A Lacanian Analysis
A video version of this article was also made with the permission of Yu Ban, the author of the original answer. Video editing: Chai Lairen (柴来人); collaboration: “Keep Your Hands Off Lab on Roof!” (别对屋顶现视研出手！).
To anticipate my conclusion, and to provoke those who call for a “return to the text,” I begin by asserting that an analysis of Look Back cannot confine itself to the manga alone. We must regard it as an attempt to inaugurate the “imagination of the 2020s” and subject it to a deeper historical-materialist critique. For such a critique—particularly its critique of the ideology at work in the plot’s allusion to the Kyoto Animation arson attack—Lacanian psychoanalytic theory is not merely applicable but necessary.1
As a further provocation, I will set the conclusion aside for now and argue that this criticism, which to some extent departs from and goes beyond the text, must begin with an analysis focused on the text—even on its technical workings, even on somewhat trivial details. This first brings us to a fundamental question, almost an ontological one: what defines manga as an independent medium, distinguishing it in particular from visual media such as animation and cinema?
I believe Tatsuki Fujimoto has answered this question fully in Look Back: what establishes manga’s independence as a medium is its distinctive techniques for representing movement. In other words, although cinema, animation, and manga are all considered visual arts capable of reproducing movement, manga has evidently developed techniques for representing movement that differ from those of the other two. It has, we might say, assembled a distinctive “manga machine.” An analysis of Look Back also allows us to give a full account of what distinguishes these techniques.
Our analysis begins with the two pages widely recognized as Look Back’s most distinctive and accomplished. Clearly, the primary artistic intention of the two pages shown below is to represent Fujino’s increasingly exaggerated bodily movements as she walks home after first meeting Kyomoto. The first problem the manga must solve is therefore this: cinema and animation—more precisely, full animation—rapidly replace one image with another over time, using persistence of vision to impart a perception of movement to the viewer. How, by contrast, can manga, a synchronic pictorial art, produce a perception of movement through one or more panels laid out across a page? The “manga machine” offers a first solution: deliberately guiding the movement of the viewer’s focus as reading unfolds in time. This solution is concentrated in the first page shown here. Three similarly structured panels are evenly distributed down the page, each composed of a figure in profile in the foreground against a different background. The difference is that, from top to bottom—the default order of sequential reading—the figure’s position within the composition moves from the right to the center, then continues to the left. As viewers read the page in sequence, their visual focus also moves from right to left. Thus ① compositional differences among panels ② guide a sequence of eye movements, which ③ the reader’s imagination synthesizes into a perception of Fujino moving laterally across the image. An illusion of movement has successfully been produced on a static pictorial plane.
In the second image, a single panel spanning two pages occupies the entire pictorial space. Viewers can no longer gain a perception of movement by directing their attention to different panels in succession. Here, however, the “manga machine” employs a different kind of technique. To understand it, we must focus entirely on the figure’s unusually contorted posture. Fujino raises both arms, twists her upper body sideways, and leans back, with one foot already off the ground. Her center of gravity has clearly shifted away from the center of her body. Yet the bent knee of the leg still planted on the ground conveys a full sense of weight: Fujino has not become weightless; she is not floating in some surreal fashion. Through imaginative investment in this contorted yet weight-bearing body, viewers clearly recognize an unbalanced, and therefore unstable, posture. Driven by a young girl’s vitality, Fujino’s body is rapidly passing from one stable bodily configuration to another. What the manga machine does is capture the contorted body’s unstable posture in a panel at the fleeting instant of transition between these two stable configurations. Of course, all of this is merely a misrecognition produced by manga’s techniques for representing movement. What actually exists is only a static double-page panel and the unstable posture it depicts. This technique, different from the one used on the preceding page, also succeeds in producing a different perception of movement. (A close reading shows that the two techniques are not, in fact, neatly separated across the two pages. They overlap, with a different emphasis on each: the figure’s unbalanced posture is already anticipated in the first page’s third panel, while in the second image she remains at the edge of the composition, suggesting that her lateral movement has not ended.) If compositional differences between successive panels represent the figure’s lateral movement, the unbalanced posture within a single panel represents the circular movement of her limbs relative to her torso. We can thus understand how manga achieves its distinctive representation of movement by breaking movement down and redistributing it across the plane of the panels in various ways. Unlike cinema’s characteristic ballistic vision or animation’s characteristic layer compositing, the techniques of movement that define manga’s distinctiveness tend toward an entirely planar design.
Once we notice how Fujimoto uses unbalanced postures to produce a perception of movement, we can also see how frequently contorted bodies appear in his manga. This image gradually exceeds its original technical purpose of producing a perception of movement and is put to more distinctly stylistic uses. In some cases, unbalanced bodily postures appear in wholly static panels. Take the two panels shown below: Fujino sits still at her desk, bent over her drawing. Particularly in the first panel, we can see her left shoulder hunched up, her right shoulder seemingly straining forward, and her hips slightly off the torso’s central axis, twisted to the left. The posture is exaggerated and awkward. To understand what the contorted body does in a static scene, we need to draw on @甚谁’s correct observation. Both panels aim for the most precise perspective possible, giving the room a degree of pictorial depth. Within this virtual spatial depth, the figure is consistently placed at the far end of the room, while near the viewer’s perspectival vantage point Fujimoto draws objects—a schoolbag or a chair—that attract the eye. The composition thus guides readers to focus first on the objects in the foreground, then gradually to move along the axis of perspective toward the figure deeper within the space. It invites them to explore the depth produced by perspective and to move their attention through it. Combined with the figure’s unbalanced posture, this visual movement creates the illusion that Fujino is not drawing at her desk but struggling to advance into the depth of the depicted space. In the context of the story, of course, viewers understand that this illusion is not intended to represent movement or evoke its perception. Rather, it refers to movement at the level of meaning, producing a metaphorical effect that allows viewers to feel how difficult it is for Fujino to advance in the manga industry. Here, the perception of movement gives way to the idea of movement. The technical use of contorted bodily postures shifts from reproducing concrete physical movement to producing textual meaning. In the process, movement is not only virtualized and conceptualized; it also detaches itself from its own spatiotemporal context and acquires complete autonomy. Alongside this development, the image of the “contorted body” begins to detach itself from its original purpose of representing movement and gradually acquires a technical autonomy of its own.
Yet this analysis still does not exhaust Fujimoto’s uses of the contorted body. In the next panel, we discover another kind of contorted bodily image. The viewer is placed in the position of a cinema screen, facing an audience sprawled every which way in their seats. Unlike the examples above, in which Fujimoto gives contorted bodies only to the protagonists, here the three incidental figures surrounding the two protagonists in the central area of the image also slump in their seats, each in a different posture. Viewers must therefore confront the proliferation and dispersal of contorted bodies across the image. Beginning with the protagonists at its very center, their attention is not concentrated in any particular direction but spirals toward the edges, following the distinctive incidental figures. This cancels the possibility of producing a perception, or even an idea, of movement. We can no longer assign the contorted bodies in this panel to any intention connected with movement. We must instead acknowledge that here the contorted bodily image is its own technical end: it finally breaks with movement altogether, becoming in itself an object of the viewer’s visual fascination and a signature of the artist’s style. Its appearance in a cinema scene is highly ironic. On the one hand, it faithfully depicts the actual postures of cinema audiences: captured by the visual attraction of the images, these people seem to have lost the ability to move and coordinate their bodies. They slump in their seats, wearing dull, ridiculous expressions. On the other hand, in Fujimoto’s stylized aesthetic practice, spectators’ vision is turned back upon spectators themselves. Now it is their own contorted bodies that are captured in the panel and presented to a second-order spectator. Spectators become the objects of the very desire to look that animates them. In this use of the technique, both the representation of movement and narrative function tend toward extinction, and manga develops a powerful impulse to return to static pictorial art. The appearance of an unbalanced bodily posture no longer presupposes, beyond itself, movement in three-dimensional space and flowing time. Instead, it is “sublimated”—in the strict psychoanalytic sense—into a representation of an object for purely libidinal investment. The technical use of contorted bodies and unbalanced postures in Fujimoto’s manga has now developed a third strategy, distinct from producing either a perception or an idea of movement: a new strategy we can call “fascination with the object.”
Once we locate this “fascination with the object” as an extension of techniques for representing movement, we can abstract the strategy from the image of the contorted body and redeploy it elsewhere. Indeed, the extensive proliferation of “fascination with the object” defines an important aspect of Fujimoto’s style. Among Japanese manga artists, Fujimoto is notably distinctive in his treatment of faces. Against manga’s general tendency to flatten the face, he often uses shading to emphasize its three-dimensionality, and produces lively expressions through the movement of facial features relative to one another. He is also fond of depicting the nose, which flattened facial styles often neglect. More remarkably still, in his endlessly repeated low-angle facial close-ups, the central focus of Fujimoto’s characteristic treatment of faces is no longer a particular organ but the absence of an organ—a pair of nostrils. Emphasizing the nose to render facial anatomy more accurately is an uncommon but by no means rare stylistic choice in manga. Such enthusiasm for foregrounding nostrils, however, is quite singular. Strictly speaking, nostrils are not among the facial features, nor do they express emotional changes by moving against the background of the face as those features do. Unlike the face’s other openings, nostrils neither direct a look outward as the eyes do nor are filled with language as the lips are. They simply leave a pair of awkward absences in the otherwise intact surface of the face, passively expanding or contracting as the other organs actively move. In other words, the nostrils Fujimoto favors do not supplement his characters’ faces with livelier emotions or more pronounced traits. On the contrary, they disrupt the pictorial integrity of the facial close-up, upsetting the balance among the features and turning the face’s Gestalt structure into a contorted vortex. It is precisely in Fujimoto’s baffling enthusiasm for drawing nostrils that we rediscover, beyond the limbs, a “fascination with the object”: a fascination with representations of imbalance, imperfection, negation, and even lack. This discovery confirms our earlier argument that such “fascination with the object” cannot be reduced to pure technical functionality—representing movement, for instance—but should be understood as a distinctive expression of stylized aesthetic appreciation (or rather, an appreciation of ugliness).
Most importantly, the “defunctionalization,” “decontextualization,” and extensive proliferation of “fascination with the object” in Fujimoto’s manga encourage us to go further, looking for the “contorted object” in the content of the plot rather than the form of the images. Interestingly, there is really no difficulty in doing so. As @tearsintherain’s answer notes, Fujimoto’s manga refuse to depict a rosy utopian social setting or provide their characters with happy endings. Instead, they attend more to life’s low points and the corners hidden in society’s shadows. What truly matters, however, is not simply this attention to an intangible “dark side.” It is that such attention always finds concrete expression through a “dark object” introduced into the plot—the Darkness Devil in Chainsaw Man, for example, or the maniac who murders indiscriminately in Look Back, an allusion to Shinji Aoba. This establishes the decisive difference between Fujimoto’s manga and Bong Joon-ho’s film Memories of Murder. Bong shows considerable restraint in his depiction of serial murder: not only does the ending withhold the killer’s actual identity, but the scenes involving the murders also avoid showing him directly as far as possible. This restraint shifts the narrative’s center of gravity from solving the serial murders to the changes in the characters’ states of mind and relationships during the investigation. It produces a folding effect: the “dark side” is drawn inward from the external crime into the characters’ psychological lives.
Some film critics, following the established habits of South Korean social-issue and crime cinema, seek to uncover allusive connections between Memories of Murder and its social background. They interpret the closing close-up of Song Kang-ho’s character, his face bewildered and lost, as an indictment of South Korea’s social problems. Such attempts at a “social” reading are undoubtedly on the right track. The problem is that they fail to attend to how the plot’s references to social problems combine with the investigation’s inward folding into the psychological realm. It is precisely through this folding that Memories of Murder rightly refuses direct reference to pseudo-concrete problems such as grassroots social unrest and judicial corruption in the dark years of South Korea’s dictatorship in the 1980s. Instead, the plot’s inward psychological folding produces a “supra-concrete” social allegory. This allegory rejects the view that immediately visible social appearances are the only site in which social problems arise, insisting instead on expressing a rejection of social reality at the level of its essence. It thereby redirects its attack from the dictatorship directly referenced by the plot toward the beginning of the new century, when the film was made. In this era, when democratization was proclaimed complete, social problems did not end with the fall of dictatorship. Instead, they were exposed at a more fundamental, more basic level. Yet in the twenty-first century, dominated by liberal-democratic discourse, we have lost the vocabulary needed to articulate these fundamental problems adequately. We can refer to them only indirectly, through psychological allegory. The restrained, circuitous mode of expression adopted by Memories of Murder not only correctly represents the fundamental problems of South Korean society in the 2000s. Through a more intricate mechanism, it also expresses the impossibility of representing those fundamental social problems—and this impossibility itself constitutes a crucial part of those social problems.
Let us now return to Fujimoto’s manga. Compared with Bong Joon-ho’s cool restraint in Memories of Murder, Fujimoto is more inclined to give the so-called “dark side” a visible image without circumlocution. At the level of plot, these images constitute his distinctive “fascination with the object.” The concrete presence of these characters distorts the course of the plot and produces heartbreaking regrets within the story, but it also opens up a space in which what has gone wrong might be put right or transcended. To return to Look Back: if Fujino and Kyomoto simply parted ways, the ensuing story could only be utterly ordinary and dull. Fujino, who in fact harbors doubts about the role of the professional manga artist, might also find it difficult to carry on for long. But the appearance of the murderer as a “dark object” reverses the story’s downward drift. Although his appearance irrevocably deprives both the reader and Fujino of Kyomoto, through a poetic coincidence it allows Fujino to rediscover Kyomoto’s quiet support for her career as a manga artist. When Fujino retreats into an imagined world, remorseful at having indirectly caused Kyomoto’s death (needless to say, this remorse is in one respect highly questionable: after all, what Fujino imagines also suggests that Kyomoto’s encounter with the murderer is “fated” and has nothing to do with Fujino), she can save Kyomoto’s life by subduing the murderer. And when, subject to the “reality principle,” she leaves that imagined world, Fujino can also accept Kyomoto’s wordless support and thereby complete a ritual of mourning, cast aside her doubts, and identify once more with her work as a professional manga artist. Ultimately, the image of the murderer brings about Fujino’s reconciliation with reality. In this sense, the “fascination with the object” that Fujimoto arranges within the manga’s plot serves as what Lacan calls a “quilting point.” This imaginary object, by no means lacking in concreteness, secures the subject’s identification with a fragmented social reality rife with contradictions. As our analysis of Memories of Murder has shown, this direct “fascination with the object,” which dispenses with elaborate technical indirection, offers only a “pseudo-concreteness” that leads to no genuine way out. What it ultimately produces is an imaginary solution to society’s “dark side.” Worse still, it may arouse a psychotic identification with this aggressive “dark side” rather than a rejection of it. In this sense, Fujimoto’s work should be situated as an answer to the question of the “imagination of the 2020s”—except that “imagination” here must be understood in the strict sense of Lacanian psychoanalytic theory, as the Imaginary. For Lacan, the Imaginary is the domain of narcissism, false identification, and self-enclosure. It offers the subject simple, clear schemata that satisfy basic needs, helping the subject evade contradictions within itself and the external world at the level of the Real.
Finally, through this whole complex, entangled analysis, moving from “materialism” at the purely technical level to “idealism” at the level of plot content, we have at last reached the ultimate, untranscendable horizon of historical materialism. Within this horizon, we should reject, in a measured way, Look Back’s profoundly moving allusion commemorating the victims of the Kyoto Animation arson attack. We should make that allusion itself a problem for criticism, and thereby predict how the pseudo-concrete images of the arson attack and its perpetrator will capture the imagination of Japanese ACG—anime, comics, and games—over the next decade. We must also recognize clearly that the attraction of these images is an imaginary compensation for the despair and despondency of late capitalism. In these 2020s of stagnation and decline, the optimistic enthusiasm for global markets and a pluralistic society that marked the end of the twentieth century has been replaced by the image of an aggressive other emerging in an atonal world. It is therefore absolutely necessary to examine dispassionately the historical limits of the “fascination with the object” embedded in Fujimoto’s manga. Only on this basis can we ultimately transcend a reality in precipitous decline.