he raves about Ozu’s (carefully preplanned) editing in a scene that uses multiple angles to show Koichi lying on his back and then standing up, commenting that Ozu “daringly cuts on [Koichi’s] movement but flips [our view] 180 degrees to the opposite side”—just one of one of the many times he praises Ozu’s “action cuts.” I have no objections to Bordwell’s specific observations, but what’s frequently missing is any convincing description of how or why these elements of Ozu’s filmmaking serve the drama of the story. I suspect Bordwell could make those arguments convincingly, he just doesn’t do that in his audio commentary, which seems to assume that because Ozu was a skilled perfectionist who “obsessively” composed each frame, that his intent-laden compositions must succeed in enhancing the film’s emotional depth and complexity—because otherwise we’d have to call many of Ozu’s shots visually resplendent and nothing more.

Perhaps the best illustration of what I’m talking about comes in Bordwell’s description of the acting in this film. When I watched An Autumn Afternoon, I scribbled down the words “animatronic” and “confined.” In this film, and in the other Ozu films I’ve seen, the actors often seem bolted in place, as rigid as something out of a Star Wars prequel. (There are extreme exceptions to this of course, one of them being the comical scene at the bar in which Hirayama’s military buddy, Sakamoto (Daisuke Kato), performs a drunken march.) Prior to listening to Bordwell’s audio commentary, I always assumed that the rigidness of the acting was indicative of the time and place in which it was made—Japan in the early 1960s—but then Bordwell’s commentary track informed me that Ozu provided “hyper-exact direction” to his actors, going so far as to regularly demand specific expressions and movements, which were often timed on a stopwatch to ensure they complied with Ozu’s planned pacing. Once I learned that, it all made sense: The acting is often rigid because the actors are constantly trying to hit their marks, even when they’re sitting down. This seems a fairly obvious and almost inarguable conclusion, even if subjectively we might disagree on how often the acting is “rigid.” Yet somehow Bordwell concludes that the acting of this film is impressively “not robotic.” To that point, I’ll give Ozu (and Bordwell) this much: Considering how much Ozu micromanaged each shot, it’s amazing that the acting is as natural as it is. But here’s a case in which I feel like Ozu is being celebrated for a meticulousness that in fact might have been detrimental to his film, whereas in the cases above, when Bordwell compliments Ozu’s “action cuts,” I feel as if Ozu is being complimented for a technical skill, without instead considering the degree to which that skill affects (if at all) the emotions of his film.

I respect that Bordwell, like many film buffs (Jim Emerson comes to mind), thrills to technical execution almost as much as to what that technical execution achieves. I’m not trying to disqualify that reaction. (If it thrills, it thrills.) But if that’s the case, Ozu should be praised explicitly (and justifiably) for his craftsmanship. I think sometimes knowing or suspecting Ozu’s intent can be an obstacle to seeing what the filmmaking actually achieves.


An Autumn Afternoon

EH: I’m totally with Bordwell on the acting in this film. The performances never strike me as the least bit “robotic” or “rigid,” except in the sense that the actors are portraying people who purposefully (try to) disguise their emotions behind pleasantries and formalities. This type of restrained performance is essential to Ozu’s aesthetic, and to the points he’s making about Japanese society. This approach to performance is as much a part of Ozu’s carefully constructed milieu as his static camera placement, finicky mise en scène and avoidance of dissolves and fades. The characters in Ozu’s films don’t generally express themselves directly, seldom diverting from meaningless social chit-chat. This is something that Western audiences understand intuitively when Ozu is depicting the relationships between businessmen, but it can be disconcerting to realize that the private, familial relationships in Ozu’s films often seem nearly as formal, nearly as bound by rules of decorum and politeness.

Furthermore, I’d argue that An Autumn Afternoon, more than most other Ozu films, depicts the cracks in this system of emotional restraint. The relationship between Koichi and Akiko is held up in many ways as a more “modern” template than the traditional relationship where feelings are suppressed and conversation is limited to trivialities. Akiko in particular is a more liberated woman than many of the female characters in Ozu’s oeuvre; she banters with her husband, negotiating to get what she wants. Ozu partly plays this off as a joke on Koichi, who seems weak and hesitant as a consequence of his wife’s forcefulness, but I think Ozu also gets a kick out of Akiko’s sharp wit. There’s also the scene in which Koichi is shown preparing dinner for the couple, which seems like a small thing except in the context of Ozu’s Japan, where women are often attentive servants to the men, who don’t contribute to the household chores at all. This film, more than any of Ozu’s others, shows a Japan in flux between traditional values and a new modern era, and the shifting dynamic between men and women—as represented in the dialectic between the shy, hesitant Michiko and the outspoken Akiko—is one site of this transformation.

As usual with Ozu, one has to be attentive to nuances to pick up on the emotional stakes of these performances: Akiko’s sly smirk and mock surprise when she’s taunting her husband; Michiko’s shyly smiling complicity with her sister-in-law’s maneuvers; Koichi’s petulant, boyish pouting when he doesn’t get what he wants. It’s frankly surprising to me that you’d denounce these performances as stiff when, to me, beneath their surface reticence, these performances are practically bursting with subtle gestural cues and slight shifts of expression that signal the churning emotions or comical subtext of these scenes. This submerged expressionism is evident in dramatic scenes as well: I already mentioned the subtle, meaningful glances during Michiko’s seemingly banal conversation with Miura, and you brought up the mysterious but nonetheless affecting emotions running through the wedding preparations. It’s equally evident in seemingly minor sequences, like the way Hirayama interacts with a secretary who’s leaving to get married, and his tenderness with her reflects his confused feelings about his own daughter of the same marriage-ready age. Far from feeling that Ozu’s meticulous, rigid scene construction unnecessarily restricts the actors, I think Ozu is trying to convey a societal structure that is restricting the characters—who, nonetheless, betray the true depths of their feelings in small ways despite all the pressures on them to hold back.


An Autumn Afternoon

JB: I think you’ve made strong points and identified scenes in which emotion slips through the cracks of the film’s rigid structure. And I agree that Ozu is trying to say something about the relationships between the characters, and indeed about the evolution of social behavior in Japan, by contrasting the more uptight manner of Hirayama and his friends with the considerably more relaxed manner (in some scenes) of his son Koichi and people from that younger generation. So it’s here that I backtrack to admit that in the same rant in which I expressed frustration with Bordwell’s lack of specificity, I made the same mistake. Because in saying that the acting in this film is robotic, I wasn’t trying to imply that An Autumn Afternoon lacks characters with character, if you will. Instead, I was trying to make the point that even the most boisterous acting—Akiko yelling at Koichi, for example—is made to fit into such a specific space that it’s as if the characters are moving on rails, as in a Disneyland ride.

That’s why I identified Sakamoto’s drunken march as an exception, because it’s one of the few times when the actors don’t seem bolted down, pivoting on one specific point. Does it really seem in character that a modern wife like Akiko would lecture her husband while leaving her lower body perfectly still, as if standing at attention? To me it seems forced—forced not by a desire to dig into themes but to ensure that Ozu’s precious compositions are preserved. (If Akiko moved a foot to the left or the right, it would throw Ozu’s shot out of balance.) Other than Sakamato’s march and the scene in which Koichi hits golf balls on the roof of his office building, there aren’t many moments in this film in which the characters are allowed to move of their own free will, which perhaps explains why Ozu has so many ancillary shots of people walking up and down corridors and in and out of doors, as if to make up for the, yes, robotic visual plotting of the other scenes.

In his commentary, Bordwell says that one of Ozu’s longtime assistants thought it was unusual, and un-Japanese, for a filmmaker to be so fixated on a specific visual approach that he would essentially force each scene into that mold. So what I’m trying to get at is the idea that Ozu’s meticulousness sometimes led him to construct the action in a way that confined his actors, that made their movements seems programmed, not for dramatic reasons but for aesthetic ones.

Take, for example, the scene in which Koichi and Miura have drinks late in the film: Ozu shoots most of the scene in his distinctive straight-ahead singles in which the speaking actor looks directly into the camera, or sometimes just above it, as if looking into the eyes of the person he’s talking to. Between Koichi and Miura are various items on a table, including a bottle of soy sauce. To the right of each man, as established in an initial two-shot, is an oversized beer bottle. The way Ozu films the scene, the soy sauce bottle jumps from one side of the table to the other in each alternating shot, to reflect each man’s perspective (the soy sauce bottle is slightly to Koichi’s left, and thus it’s slightly to Miura’s right). The beer bottles, on the other hand, only appear one at a time in each alternating shot, and always to the character’s right, which means always on the left edge of the frame.

Bordwell seems to suggest that this is a continuity error, but it isn’t; the shot structure implies that each character is looking over his own bottle (and glass) of beer to see his friend’s bottle, whereas the soy sauce bottle appears in each shot—and jumps from one half of the screen to the other—because it’s in the middle of the table. Still, this scene exemplifies Ozu’s obsession with visual patterning and, in my opinion, demonstrates how he painted his actors into corners. The actors playing Koichi and Miura have no choice but to sit in the same erect posture, not because it’s “in character” (in the scene on the rooftop, they’re much more relaxed) but because that’s what Ozu’s visual aesthetic requires. Bordwell theorizes that perhaps Ozu is trying to suggest the men are mirrors of one another, but he doesn’t seem convinced, noting in the end that Ozu probably shot the scene this way because it’s “cool.” And that’s fine by me. As I said before, if it thrills, it thrills. But whereas you feel that Ozu created this kind of patterning to convey a social structure that restricts these characters, I don’t buy it. Or, more specifically, I don’t think that was the main impetus for the rigid structure of Ozu’s compositions, which I feel often overcame the emotions and story he was supposedly trying to evoke and explore. I don’t expect to convince you of this point. I merely mean to explain why I feel as if Ozu’s compositions don’t always translate into emotion, mood, theme or social commentary. Sometimes I think they reveal the director’s passion for composition itself.


An Autumn Afternoon

EH: That makes sense to me. It seems like, rather than thinking the performances themselves are rigid or robotic, it’s the blocking of the actors and their carefully managed arrangement within the frame that strikes you as restrictive. I don’t see it as a problem—for one thing, there’s no reason it can’t be both an aesthetic/pictorial choice and a reflection of Ozu’s themes of societal rigidity—but I understand what you’re seeing. To some extent, we’re seeing the same thing—Ozu denying his actors much mobility within the frame—and disagreeing about its effects. You say that “there aren’t many moments in this film in which the characters are allowed to move of their own free will,” and that’s a very telling description. The central theme of the film is precisely that lack of control over one’s life. Hirayama is caught between two bad choices, the loneliness of a solitary old age versus the misery of forcing his daughter to sacrifice her own independent life for him, likely leading towards a future very similar to the Gourd’s unhappy situation. If these characters often seem like they’re on rails, unable to move freely, that’s because they are; they are tied to seemingly unavoidable destinies mandated by culture and tradition. Hirayama looks to the past so much, with his school reunions and reminiscences of his military career, because he has so little to look forward to in the future. Even Michiko, who at times shows signs of modernity and independence, winds up with little role in her own fate, acquiescing to her father’s choice of husband when her own choice falls through. Ozu purposefully never shows her with her new husband to emphasize how little it matters, to anyone, who she marries, as long as she marries someone to fulfill the role that’s expected of her.




1 2 3