and fiction films (Tsumura 1940: 132). “But in the world of art,” Tsumura contends, “there is nothing more dangerous than seeing the world by applying an ideology called ‘materialist dialectic’ [yuibutsu benshōhō] … Needless to say, I do not trust the success of the application of such a man‐made calculator in the world of art, so I can never agree with Mr. Rotha” (Tsumura 1940: 143). Seen from Tsumura's blatantly anti‐Marxist perspective, Rotha's term dramatization means a shallow interpretation of actuality through its economic basis alone, and for that reason, Tsumura refuses to see any positive value in Rotha's documentary theory by concluding that “the world of documentary is nothing but the materialist world that has nothing to do with the human spirit” (emphasis in the original) (Tsumura 1940: 133). As his final words, Tsumura warns the reader that he is not totally dismissive of the potential of nonfiction films in general, but never forgets to stress that its healthy development must lie in the producer's proper understanding of its stylistic, thematic, and ontological difference from fiction film (Tsumura 1940: 149).
Cinema of Facts: Imamura Taihei
It should be remembered that Tsumura published his counterargument when Rotha's prestige in wartime Japan was at its zenith; it thus provoked significant criticism from his fellow critics. This criticism, however uncompromising, focused mostly on Tsumura's misunderstanding – or rather, intentional dismissal – of Rotha's basic terms and political standpoint, therefore leaving behind his philosophical questioning of the categorical difference between reality/Realität and actuality/Wirklichkeit (Takakiba 1940: 525–528). This theoretically unfruitful situation did not change much until after World War II, as was also the case for the Marxist film critic Imamura Taihei, who commented only sporadically on Rotha before 1945. During the war, Imamura had already earned a reputation as “the best film scholar of our time who has most beautifully proved the theoretical character [rironteki seikaku] that we could ascribe to the Japanese” (Ōkuma 1938: 411) for his very prolific and systematic writings on cinema. Indeed, Imamura published 10 monographs between 1938 and 1943, including The Form of Film Art (Eiga geijutsu no keishiki) (Imamura 1938), Theory of Documentary Film (Kiroku eigaron, 1940), and Theory of Animated Film (Manga eigaron) (Imamura 1941), to name but a few. Today, Imamura is recognized as the founder of Japanese animation theory (Lamarre 2014), but the focus of his theoretical writings always revolved around the camera's ability to offer an immediate and objective “document” of events or phenomena that took place before its lens, or around what Michael Renov has once called “a direct, ontological claim to the ‘real.’ ” (Renov 1986: 71). It thus seems natural that Imamura, unlike Tsumura, found common ground with Rotha – there is a portrait of him reading Documentary Film (see Figure 3.1) – and detailed his favorable interpretation of the book in the last chapter of his 1952 Introduction to Film Theory.
Figure 3.1 Imamura Taihei.
Imamura begins his chapter by pointing out his fellow Japanese critics' conventional misunderstandings of Rotha's documentary theory. While these local interpretations frame Rotha as a kind of “formalist” who rejected fiction film tout court, Imamura rightly reminds the reader that what Rotha criticized was not fiction film as a whole but the capitalist basis of the film industry that had privileged this genre as the most important form of film practice. Similarly, Imamura speaks highly of Rotha's promotion of documentary as the most effective tool for mass propaganda, writing that Rotha's documentary theory “reflects [his] strong social consciousness … and is based on the contemporary mass public's most urgent/real demand,” and that it thus should be respected above all for its unflinching aim to “enlighten the people politically, to turn their eyes to fundamental contradictions in the modern social system” (Imamura 1952: 153–154). Highly sympathetic to Rotha's promotion of documentary's mission of social reform, Imamura goes even so far as to declare that “under Rotha's opinions lies the idea of socialism, and in this sense his ‘documentarism’ shares commonalities with that of the Soviet Union” (Imamura 1952: 154). As a Marxist writing before Nikita Khruschev's critique of Stalin's cult of personality, Imamura uses a comparison with the legacies of Soviet Union's revolutionary model of film practice in the 1920s – which had huge impact on the Japanese left in prewar Japan despite severe state censorship – as the highest compliment for Rotha's political consciousness.7
Despite these general accolades, Imamura is highly critical about Rotha's own theorization of documentary. For instance, he condemns Rotha's inclusion of Potemkin and other examples from the Soviet Union within the category of documentary, arguing that these films should be treated as what he calls “semi‐documentary.” Although they use some basic methods borrowed from documentary, films that present us with a reenactment or reproduction of historical events, he argues, must remain in the realm of fiction because “they did not record the facts [jijitsu] themselves” that took place in front of the camera (Imamura 1952: 160–161). Moreover, after concisely paraphrasing Rotha's idea of dramatization – “rather than showing the fact as it is, one should change it at one's discretion by filtering it through our subjectivity so that certain principles behind it are revealed” – he abruptly dismisses it by saying “this statement is so illogical that it is hard to grasp its meaning” (Imamura 1952: 166). Likewise, Imamura's unsympathetic judgment is applied to Rotha's application of dialectics. “It seems apparent,” he writes, “that this [Rotha's reference to dialectics] is nothing more than a mere repetition of Eisenstein's dialectic of collision, and in this instance, too, the term ‘dialectic’ serves as a magic spell. In depicting anything, this position deems it to be enough to see the collision of things, and, as a result of this collision, lead to the dialectical formula composed of thesis, antithesis, and synthesis. This is the most foolish part of Rotha's Documentary Theory” (Imamura 1952: 176).8 Imamura concludes his reading with a highly self‐contradictory remark: “As I have shown, Rotha's Documentary Theory contains a lot of inaccurate terms used in an idiosyncratic manner, but if we try to understand what he meant to say as a whole, then we come to realize that it has very correct views” (Imamura 1952: 167).
Imamura's schizophrenic interpretation of Rotha's text was inseparable from his own determination to fight against domestic opponents of documentary film practice. In other words, he deployed Rotha as a mouthpiece through which his own theory of documentary could be disseminated. What are, then, main characteristics of his theory, and how do they differ from Rotha's? First, in contrast to Tsumura's discussion of the philosophical difference between actuality and reality, Imamura introduces a third term “fact” (jijitsu) as the foundation for what he considered to be kiroku eiga (Imamura uses this category in the same way as Tsumura, meaning nonfictional films in general). To be sure, Imamura does not ignore the creative intervention of the human agent in the capturing of the fact, as he likens the operation of montage to that of the human cognition in his earlier publications such as Theory of Documentary Film (Imamura 1940: 26). But his film theory in general repeatedly emphasizes that the factuality of filmic representation is guaranteed by the mechanical nature of the photographic image, which is able to capture what has been invisible or unknowable to the human perception, to grasp an object's motion as it simultaneously moves before the camera, and to reproduce identical images at all times. In Imamura's view, it is this mechanical nature of the photographic image that distinguishes cinema from the traditional arts, and, just like the late Kracauer, he argues that “one can understand film's property by knowing the features of the photography that constitutes this medium's basic units, structural elements, and historical origin” (Imamura 1957: 97–99). And as long as it makes use of those mechanical/photographic features properly, kiroku eiga – or what he now calls the “cinema of