The History of Mushi Pro – 01 – The Road to TV Anime (1960-1965)

This is part 1 in the History of Mushi Production series. You can read the introduction here

Osamu Tezuka, the God of Manga, is also the father of modern anime – not only did he coin the word, his Tetsuwan Atom was the first animated TV serial in Japan, and pioneered the “limited animation” techniques still associated with anime. Or so the story goes. Actually, things are far more nuanced: the goal of this first article is to show that, and narrate the events that led to the completion of Atom in all their complexity. I hope to achieve that in mostly two ways: avoiding teleology – that is, the idea that Tezuka’s goal was to make TV, “limited” animation from the start – and shifting the focus away from Tezuka as an individual.

For this reason, I will skip the obligatory accounts of Tezuka’s fascination with animation from an early age, the parallels between the so-called dynamism of his manga and animation, and the question of whether or not the reason Tezuka did manga in the first place was to earn money to create an animation studio. On that latter topic, I will just mention that the one who examined it the most thoroughly among the sources I consulted, Nobuyuki Tsugata, concludes in favor of that possibility [1]. With this out of the way, I want to move on immediately to Tezuka’s first direct involvement in animation production: that is, Tôei Dôga’s 1960 film Saiyûki.

Saiyûki and the first Atom proposal

Created in 1956, Tôei Dôga was the subsidiary of live-action films production and distribution company Tôei, and released its first feature-length work, Hakujaden, in 1958. Already interested in animation, Tezuka closely followed Tôei’s creation, and made a first visit to the studio in 1956, back when it was still occupying its predecessor’s Nichidô’s facilities [2]. But nothing happened until late 1957. Towards the end of Hakujaden’s production, Tôei started planning its two next works: one, adapted from a Japanese story, would be Shônen Sarutobi Sasuke, released in 1959; the other, adapted from a Chinese one, was Saiyûki.

The idea to invite Tezuka on Saiyûki came from young Tôei producer Daisaku Shirakawa. When Saiyûki’s title came up in discussions, Shirakawa immediately thought of Tezuka’s manga Boku no Songoku, and suggested using it as a base for the film. Once the idea was approved by higher-ups, Shirakawa contacted Tezuka, who immediately accepted – he was starting to seriously consider creating his own studio at that point, and had already bought a plot of land in Fujimidai, not far from Tôei, where he would build his house and workshop [3]. However, Tezuka was not content with just providing the original work for the film: he wanted to be directly involved in its creation.

The mangaka’s work on Saiyûki consisted of creating story boards. Unlike storyboards (ekonte), story boards were made before the writing stage, and consisted of sketches delineating the story and main scenes. These were not in use in Tôei, and Tezuka most likely took the idea from Disney Studio’s organization, in which story boards were one of the main steps of the creative process. Shirakawa agreed and, from September 1959 onwards, Tezuka started regularly visiting Tôei to draw and discuss his work with Shirakawa and director Taiji Yabushita. He was extremely productive: Tezuka stated that he drew around 500 pages [4]. 

Tezuka (center) commenting his story boards in Tôei. On the right is director Taiji Yabushita [Tsugata 2007]

The only problem was, this took time – the story boarding phase may have lasted up to a year. This was due not only to Tezuka’s productivity, but also to his schedule: busy working on manga, he never had any opportunity to fully devote himself to Saiyûki’s production. Before moving on further, I’d like to state things on this topic strongly once, to avoid having to repeat them further on. During the entire length of time narrated in this series, Tezuka had multiple manga series going on at the same time (you can check this chronology to get an idea). Although he had assistants and managers, he constantly had to think about stories, deadlines, and draw: most testimonies portray him as constantly hounded by editors. On top of that, he had to attend various events (lectures, fanclub meetings, etc) and always endeavoured to stay in touch with the latest trends in film, animation and manga. It would be reasonable to assume he was in a constant state of overwork and exhaustion. This, more than any alleged lack of business sense, is in my opinion the cause for some of the biggest mistakes Tezuka made in his animation career. He simply never had the time to consider things, leading him to make what appears in hindsight like misguided decisions, or give his complete trust to people who would not meet his expectations or simply betray him. Certainly, he was not passive, but he was often a victim of events or decisions which he had no responsibility for, or the consequences of which he could not predict.

In Saiyûki’s case, the solution Tezuka found was to send people to Tôei in his stead. Around April 1959, his assistant Sadao Tsukioka and friend Shôtarô Ishinomori began visiting the studio, where they started the work of cleaning up the story boards and transmitting messages between Yabushita and Tezuka [5]. But, even cleaned up, Tezuka’s boards were considered unusable as is. Veteran live-action scriptwriter Keinosuke Uekusa was called on to write a proper scenario, after which Yabushita did the sequence division, ultimately finalized into a proper storyboard by Sadao Tsukioka. It is at this stage that the first major disagreement between Tezuka and the Tôei side happened: the mangaka wanted to end the film on the death of the female character Rin Rin. Shirakawa and Yabushita strongly disagreed, arguing that a children’s film could not end in such a way, and in the end Tezuka gave in [6].

The other issue was with character designs. Tezuka had done them himself, presumably based on his original boards. But they were considered unusable as well, and went through a complex series of revisions, closer to Tôei’s original process: on Tezuka’s suggestion, Tsukioka reworked them and even did some original designs, then each animator did their own proposals and modifications at the animation stage [7]. In the end, Tezuka did not recognize the final film, stating that only half of it came from his ideas [8]. Without Tezuka’s boards, it’s hard to tell how much of that is exaggeration; but from outside, Saiyûki appears notably different from the rest of Tôei’s early production, and Tezuka’s touch is, in my opinion, visible all over it.

A chronology of Tezuka’s involvement with Tôei and of Saiyûki‘s production

As I have already commented on Saiyûki many times before, I will keep things brief here. It is a special movie in Tôei’s production for three reasons. First, it represented the true coming of age of the studio’s first homegrown generation of animators, represented by Yasuo Otsuka and Daikichirô Kusube. Second, it contained an unprecedented variety of tones and styles, which I believe can be directly attributed to Tezuka’s input and the collective process enabled by story boards. Third, it pioneered multiple expressive and then-experimental techniques that would be systematized in Mushi’s production, such as intense framerate modulation or dynamic deformations. For Yasuo Otsuka, Tôei’s animators enjoyed an unprecedented degree of freedom on the film, whereas for Shirakawa, it was from Saiyûki onwards that an “anti-Disney” movement started appearing within the studio [9]. It was led by animator Makoto Nagasawa, who had done some of the most experimental parts of the film. These elements would play a key part in the intense rejection that Tôei’s following film, Anju to Zushiomaru, faced from within the studio.

In spite of all that, Tezuka was obviously dissatisfied with Saiyûki. Although he agreed to continue collaborating with Shirakawa for 2 more films in Tôei, this experience is largely thought to have been the final trigger for the creation of Mushi Pro (though, as I have explained, he was already close to making his idea a reality before Saiyûki). But before that, a significant event happened. By the end of Saiyûki’s production, Shirakawa made the following offer to Tezuka: a TV adaptation of his manga Tetsuwan Atom. The mangaka agreed. Then, the producer went to Tôei’s higher ups with a plan for a “TV slideshow” (kamishibai), which faced immediate and unconditional rejection – Tôei’s creative leader, Zenjirô Yamamoto, reportedly called Shirakawa an “idiot” for suggesting the idea: an animator’s job is to make things move, so why have them work on still pictures? [10] The plan was abandoned, never to resurface again – or so Shirakawa thought.

Tezuka and Japanese animation in the 1960s

The next development, then, was Tezuka creating his own studio. But this did not happen from nowhere or on a whim. By the time Saiyûki came out in August 1960, the mangaka was already closely involved with the budding Japanese animation industry beyond Tôei Dôga.

Tezuka’s first attempt to work in animation dated back to 1947. In August, during the summer holidays, he traveled from Osaka to Tokyo to offer some of his manga, including the famous Shin Takarajima, to Tokyo publishers. On the way, he stopped by Iwao Ashida’s Cartoon Films Institute, a studio which dated back to before the war, to obtain a job as an animator, but was refused. Ashida argued that Tezuka’s style was more fit for manga [11]. 

Things stayed at a standstill until 1956, when two things happened simultaneously: Tezuka started to pay visits to Tôei, and was invited to join the “Japanese Association of Animation Artists”, which had been created by Tôei’s Zenjirô Yamamoto (the activities and goal of which remain unclear) [12]. Although the exact circumstances are fuzzy, Nobuyuki Tsugata hypothesizes that Tezuka was invited by fellow mangaka Ryûichi Yokoyama. Yokoyama had been to the US, visited Disney Studios, dabbled in animation before and during the war and, in January 1955, created his own animation studio, Otogi Production [13]. He knew Tezuka since at least 1951 and the latter regularly visited Otogi Pro [14]. By the late 50s, Otogi had expanded its activities to CM production and would create the first animated series for TV in Japan, Instant History, in 1961. It was therefore Tezuka’s go-to destination when he decided to start his own thing: in August 1960, once Saiyûki was completed, he visited Otogi, accompanied by Shirakawa, and announced his intent to make animation. Both Shirakawa and Yokohama tried to dissuade him, to no avail [15].

At the same time, another significant development was happening in Japanese animation. In 1960, artists Yôji Kuri, Ryôhei Yanagihara and Hiroshi Manabe created the “Three Animators Group” (Animation San Nin no Kai) and started yearly screenings of experimental animation works in November. Chief representatives of the “anti-Disney” movement which swept over Japan, they introduced the works of Western experimental animators, notably Norman McLaren, in their country. Naturally, Tezuka attended their screenings – he reportedly did not find the first very interesting, but started becoming increasingly curious from the second on. It is probably at this second screening, in November 1961, that Kuri reportedly taunted Tezuka, telling him, “now we’re waiting to see your work” [16]. At that time, Tezuka’s work was indeed coming: the mangaka had created his animation studio on June 1st, 1961.

Tezuka Production, Animation Division

As seen above, Tezuka had finally decided to create his studio by August 1960. In December, he first met with Eiichi Yamamoto, who had quit Otogi Pro when he heard Tezuka was recruiting [17] and, in the months that followed, with former Tôei animator Yûsaku Sakamoto. He also bought a multiplane camera, which he installed in his garden’s shed. By June 1961, the studio counted 5 members: Tezuka, Yamamoto and Sakamoto, Kazuyuki Hirokawa, a highschool graduate member of Tezuka’s fanclub who had some experience with photography, and Chizuko Watanabe, who had worked as a cel painter in a CM production company [18]. They gathered and worked in Tezuka’s house, under the name “Tezuka Production Animation Division”. The stated goal was to make “art animation”, but Tezuka seems to have already been aware that they would need to produce “entertainment works” alongside to obtain funds – he just hadn’t envisioned the shape these would take yet [19].

A few days after the team had first gathered in Tezuka’s house, the mangaka submitted a plot proposal, and work started on Aru Machikado no Monogatari. Tezuka would be in charge of the storyboard, on which he could afford to spend 2 hours a day, while Yamamoto and Sakamoto did the character designs [20]. Tezuka didn’t want to do them himself, seemingly for two reasons: one, because he envisioned animation work as collective and didn’t want to center things on him; two, because he believed his own style was too “cartoony” and didn’t fit such art animation. Tezuka’s storyboard was then reworked by Yamamoto and Sakamoto, as it only transmitted a “rough image” and the indications were too basic to be used as is [21]. This would remain a characteristic of Tezuka’s storyboards for some time. The storyboarding and designs were completed in October 1961, and  animation began.

By that point, “Tezuka Animation” counted 15 members. New people kept coming in, most of them from Tôei. It is important to stop and explain the reasons for this continuous movement from Tôei to Tezuka/Mushi Pro, which did not stop until at least 1963. The film that followed Saiyûki in Tôei was Anju to Zushiomaru. Conservative in both narrative and aesthetics, it faced heavy resistance from the staff. A first union had been formed in March 1959, but had been disbanded after 6 days by Tôei’s management [22]. Anju to Zushiomaru single-handedly revived it: after some time spent in clandestinity, a new union was created in March 1961, apparently following from a proposal by Yasuo Otsuka. It had two demands: improvement of salaries, and the opportunity for the production staff to offer its own projects to the planning department. 

Makoto Nagasawa, the leader of the “anti-Disney” movement within Tôei, was elected chairman and the union went on strike. President Hiroshi Okawa’s answer was swift and brutal: the studio was locked out, and some of the most agitating elements were blacklisted and either left or were fired [23]. Most of them joined Mushi in the following months. The union managed to live on by affiliating itself with the national Federation of Theater and Film Workers’ Union and concluding a deal with Tôei’s management.

Even aside from this development, Anju to Zushiomaru caused discontent. In November 1960, after reading the film’s script, Gisaburô Sugii and Yûsaku Sakamoto both determined that they’d leave Tôei together [24]. Sakamoto joined Mushi after a few months spent in a CM company, whereas Sugii remained in Tôei until early 1961, and then spent 6 months living on his severance payment, pachinko, and 1985 yen gathered by the union to support him [25]. It was then that Sadao Tsukioka, who had joined Tôei, visited Sugii and told him about Tezuka’s studio, which he visited and joined. Sugii then invited his closest friend from Tôei, with whom he had lived for some months and written manga to make ends meet in their early time in the studio – cel painter and aspiring director Shigeyuki Hayashi, whom he nicknamed Rintarô [26]. Daisaku Shirakawa encouraged his wife Shigeko to join as a cel painter. However, it was Sakamoto who attracted the most recruits: he regularly visited Tôei, just had to say that Tezuka needed people, and brought new faces to the studio on his way back [27]. Later, he would also go to Osaka to scout the members of CM studios working there, ultimately leading Sadao Miyamoto and Renzô Kinoshita to join [28].

The issue became putting all this new staff somewhere – at first, Rintarô worked from home because there was no more room in Tezuka’s house [29]. In September 1961, construction of a studio started on the land Tezuka owned. It was completed in April 1962. By that point, things had moved on to the next stage: Tezuka Animation was renamed Mushi Production, and the Tetsuwan Atom project was on the rails.

Plans of Mushi and Tezuka’s house over the modern parcels. Right is a picture of Mushi Pro (building on the left) and Tezuka’s house (back right) in 1970

Mushi Production and Tetsuwan Atom

Yûsaku Sakamoto had not met Tezuka during Saiyûki, but he had heard about Daisaku Shirakawa’s proposal to make a TV version of Atom. Building onto it, he had proposed a movie version of Jungle Taitei, but also faced a refusal. Now in Mushi, he was also aware of the necessity to make commercial works, as nobody in the studio believed they could sustain their activities on Tezuka’s manga alone. Finally, he knew about the American “TV cartoons” which aired on Japanese TV – that is, Hannah Barbera’s productions like The Flintstones and The Yogi Bear Show [30]. In February 1962, he individually approached Eiichi Yamamoto and Gisaburô Sugii to hear their thoughts about a TV adaptation of Astro Boy. They both thought it was good and, in the following days, Sakamoto and Yamamoto shared the idea with Tezuka, who immediately approved it [31].

While the existence of Hannah Barbera’s series made the existence of a Japanese animated TV series possible, all three men agreed on one thing: they wanted to do something completely different. Tezuka found the US cartoons “boring”, describing them as filled with “poor jokes and filling up airtime with mouthflapping” [32]. In his autobiographical novel, Yamamoto puts pretty much the same words in Sakamoto’s mouth [33]. While Tezuka’s opinion of Disney fluctuated a lot, going in contradictory directions over time, it seems that he leveraged similar criticism at the Burbank studio in the early 60’s: he described their films as “picture stories” (ebanashi) which hid their complete lack of narrative depth behind visual technical perfection [34].

The choice of Tetsuwan Atom was then calculated as an anti-Hannah Barbera and anti-Disney move: the flagbearer of “story manga” would also be the first “story anime”, not carried by gags but by actual structured narratives. This motivated the choice to produce 30-minute episodes – length was necessary to develop real stories.

However, things weren’t settled from the start: while Sakamoto had initially suggested Atom, the three men also considered adapting Tezuka’s Zero Man instead. The final choice of Atom was made by Tezuka, but it was largely influenced by another individual, who quickly became one of the most important members of Mushi: Kaoru Anami. Anami was an executive in the Tokyo branch of advertising company Mannensha and the husband of former Tôei animator Kazuko Nakamura, who joined Mushi in May 1962 (she would most often be credited under her husband’s name, as Kazuko Anami). Through her, Anami heard about the Atom project and met Sakamoto the same month. He was extremely enthusiastic and offered the following: Mannensha would find a sponsor and TV station for Atom, while he would eventually join Mushi Pro (which he did in July 1963) [35]. After meeting with Yamamoto, Anami discussed his plans with Tezuka. There were two reasons they should chose Atom instead of Zero Man, Anami argued: one, Atom was more popular; two, it was a better vessel for selling merchandise [36].  In fact, it wouldn’t be the first time Atom appeared on television: a puppet play and a live-action series version had already aired in 1957 and 1959.

It is at this point that work on the pilot film for Tetsuwan Atom began, in parallel with Machikado’s animation. Mushi was reorganized in two groups: the TV division, led by Sakamoto, and the film division, led by Yamamoto. At this point, Mushi counted 41 members [37], a dozen of which were put to work on Atom. Tezuka did the storyboard and the rough animation, refined into proper keys by Gisaburôo Sugii; second key animation was done by Motoaki Ishii, Rintarô and Yamamoto; in-betweens by Kiyomi Numamoto, Kazuko Nakamura and Tsunako Miura [38].

The myth of “Tezuka’s curse”

One of the most important issues relating to Tetsuwan Atom is its budget. Under the nickname of “Tezuka’s curse”, it has always been said that Tezuka sold Atom under its actual price, thus dooming the anime industry to low budgets, low salaries, and generally poor working conditions. Given how important Tezuka’s decision supposedly was for the entire future of the anime industry, there is a vast amount of literature on the topic, to the point that things are becoming confused. Tezuka himself never denied the facts, but consistently stated that he had no choice and was implicitly forced by the sponsor and TV station [39]. But, even before I review the facts, I want to make my conclusions clear: 1) Tezuka knew what he was doing and 2) his decision had virtually no impact on the organization of the anime industry and its budgets.

First, let us begin by reviewing TV anime’s financing and distribution model, created by Kaoru Anami based on live-action TV series, which would more-or-less remain the same until the 90’s. As Anami came from an advertising company, these were at the center of a triangular system: the ad company bought TV episodes from the animation studio and sold them to the TV stations. They would then be aired on a slot bought by the sponsor, which additionally acquired merchandising rights, in this case from the animation studio itself.

For Atom, there were two people present at the negotiations on Mushi’s side: Tezuka and his manager Yoshiaki Imai. But most of the actual business talks were led by Anami, with the final decisions being made by Tezuka. At first, Anami approached TV station Nippon TV and Morinaga Foods. But he also discussed the project with Daisaku Shirakawa, whose brother worked in Fuji TV – when he heard about it, Shirakawa instantly contacted his brother and told him to buy Atom at all costs. Fuji immediately approached Mannensha, and the deal with NTV was called off; for the sponsor, Morinaga had failed to follow suit, and a partnership with Meiji Seika was concluded [40]. With all the actors in place, in Fall 1962, discussions about the budget started.

First, the president of Mannensha’s Tokyo branch, Ichirô Kimura, offered an amount of around 300,000 yen for an episode. Tezuka found this much too low and refused [41]. It seems that Kimura thought the cost of an episode would be between 150,000 and 200,000 yen, so this must have sounded like a generous offer – Tezuka brought him back to reality. Following this, Meiji and Fuji directly approached Mushi instead of going through Mannensha, and offered a price of 1,200,000 yen per episode. Once informed, Anami and Imai were ready to agree, but not Tezuka: he lowered things by half, concluding the deal at 550,000 yen [42].

Although it’s hard to tell what estimates Tezuka had at his disposal, he was well aware that 550,000 yen was not enough: the first episode of Astro Boy cost 1,300,000 yen, the second one 900,000 yen, and the prices later oscillated between 1 and 2 million yen [43]. So, why did Tezuka make that choice? Meiji and Fuji based their offer on the price of a live-action drama’s episode – between 450,000 and 600,000 yen [44]. Understanding that animation cost significantly more, they just doubled the number. Hearing this, Tezuka reasoned the opposite way: he wanted the price of an Atom episode to be around the same as that of a live-action episode for two reasons. First, to avoid putting his partners in a bad position and create unnecessary competition. Two, by giving such a low price which he knew put Mushi at a financial disadvantage, he hoped to discourage any other animation studio from trying the adventure, in order to give Mushi an effective monopoly on domestic TV animation [45]. Whatever one thinks of Tezuka’s decision here, it is impossible to say the mangaka didn’t know what he was doing: it was a conscious, motivated and justified move.

Considering that TV anime instantly developed following Atom, Tezuka’s dumping strategy was an obvious failure. But the reason wasn’t that studios went ahead anyways with such low budgets: it is that budgets were never so low in the first place. While we don’t have exact numbers, it seems that episodes of Eightman (TCJ) and Wolf Boy Ken (Tôei) were sold between 550,000 and 1 million yen – probably still under budget, but far less than Atom would initially have been [46]. On the other hand, according to Sôji Ushio, P Production’s Zero Sen Hayato’s episodes were sold at 3 million yen each [47]. Later, Mushi’s Jungle Taitei was sold at around 5 million the episode, twice its actual budget [48]. 

Moreover, budgets were renegotiated: according to Shôzô Sudô, the initial contract signed between Mushi and Mannensha only applied to 2 cours (which seems to have been Atom’s initial planned run) and the amount was doubled afterwards; while they give different numbers, Hideaki Itô and Kenji Nagao similarly note that around August 1963, the 550,000 yen were raised to 750,000 [49]. Tezuka’s initial price was used as a negative standard, but it never had any currency for more than a few months.

That’s not all. Sales department member Shôzô Sudô revealed that Anami and Imai were not satisfied with this. So, without Tezuka’s knowledge, the three men went to Meiji arguing that 550,000 yen were not enough, and came back with a second contract stating that the sponsor would give Mushi an additional million yen per episode. There were, in other words, two contracts, one of which remained secret until 2007 [50]. If this is true, this raises two questions: why did it remain secret for so long? did Tezuka ever know about it? Nobuyuki Tsugata, who reports Sudô’s story, provides no answer, which is in my opinion the biggest weakness of this version of the events. But it is easy to fill the gaps.

For the first question, the two-contracts deal was at best irregular, if not shady. It would make sense that none of the persons involved would discuss it openly. Moreover, while I don’t know when Imai died, it seems that Sudô was the only one among the three persons on Mushi’s side to outlive Tezuka long enough that telling this story wouldn’t hurt anyone’s reputation. As for the second question, Tezuka never knowing about this seems strange: as the president of Mushi Pro, he should have noticed such an amount of money coming in for no good reason. But it is important to remember that he had other things on his mind and that, while he attended business negotiations, he rarely led them himself. It would therefore have been easy for Anami or Imai to hide the contract from Tezuka and attribute the profits to another cause. As we will see later on, it wouldn’t be the last time Anami acted on his own.

But then, even if we assume that Sudô’s memory was so confused that he made up the whole thing, Tezuka had all reasons to believe that the low price at which he sold Atom would not cause major troubles for Mushi in the long run. First, there were all the cost-cutting measures used on the show, notably the cel bank system, which ensured that, as it went on, Atom would cost less and less money. Second, there was all the so-called “pocket money” Tezuka earned from his manga, most of which he poured into Mushi. Then, in Spring 1963, Tezuka  signed a deal with US network NBC Enterprises, selling 52 episodes of Atom at a unit price of 3 million yen [51]. Finally, the contract signed with Mannensha, Fuji and Meiji contained one important clause: Mushi retained all the copyrights to Atom, and would therefore fully benefit from all the merchandising that came from it. It seems that this was Tezuka’s personal idea [52].

Osamu Tezuka (right), Bill Schmidt from NBC (left) and agent Kiyoshi Fujita (left of Tezuka) [from here]

It is at this point that we must discuss the real curse plaguing the anime industry, which would have been avoided had other anime studios done like Mushi: the fact that animation studios don’t hold any of the copyrights of the properties they produce, effectively making them subcontractors for TV stations and sponsors. I want to emphasize the fact that this did not apply to Mushi, at least in the first years: Tezuka himself earned royalties as the original author, and his studio was the copyright holder, an equal partner in the relationship and the initiator of the deal. But as soon as other studios like TCJ entered the game, things would be reversed, as TV anime series became adaptations commissioned by stations and sponsors instead of coming from studios themselves.

With this in mind, we can understand Tezuka’s strategy better: by underselling Atom, he effectively made a bet. Yes, it put Mushi in the red – but only for a short time. If it worked, it would all be compensated and more. The studio just had to bear it for a few months. And the fact is that this partly worked: Mushi soon lost its monopoly on TV anime, but Atom remained extremely popular, and it seems that the studio recouped its costs thanks to merchandising. Business-wise, Tezuka had succeeded.

Before moving on, I want to discuss one important implication of all this. Remember that Atom’s initial goal was to make profit as an entertainment work so that Tezuka could fulfill his initial objective, that is to make art animation. But with Tezuka making such a bet, this became untenable: Atom might profit in the long run, but that profit remained extremely fragile, perhaps more than the one Tezuka made by selling his manga – even if the amount would be much higher. Certainly, neither him nor the other people at Mushi Pro could have predicted the changes Atom would have on the studio and the entire Japanese animation industry, but there’s no way they couldn’t see what I just mentioned. In other words, underselling Atom represented Tezuka’s choice to move away from art animation towards entertainment works. Tezuka would regret that decision later on, but he was responsible for it, and nobody else.

Anyways, now that the budget issue is out of the way, it is time to move on to the second element of “Tezuka’s curse”, that is low salaries. First, what do we know of salaries at the time? It is said that in the early 60s, a university graduate’s best first salary would range between 18,000 and 22,000 yen [53]. In Tôei, we have the following range: when they joined as contractual workers, Gisaburô Sugii (in-betweener) and Rintarô (cel painter), neither of whom had graduated from highschool, both earned around 4,000-5,000 yen. When they left 4 years later (perhaps now full-time employees), their salaries had doubled [54]. It seems that the entry salary for full-time university graduates in Tôei was 12,000 yen, although Hayao Miyazaki, who was in such a case, earned 19,500 yen [55]. 

In comparison, Mushi paid much higher than the rest of the industry. When he joined, Yamamoto earned 30,000 yen a month, more than double what he got in Otogi Pro; it would be reasonable to assume that Sakamoto got the same offer. It seems that Sugii and Rintarô were paid a bit less: 20,000 a month for the former and 21,000 for the latter. Akihiro Kanayama, who joined some time later, said he earned “around as much as a university graduate” – so probably around 20,000 yen as well. Closer to Tôei’s rates, Kazuhiko Udagawa claims he earned around 13,000 yen, though that number reached 35,000 later on [56]. 

Sudô, who initially joined in the photography department, got 70,000 yen a month. He also tells the story of someone who asked for 550,000 yen as a joke, and actually got them [57]. Scripts were extremely well paid as well: Masaki Tsuji and Aritsune Toyota give the number of 50,000 yen per script, with Toyota adding another 100,000 for a synopsis [58]. Finally, all the way down the ladder, Yoshiyuki Tomino mentions that a newbie production assistant could hope for around 10,000 yen a month. On the other hand, Jun Masami, who joined later in the same position, noted that he earned 17,000 yen [59]. 

Make no mistake, these are ridiculously high numbers. I can’t tell whether salaries remained so high for the studio’s entire existence. But these figures remain somewhat absurd and, even if Mushi earned a lot from merchandising, it was probably not enough to cover all the costs incurred by such salaries. It is important to note that, according to Yamamoto, the one who negotiated his salary was Imai, not Tezuka – although certainly they must at least have discussed it together before. But, assuming that Tezuka was the one who decided on such numbers, I think the reason was partly that he did not care about money, and partly pure generosity. Considering the situation in Tôei, Tezuka didn’t really need to raise salaries that much to attract staff – had he only considered things in terms of profits and losses, this wouldn’t have made sense. What he had in mind was therefore something else.

Machikado, Atom #01 and “Tezuka anime”

When the negotiations happened and the contract was signed, Atom‘s 15-minutes pilot was completed: it naturally served as a basis for the talks. The airing date of January 1st, 1963 was decided, and Mushi’s staff decided to use the pilot as the first half of the first episode that was to come out a bit more than 3 months later. The plan was to have 5 episodes ready by that date. It seems that, including the pilot, episode 1 was completed in 10 weeks [60]. Additionally, on November 6th, 1962, Mushi held its first screening event, showing Machikado, the barely-completed first episode of Atom, and the short Osu, a Yamamoto project that probably kept him busy during the short time between the completion of Machikado and his work on Atom [61].

Because of the obvious differences in content and format, it would be easy to picture Machikado and Atom as two different works, one “artistic” and the other constrained by the limits of the TV format. However, they were made at the same time, by the same people. I actually think they both equally showcase Tezuka’s vision in animation. For that reason, I will analyze them together. On the other hand, it would be best not to focus things too much on Tezuka, whose direct involvement in both works was limited. This was for two reasons, which I already mentioned: he was too busy with manga, and probably conceived animation as fundamentally collective, something that should not be centered only around him.

In its themes and execution, Machikado is not particularly original. But it does have remarkable stylistic features which deserve commentary. In my opinion, the film is formally built around the idea of depth: the flatness of the characters contrasts against the backgrounds, drawn in a far rougher style. The entire film revolves around vertical (the little girl at her window/the posters down in the street) and horizontal (the gutter up on the roof/the pans over the posters) movement. Shots of the street as a whole often emphasize the perspective. But what is most notable and must have felt most new at the time is the use of the camera, fairly distinct from Tôei’s use of the multiplane.

First is the use of zooms, not to emphasize the non-photographic nature of the image but to create depth, as in the scene where the tree tries to send its seeds far down into the street corner. Machikado also features two prominent instances of complex camera movement. The most impressive is the second one, the penultimate shot: lasting for 1 minute, 20 seconds, it begins as a subjective camera from the point of view of the little girl. After a pan up and a zoom, the camera “moves back” while “turning” upwards, revealing the body of the little girl – clearly imitating live-action camerawork. But then, in an unexpected move, the entire screen is rotated – a disorientating technique unusual in live-action – before a long pan fully exploiting the possibilities of the multiplane camera. It seems that 6 layers of background and 2 cels were put on top of each other for this exceptionally intricate shot. 

Moving to animation, it seems that framerates were carefully considered in terms of aesthetic consistency. All the human and animal characters, mostly animated by Sakamoto, (except for the moth) are on 1s or 2s, in a realistic fashion characteristic of “full” animation. On the other hand, the posters (mostly handled by Shûji Konno) are decidedly “limited” – there are sometimes only 2 to 3 new frames in a second, with extremely wide spacings and no poses between each extreme. Then, there is the moth, mostly animated by Gisaburô Sugii, which obviously owes a lot to Makoto Nagasawa’s work on Saiyûki while announcing many of the features of Atom’s animation. First, it is highly modulated – sometimes on 1s and 2s, sometimes on 3s, 4s or even lower. Its movement is irregular and, for that reason, both expressive and fun. What makes it even more entertaining is the systematic use of deformations and smears, which would have been an experimental technique in the early 1960s, especially compared to Tôei’s output. While these smears are sometimes created by drawings alone, more often than not, they were expressed through brushwork applied directly on the cel.

In comparison, Atom’s animation is not so thought-out. But it is far more complex than most commentaries have noted: most times, when the characters move, they move on 2s, and the episode opens on an extremely frame-consuming technique, background animation. In fact, Eiichi Yamamoto notes that Tezuka exceeded the amount of frames they had set for the first episode, and that most of the show’s “limited animation” techniques were developed on the spot to compensate for the excessive movement of Tezuka’s work [62]. This is consistent: among the 70 episodes of Atom that I’ve seen, those directed by Tezuka regularly feature the most movement. In this specific episode, many motions are on 1s or 2s, notably cycles and effects, but also some of the most lifelike performances such as Pr. Tenma’s mad laughter when he has the idea of reviving his dead son Tobio. 

Because of the radical pose-to-pose spacing and almost complete lack of follow-through, it is true that movement often looks very stiff. However, this is not always without a good reason. The best example would be the moment when Atom wakes up for the first time: the framerates are extremely slow (sometimes as low as 10s), but are modulated, occasionally switching from 3s to 2s to 5s, all to express the progressiveness of Atom’s awakening and the awkwardness of his movement. Initially intimidating, the animation becomes charming and cute, largely thanks to squash-and-stretch and the adorable, relatable expressions on Atom’s face. There is no other word to describe this sequence than “excellent animation”. In terms of pure animation, this episode is no doubt one of the best made in 1960s anime.

Atom #01 is carried even further by its storyboarding. The opening scene, entirely silent until Tenma’s heartbreaking cry, is a lesson in efficiency and visual storytelling. The shot compositions are varied, diverse and complex, whether that is through framing techniques, use of silhouettes and shadows (which are often animated), and camera movement. This episode also remains a recognizable Tezuka work in its pacing, as gags are always inserted in the most unexpected moments – the most surprising one being at the most dramatic scene of the episode, when Tenma chases Atom away and is then thoroughly ridiculed. On the other hand, the editing is often basic, just there to link shots together. 

Whereas most commentators downplay or ignore the obvious visual strengths of Atom #01, the element that is most often discussed is the sound. Machikado’s use of synchronized music already showcases Tezuka’s tendency to use sound for dramatic effect, but Atom pushes that a level further, not only through music, but sound design proper. From the first episode onwards and for its entire run, Atom is in fact a very loud show, which may become annoying to the modern viewer. But its soundscape is incredibly diverse, as new effects always come up to recreate, for example, the sound of different robots walking in different ways.

Taken together, then, Machikado and Atom #01 do not, in my opinion, represent Tezuka’s choice of “limited” animation. Instead, they showcase an understanding of animation as a total artform, encompassing all the techniques available to create movement – from framerate modulation to camera movement -, a highly developed visual language, a focus on story and a careful, if not always subtle, use of music and sound. As production progressed and the schedule became tighter, Atom’s animation became less intricate – though it is ultimately similar to any other anime, where certain scenes exhibit particularly good animation and others are just still. Considering the level of 60s TV anime, Atom’s animation is far from below standard. But one thing that did not change was the creativity of its direction, which kept finding new ways to create emotion, drama or gags. It would not be an exaggeration to say that anime’s visual language was basically created on Atom, and that there are few shot compositions, however bold, that were not first tried on this epoch-making show.

The visual inventiveness of Atom: screens from the first two cours of the show

Atom‘s production

Atom’s run stretched over 4 years, from January 1st, 1963 to December 31st, 1966. During such a long course of time, Mushi Pro considerably expanded, and this makes it impossible to consider its production as one single, continuous process. For ease of explanation, here, I will therefore break it down in multiple “phases”, each of which had a slightly different organization, staff and style. But it should still be noted that these divisions are not absolute: new people kept arriving in Mushi, meaning that the staff and the way they worked kept changing in ways more complex than my summary accounts for. Moreover, as in any anime production, while Atom has an overarching aesthetic, the small variations between each individual episode are endless, and I will not describe them all here.

The first phase: Tezuka’s experimental pipeline (#01-#03; 07/1962-01/1963)

When the production of the Atom TV series was approved, Tezuka decided to take charge of the first three episodes, which would act as a sort of pilot or standard for all the rest of the show [63]. While I have said before that Tezuka probably conceived animation as collective, when he was actually put in charge of something, it all became centered around him – not because he had any authoritarian tendencies, but because he probably didn’t know any other way to work. As such, the production of Tezuka’s episodes was very unusual, and completely different from what Atom’s process would become. This is why I call it “experimental”, as Tezuka himself and the rest of the staff seem to have mostly played it by ear.

As Tezuka adapted his own manga, he wrote neither scripts nor storyboards: he directly started with the animation, and wrote the dialogue as he went – the actual script used for the voice recording had to be written by the production assistant, Eiichi Kawabata. Tezuka’s key animation was also quite unusual: he reportedly made no timesheet and did not number the cuts or the frames, making it almost impossible to understand by anyone else other than him [64]. As a result, he had to personally supervise all the following steps in the production process, from clean-up to photography. But there was a problem: Tezuka spent most of his waking hours on manga, and could only attend animation work irregularly. If episode 1 took so much time, it was largely because of Tezuka’s schedule.

Tezuka attending photography work circa 1963 [Tsugata 2007]

This account by Yamamoto needs to be confronted with Tezuka’s actual drawings, some of which have thankfully been published. The images below are from episode 2. Some, at least, include cut numbers, though that is not the case for all. Moreover, none of these – and this also applies to the drawings from episodes 1 and 3 included in the same book – contain dialogue, which must have been written on separate paper. However, we do see indications that would usually belong to the storyboard, such as the length of shots or camera movement (see, under “Cut 354”, the following indication: “slow Track-Up / 1 second”). Some of these drawings are remarkably rough, and it’s apparent that Tezuka had no idea what to do with backgrounds – on multiple drawings, we can find “backgrounds?” written.

Some of Tezuka’s animation drawings for #02 [from Tezuka, 2018]

As soon as episode 2, it seems that Tezuka started to leave more leeway to the animators working after him [65]. Things got more efficient: episode 2 was made in 8 weeks and episode 3 in 7. This compression had results on the finished work: episode 2’s animation is already far less ambitious than that of episode 1, and episode 3 is no different from the standard of the second phase. Realizing that things could not go on like this, Sakamoto and Yamamoto appealed to Tezuka on the necessity of reorganizing the production as early as November 1962 [66]. The mangaka, well aware of the trouble he caused, agreed: this was the beginning of Atom’s second phase.

The second phase: the establishment of the rotation (#04-#47; 11/1962-11/1963)

Put simply, what Sakamoto and Yamamoto did was to adapt Mushi’s pipeline to the necessities of industrial production. They completely restructured the studio with themselves on top: Sakamoto became Atom’s chief director and the effective leader of the studio, and Yamamoto chief of the “literature and direction” section. Under them, they assigned chiefs to each section: animation, ink and paint, backgrounds, photography, production and management. Moreover, they created a precise planning document for the production, which initially laid out the exact rotation and schedule for the first cour of the show, up until episode 13 [67]. Finally, they established the team of 6 episode directors who would work in rotation: these were Sakamoto, Yamamoto, Sugii, Motoaki Ishii, Shûji Konno and Tezuka himself. In such a system, each director would have up to 6 weeks to produce their episode. As things advanced, new people were promoted to episode direction, thus making the rotation more spaced out.

This system rested entirely on one person, the episode director, and one document, the storyboard. Like Tezuka, directors didn’t write scripts but worked directly from the manga (sadly, I don’t know following what process manga chapters were chosen) and wrote the dialogue in the storyboard. These were then checked by both Tezuka and Sakamoto. After that, they did most of the key animation themselves (even though the credits always provide multiple names and directors probably helped each other out, meaning that no episode is entirely solo), which was then cleaned-up by second key animators and in-betweened [68]. Everything proceeded until the production of a rush film, which was checked by the director; it was at this stage that they would ask for retakes, after which recording, dubbing and mixing were made.

To me, three questions remain regarding Atom’s production process. The first relates to character designs, notably those of guest characters. Settei, or standardized reference materials, existed, as many remain today and have been published here and there [69]. But who drew them? According to scriptwriter Aritsune Toyota, who started working during the third phase, writers made written descriptions of the characters, which were then put to images by each director in their storyboard and adapted by animators [70]. In the 1999 collection of Atom storyboards, it is stated that “most of the settei were drawn by Mushi Pro’s staff, but in many cases, Tezuka’s original drawings were traced by the key animators” [71]. This is supported by Itô and Nagao’s research on the production documents: it seems that the first settei for Atom’s expressions was only completed for #14, based on Tezuka’s drawings for episodes 1-3 [72]. Those in charge of that work were probably either the episode directors or their assistants.

Settei for Atom, Pr. Ochanomizu, and guest characters from episodes 112 & 128 [Tezuka 1999a & 2018]

The second question is that of “second key animation”. Its existence is attested in credits, but direct testimonies from the production are remarkably vague on the exact nature of the task. Yamamoto mentions that he did “first key animation” in Otogi Pro, thereby implying the existence of second key animation [73]; for Mushi, he only mentions “key animation assistants” [74], but not exactly what their role was. On the other hand, Sugii says that episode directors did key animation and timesheets, and that clean-up and in-betweening (nakawari, not dôga) were done by other people, without specifying whether these tasks were clearly separated [75]. In any case, the presence of second key animation makes sense for two reasons: one, because it was used in Tôei and probably imported by its animators, and two, because Atom’s schedule encouraged such a division of the pipeline. It is possible that there was actually a three-step process: “rough” rough key animation by the episode director; proper key animation by key animators; and then cleaned-up second key animation. At least, that’s how things happened during the experimental phase, and they may have remained like that in the early stages, when Mushi’s workforce was still small. But then, the disappearance of the credit later in the production raises yet more questions: why did it disappear? who was in charge of the clean-up and applying corrections/retakes after that point? 

Finally, there is the issue of layouts. Gisaburô Sugii likened Tezuka’s rough animation to layouts, which probably applies to the work of episode directors as well [76]. But, given what Tezuka’s drawings actually looked like, this is both an approximation and an anachronism. In Tôei, layouts were made by the background staff and revised by animators [77]. It seems that, in Mushi, layouts were entirely in the hands of animators.

We can take a look at a document included among the Atom DVD bonus materials. Described as a “background blueprint” from #34, it is attributed to animator Jirô Tsunoda. This is, without any possible mistake, a layout: both background and cels are drawn together and the camera movement is planned out. Because this is the only such document I could find and from an outsourced episode, it is hard to draw definite conclusions. However, the existence of layouts drawn by animators is further supported by one testimony, that of background artist Hiroya Hachimura. According to him, the episode director and chief of the backgrounds division first discussed the storyboard, providing general indications for the background artists. Then, individual animators sent their “genga” with sketches of what each background was supposed to look like [78]. While the genga terminology doesn’t match (the term is usually used to refer to key frames), the description perfectly fits the “background blueprint” I just discussed.

The questions I just raised about Atom‘s pipeline are relevant to my current inquiry, which is to reconstitute how work was organized in Mushi. But they were as well for Mushi’s staff: character designs had to be unified, key animation had to be cleaned up, and animation, backgrounds and photography had to be coordinated. These were necessities for the industrial production process that was being put in place, and have to be assumed in order to make sense of the highly standardized object that is Atom. Thankfully, Itô and Nagao’s research on Atom uncovered a document from much later in the production (#179) that shows how Mushi answered them. An “organization chart” drafted by producer Tatsuo Ikeuchi, it is a modelization of the staff division for an episode. I will anticipate a bit on the chronology and proceed to analyze it here.

At first glance, this chart gives the impression of an incredibly complex, almost bureaucratic organization, in which tasks are divided to the extreme. But this document is just that – a document, formalizing the production but not giving a direct picture of how things were on the ground. Moreover, what it modelizes is not the pipeline itself, but the people in charge for each section. If I’m not misreading it, it seems that section chiefs are shown on the right, while on the left and center are shown the tasks and people actually involved in production.

With episodes being produced in around one month, things must have felt far more confused than in this chart. In fact, we can note that, in spite of the extreme breakdown of functions, most of these are in the hands of the same people: for instance, the entire “production” (seisaku) side is represented by only two individuals, Kurokawa and Kishimoto. Some others seem to be standing in two hierarchical levels at once, such as Ochiai, both a cel painter and the chief of the ink & paint division, or Sakura, both technical director and chief of the photography division. In other words, while this chart makes Mushi’s overall division into departments clear, the actual functions and tasks must have been far more confused in reality.

With this being said, let us turn to the three questions I raised earlier. The character design seems to have been a particularly complicated affair, as it is apparently divided between the settei office, the “materials” (shiryo) and art divisions – note how the function of “art director” appears notably different from what it means today, having apparently no relationship to background art. However, it appears that it all emanated from the writing and storyboarding phases, only to go through a complex reviewing and revision process. If I were to oversimplify it and put it linearly, I would say it probably happened as follows: written descriptions were included in the script, put to drawings in the storyboard, made into proper blueprints by either the art director or Tezuka, and then circulated to the animation director and production staff.

Second key animation is clearly indicated in this chart, providing one more proof of its existence – even though, by that point in the production, the credit itself had disappeared. Since no names are given for the animators, it’s hard to tell who drew the second key animation, but if I’m not mistaken, the chart seems to indicate it was done by the in-betweeners.  More interesting to me, however, is the role of the animation director in this: it appears they not only supervised the key animation, but also the second key animation, in-betweens, as well as all the ink & paint and background art work. They were therefore a sort of super-supervisor in charge of almost all the visual elements.

This, then, may provide an answer to the question of layouts. If the animation director was in direct contact with the background artists, it would make sense for them to supervise or even draw the layouts as well, since their function is precisely to coordinate animation and backgrounds. They would then be transmitted to the photography team alongside timing sheets via either the technical director (who may also have drawn or at least supervised the layouts) or the production staff. But this remains highly hypothetical as this chart indicates no “layout” task separate from any other step in the production. In any case, this document also illustrates the central place of the episode directors, who served as both storyboarders, possible character designers, and animation directors by the end of Atom’s production.

Although episode directors were at the core of the production process, the position of enshutsu wasn’t yet the essentially technical job it is today. It was largely conceived as creative, the natural extension of an animator’s work – all episode directors in Atom’s initial rotation were animators, and the two positions were virtually indistinguishable at first. There was therefore a strong animator-centrism in Mushi, best formulated by Yoshiyuki Tomino who reports that production staff self-deprecatingly used to say that “those who are not animators are not humans” and that the studio’s other, established directors told him to give up on the position early on. Less dramatically, Akihiro Kanayama also noted that “directors who couldn’t draw weren’t directors”. In another anecdote, Atsumi Tashiro said that, when he told that he wanted to work in sound during his initial recruitment interview, the answer he got was “we’re making drawings here” [79]. This was perhaps one of the most essential differences with Tôei where, from the early 60s onwards, a clear distinction was established between animators and directors, the latter of which were never recruited for their drawing skills. Animators becoming directors in Tôei was almost inconceivable, just as in Mushi, production assistant Tomino becoming a director was exceptional, and only justified by his unmistakable drawing ability. 

This lack of technical expertise could cause problems: for instance, Eiichi Yamamoto recounts in his autobiographical novel how traumatic the recording sessions for #01, which he attended in Tezuka’s place, were. He had no idea what was going on, was basically bullied by the recording director, and finally appealed to Tezuka who, busy on manga and just as helpless, told him to trust the decisions of specialists [80]. It’s impossible to know how much fiction is added for dramatic effect, but the fact is that Mushi’s staff was totally uninvolved in everything related to sound.

At first, recording and dubbing were under the ad hoc supervision of Fuji TV producer Takaharu Bessho. After a few episodes, he became too busy and let Aoi Studio mixer Mitsuru Kashiwabara and Mushi production assistant Atsumi Tashiro take on the job [81]. Sound production was entirely outsourced to Aoi Studio and its own recording director, Matsuo Ono. Ono was a pioneering artist who, taking inspiration from musique concrète, created Atom’s unique soundscape. It contributed to the show’s sense of realism and SF-centered approach, but was also futuristic in some ways: the sound of Atom’s footsteps was the first use of electronic sound effects in anime, and perhaps Japanese TV history [82]. Voice recording was a particularly challenging task as well: tapes weren’t in use yet, meaning that, every time there was a mistake, the actors had to restart everything from the beginning. According to Atom’s actress Mari Shimizu, the recording of a 20-minutes episode could take up to 8 hours [83].

As Tashiro’s case illustrates, Atom’s technical lifeline was maintained by the production staff, divided as we saw above in various functions, from materials to animation and sound. Their first and most essential task consisted of managing all of the studio’s cels, sorting and numbering them so that they could be reused through the bank system. Not only did this avoid the studio becoming a complete mess, it ensured that episodes would remain on schedule and budget. Yoshihiro Nozaki, a production assistant on Atom, also explains that his tasks including taking care of “manga materials”: Tezuka seemingly didn’t keep copies of his own work at home, meaning that the production staff had to scour all bookstores in Tokyo to buy and store manga in the studio, where they could serve as reference for the staff [84]. Others, like Atsushi Tomioka or Keijirô Kurokawa, also mention how they served as Tezuka’s managers of sorts, sometimes driving him here and there or buying time with his editors while he did animation [85].

Aesthetically, it is during the second phase that Atom’s style was established. The approach to animation started changing: whereas Tezuka tended to use many drawings, modulating them and compensating with stills to save on the number of frames, Mushi’s other directors favored a more distinctly pose-to-pose approach, in which many extreme poses succeeded each other without transitions. They would also more systematically use deformation, going further than the squash-and-stretch appreciated by Tezuka and making smears more frequent. In the end, both philosophies coexisted, if only because bank sequences and cycles kept being animated on fluid 2s. In spite of that, Mushi’s initial members had difficulty adapting to Atom’s style: Sugii notably said that, until the completion of #01, he felt extremely uncomfortable making animation that moved so little, a feeling probably shared by most artists in the studio [86]. Indeed, during Mushi’s first year of existence, the huge majority of the staff came from Tôei. Contradicting the company’s image as a “mangaka’s studio”, Sugii stated that, at first, Rintarô and him were the only ones in the animation division to have had any experience drawing manga [87]. It wasn’t until the recruiting campaign of September 1963, when people like Osamu Dezaki or Masaki Mori joined, that the balance started to shift significantly.

Examples of smears in Atom

If the animation started to change, in terms of storyboarding, the experimental mindset initiated by Tezuka’s early episodes was shared by the team of directors. As I have already mentioned and shown, Atom is visually incredibly diverse, making full use of the possibilities of animation and the black-and-white format. This was partly allowed, according to Rintarô, by the fact that Atom was an adaptation and that at first, episode directors didn’t have to worry about the script: the stories and themes were ready-made, so directors could fully concentrate on visuals [88]. On the other hand, it must be noted how, in conformity with Tezuka’s original choice to focus things on the story, Atom’s episodes are also carried by their writing. The manga provided excellent source material which, even in less competent hands, would have made for great stories. Changes were naturally made in the course of the adaptation process: stories were compressed and made tighter. An excellent example would be #07, Yamamoto’s rewriting of the initial Atom Taishi story, which maintains all of the shock and intensity of the original while removing all side storylines and non-essential characters. The spirit of Tezuka’s manga remained: a focus on elaborate SF settings, dramatic storylines, ecological and social themes, and an abundance of visual gags.

The third phase: Mushi’s expansion and the new pipeline (#48-#97; 11/1963-12/1964)

Between August and December 1963, a series of changes in Mushi Pro caused Atom’s production to evolve as well. First, the studio went through exponential growth: staff from other studios and soon Tezuka’s former assistants kept pouring in, but that was not enough. So, from the summer of 1963 onwards, Mushi started putting out public recruitment campaigns in the newspaper. In September 1963, 40 new employees joined, and the studio numbered more than 80 members [89]. A third floor was added to the main studio, and two new “branches” were added: the “first branch”, nicknamed “the Lalamy Ranch” (from a popular drama which ran on NHK at the time), was located a 5-minutes walk away from the studio. At its peak, it housed 30 people, among which 21 animators. The “second branch”, similarly a rented apartment, was even closer to Mushi and would become Mushi Pro Studio 3 in late 1964 [90]. As yet another wave of recruits arrived in January 1964, plans for a second building across the street from the main studio, Studio 2, began being drafted; it would start operations in August 1964 [91].

With all this new staff, Mushi started launching new projects. Atom, initially planned for only 26 episodes, was extended for a second year and Kaoru Anami drafted a new planning document for it, which, among other things, presumably renegotiated the show’s budget [92]. Moreover, a contract was concluded with Fuji TV for a new series, titled Mushi Pro Land: it would consist of 1-hour long episodes aired once every two weeks. Under Yûsaku Sakamoto’s direction, the project included Shin Takarajima (not an adaptation of Tezuka’s manga of the same name, but a SF take on Stevenson’s novel), Zero Man, Ribbon no Kishi and Hi no Tori. The name was taken from Disney Land, a live-action program by the Burbank studio which had started airing in Japan on NTV in 1958 [93]. However, only the first episode was produced and the program was canceled in April 1964. Due to Mushi Pro Land’s production, the first major reshuffle in Atom’s staff happened, as Eiichi Yamamoto took Sakamoto’s place as the series’ chief director [94].

Settei for Mushi Pro Land by Tezuka (?). Top is Ribbon no Kishi, bottom is Zero Man [Tezuka 2018]

This wasn’t the only change. By late 1963, Mushi had enough animators that episode directors no longer needed to do the key animation for their episodes themselves. Their role evolved towards supervision, and they started to occupy the function of modern animation directors [95]. Then, a few months later, the task of animation supervision was formally separated from that of episode direction: according to Itô and Nagao’s research, between episode 61 to 93, the “key animation” (genga) credit must be understood as meaning “animation direction” [96]. They sadly do not explain how they reached this conclusion, why the sakuga kantoku terminology didn’t appear yet, and why the role was given back to episode directors later on. Moreover, the question remains whether this new task naturally grew out of Mushi’s pipeline – which might explain the lack of change in credits – or was borrowed from Tôei, which had invented it.

Indeed, in early 1964, animation supervision was a novelty. The title had officially appeared in credits for the first time on Tôei’s Wanpaku Oji to Orochi Taiji, released in March 1963. Most of the ex-Tôei animators in Mushi hadn’t worked on the film and had therefore never experienced their drawings being corrected by a higher-up. The same applied to ex-mangaka, who had an even more individualistic mindset. And, unlike in Tôei’s rationalized system, this seems to have caused friction in the animator-centered Mushi: Yamamoto reports a violent altercation between animator Shinji Nagashima and episode director Rintarô, in which the former judged that the latter’s corrections made his drawings worse than they had originally been [97]. Whether this really happened like Yamamoto tells it, with these two specifically, matters little in this case: the anecdote itself is credible and tells volumes about the atmosphere in the studio.

Finally, the last major change that happened to Atom’s production was the arrival of scriptwriters. When the show started airing, around 56 chapters were available to adapt [98]. This posed no issue as long as the show was supposed to last 26 or even 52 episodes, but when it was extended for a second year, it became necessary to write original stories. Around November 1963, a separate literature department was created, first occupied by members of the directors team [99]. Sometime in early 1964, former production assistant Arashi Ishizu was assigned as its chief. Tezuka seems to have been worried that Mushi’s in-house members did not have enough sense for SF, and that Atom’s stories would lose in realism. He therefore contacted one of his acquaintances, SF novelist Aritsune Toyota, who had written anime’s first original episodes on Eightman, and offered him to work for Mushi [100]. Toyota then invited another of Eightman’s writers, Masaki Tsuji, who knew Tezuka well since he had planned and written a joint manga-TV drama project between him and NHK in 1961 [101]. Episode directors kept writing scripts as well, but Ishizu, Toyota and Tsuji became the core members of Mushi’s literature department – although it seems that Ishizu was the only in-house member of the team.

Arashi Ishizu in front of Mushi Pro studio 1 [Mineshima 2012, n°3]

According to Toyota, the writing process was quite complex and broken down into 4 phases: working out the plot; writing a synopsis; creating a first draft; and then the final script [102]. At least two of these phases, the synopsis and the script, were supervised by Tezuka. According to Tsuji, the mangaka initially left a lot of freedom to writers over the initial plot ideas. But when actual synopses or scripts arrived to him, he was unforgiving: he would often reject them 3 or 4 times, constantly asking for revisions. Needless to say, this did not have a positive impact on schedules, especially when Tezuka gave most of his commentary later than he was supposed to [103]. As the ones in charge of the first step in the production process, it seems that writers were regularly pressured by production staff, who were themselves helpless when it came to Tezuka.

The writing process may have been difficult, but the results weren’t disappointing. The most popular episode of Atom, #84, which obtained a rating of 40.7%, was an original one written by Toyota [104]. It deserved its success: not only does it feature some excellent animation, the writing is some of the best in the series. Generally, Toyota’s episodes did not betray his reputation as an SF author. They kept the futuristic and grounded dimension of Tezuka’s work, and built on the moral complexity of antagonists to make them fully-rounded characters – as is the case of the Dolphin people in #84. Paradoxically, the only one this made unhappy was Tezuka himself, extremely proud and easily vexed, who didn’t understand why episodes adapted from his manga didn’t work as well [105].

The fourth phase: the outsourcing rotation and the new generation of directors (#98-#193; 12/1964-12/1966)

As time went by, Mushi Pro kept expanding. By Fall 1964, the studio counted 230 employees [106]. All these people had to be paid, which meant the company also had to expand its activities. Moreover, with NBC refusing to buy a third batch of Atom episodes (having them buy a second year worth of broadcast had already been an uphill battle), it became necessary for Mushi to launch other projects. These included CM work, collaborations with hybrid live-action/animation adaptations of Tezuka’s manga, and, from January 1965, two new TV series, Jungle Taitei and W3. Under Yamamoto, most of Atom’s original team – notably the directors like Rintarô, Shinji Nagashima or Chikao Katsui – moved on to studio 2 to work on Jungle Taitei, while the remaining staff was in studio 3 for W3 [107].  Atom‘s new chief director seems to have become literature department chief Arashi Ishizu. When he left Mushi in mid-1965, it seems that he was replaced by Tezuka – who kept checking scripts and storyboards as he always had – and Yoshiyuki Tomino [108].

As a result of these changes, Atom’s studio 1 lost most of its staff, and the production could only continue thanks to outsourcing. This wasn’t the first time Mushi resorted to this: back in June 1963, Tezuka had asked mangaka friends of his gathered in Studio Zero to produce an episode of Atom in order to give Mushi’s staff a well-deserved holiday [109]. As is well-known, the results were disastrous: Zero only had one experienced animator, former Otogi Pro Shin’ichi Suzuki, and while #34 certainly moved, the characters were consistently off-model. Tezuka was appalled by Zero’s work and resolved never to work with them again. By late 1964, however, the situation was different, and Mushi had to rely on the help of other studios. At first, there were only two: Art Fresh and P Production.

Art Fresh was basically an extension of Mushi, but its origins were somewhat complicated. In August 1964, rival studio Tokyo Movie started its first TV series, Big X. It so happened that a good friend of Tokyo Movie president Yutaka Fujioka was the father of Mushi animator Atsushi Takagi. Tokyo Movie had basically no experienced animators, so Takagi’s father urged his son to work on the side on Big X. Unable to refuse, Takagi and a few friends started moonlighting for Tokyo Movie. When they learnt about it, Yûsaku Sakamoto and Eiichi Yamamoto convoked Gisaburo Sugii, who had invited Takagi to Mushi in the first place and told him their intent to fire Takagi. Sugii asked them to wait, heard the circumstances from his friend, and decided to take responsibility with him: they gathered the Big X team, left Mushi and moved to Takagi’s house under the name Art Fresh. A few months later, financed by Takagi’s father, they moved to a bigger location [110]. Probably feeling betrayed, Sakamoto and Yamamoto refused to work with Art Fresh, but Tezuka, very close to Sugii, bypassed them and decided to have Art Fresh take charge of Atom’s new batch of outsourced episodes [111].

Mushi’s other main subcontractor was P Production, which had just finished work on their first TV series, Zero Sen Hayato, in November 1964. The studio’s creator and leader was Sôji Ushio, a veteran active since before the war, and a good friend of Tezuka’s. At first, P Pro was only supposed to work on Atom for 1 cour, with the following rotation in mind: one episode would be in-house Mushi, one episode would be Art Fresh, and one episode P Pro – and then repeat. However, the contract was extended and P Pro ended up staying for 3 cours [112]. By that point, the rotation had expanded: other studios by ex-Mushi employees were brought on, such as Masami Hata’s Studio Fantasia or Moribi Murano’s Anaguma Pro, but also completely separate companies like Hatena Pro, Wako Pro, Onishi Pro…

Comparing various testimonies, it seems that the production system varied according to each subcontractor. P Pro only took charge of the animation based on storyboards made in Mushi, and the ink & paint, photography and editing were also done in-house. It seems that there was no more animation director system by this point: retakes were decided (presumably by the episode director) after the completion of the rush film, and, on top of corrections, P Pro received a fine for the inadequate sequences [113]. On the other hand, Noboru Ishiguro’s Onishi Pro seems to have done the ink & paint [114]. Finally, Seiji Okuda said that Art Fresh took charge of every step until the end of the animation, had its own animation directors (which were, in this case, the episode directors), and that there weren’t even production assistants [115].

The beginning of outsourcing is inseparable from another major development in Atom’s production: the arrival of a new team of episode directors, promoted to replace the initial group who had left for Jungle Taitei. Its two representative members, with completely different profiles and approaches, would become some of anime history’s greatest directors: Yoshiyuki Tomino and Osamu Dezaki. Besides the fact that they debuted at roughly the same time, what united them was the fact that the episodes they directed were very different from the rest of Atom, and already prefigured the visual and narrative motifs associated with their later careers.

Yoshiyuki Tomino (left) and Osamu Dezaki (right) in the mid-60s [Tomino 2019; Takarajima Editors Room 2018]

Tomino entered Mushi Pro in February 1964 as a production assistant [116]. He was an anomaly in the studio, perhaps its only member at that point who had graduated from film school. The first episode he was in charge of was #65, and from there he progressively went up the ladder: he became assistant episode director on #76, and debuted as both scriptwriter and episode director on #96 under a pseudonym. It seems that he had been tasked with the script as assistant director, but ended up writing the storyboard when Tezuka noticed his talent for drawing [117]. By the end of the production, he seems to have become the chief of Atom’s directors team, and possibly the unofficial chief director. His official debut was #104, P Pro’s first episode. As Tomino couldn’t directly consult with the other staff members who weren’t on-site, most of his supervision work was done through written notes transmitted by production assistants.

The episodes made by Atom’s initial team of directors were remarkable for their constant visual experimentation and use of flashy techniques of framing, lighting and composition. In comparison, Tomino’s episodes feel very subdued, with very little use of the aforementioned tricks. Coming from a live-action background, Tomino instead approached things in terms of staging, with the same attention to space that would characterize all of his later work. So, for instance, he would repeatedly use in-depth compositions, in which two or three characters are all placed at different distances from the “camera”, and possibly on different cels. While such shots were already common in earlier episodes, Tomino made them his trademark, as they allowed him to make longer shots and avoid the constant cutting characteristic of Atom and early anime editing, especially during dialogue scenes.

Trademark Tomino shots from #104, #111 and #121

Tomino didn’t always write his own episodes, but even those in the hands of other writers bear his mark – writers such as Aritsune Toyota probably shared Tomino’s taste for SF and Hollywood cinema. One of their main traits is how they tend to lessen Atom’s role, or make his powers relatively inefficient. For example, episode 104 tells how Atom becomes hated by the entire population because of a fake double abducting children; or, in episode 121, he is almost entirely passive and the focus is actually on the guest character. The same episode 121, written by Toyota, is pure SF adventure and feels like the kind of story one would read in a 1950s US pulp magazine. A scene of planetary exploration and encounters with strange aliens feels directly borrowed from Forbidden Planet.

Tomino was most probably not very involved in the animation for his episodes, both because most were outsourced and because he personally never had any taste for doing animation himself. But those animated by P Pro deserve scrutiny. #104, while not as awful as Zero’s #34, certainly stands out: most characters are off-model, and the mob characters are worlds away from Tezuka’s style. In fact, the A part of that episode features some of the most interesting animation in the entire show, because it rests on totally different principles than the ones used in Mushi. Instead of being stiff but expressive, the movement is jerky but charming, tends to make bodies more elastic and dynamic through wide arcs . Perhaps this didn’t please either Tomino or Tezuka, as such animation quickly disappeared and P Pro’s episodes became less idiosyncratic.

Osamu Dezaki joined Mushi in September 1963. He had spent most of his time in middle and high-school writing manga and, in 1959, won a newcomer award for his first publication in the magazine Machi [118]. Dezaki’s manga career was very brief, but his work was popular enough that many people, including Akio Sugino and Seiji Okuda, knew him by name when they first met him [119]. However, that didn’t guarantee his recruitment: when Yûsaku Sakamoto interviewed him, he intended to reject him, but Sugii vouched for him and managed to get him a position as in-betweener [120]. At that point, Dezaki didn’t know anything about animation, but he had experience drawing flipbooks and quickly rose up the ranks: his first in-betweens were on #39, he moved on to second key animation on #51, key animation on #66, and finally debuted as episode director on #112. But this wasn’t his true debut: his mentor Atsushi Takagi had brought him to moonlight together on Big X, which is where he did his first storyboards. When Takagi left Mushi for Art Fresh, Dezaki naturally accompanied him [121].

Dezaki’s first episode on Atom, #112, is already the work of a mature artist. According to Seiji Okuda, Dezaki served as both episode and animation director, on top of doing most of the key animation himself – the other members of Art Fresh credited as key animators actually did the second key animation and in-betweens [122]. It seems Dezaki did animation direction himself on most of his episodes, which explains the extremely high level of the drawings. While his style was always close to Tezuka’s, his choice of camera angles and his dynamic, natural posing made his animation extremely energetic, and sometimes seems to prefigure Akio Sugino and Moribi Murano’s work on Sabu to Ichi in the way it stretches out bodies.

Some of the excellent posing and perspectives from #112

Where Dezaki truly learnt from Tezuka, however, was in conceiving animation as a total work of art: he explored each of its possibilities beyond just character movement. He regularly let the background art do most of the job, and already experimented with the possibilities of the multiplane camera and filter work. Moreover, he used sound like no other director on the show’s rotation. For instance, the opening scene of #112 is a masterpiece of sound design, where the naturalistic chirping of the birds is cut by more SF-like sound effects, all of which perfectly complement the long pan to create both a quiet and mysterious atmosphere. Similarly, during the climax of the episode, Atom’s triumphant theme starts blaring out at each attack, only to be suddenly interrupted when these attacks prove ineffective. For Dezaki, sound wasn’t just a complement to the visuals, but an essential cinematographic tool to create rhythm and support the narrative.

Although he did not write them himself, Dezaki’s episodes are among the darkest and most dramatic of the show – to the point that Tezuka himself told him to make things lighter during storyboard check [123]. Just like Tomino – although their styles are very different – Dezaki put few gags in his episodes and tended to shift the focus away from Atom. In #112, the true main characters are actually Pr. Ochanomizu and the guest Samson, a typical Dezaki tragic hero.

As creative as the last stretch of Atom could be, in this same period, the show’s ratings started going down and, with them, Mushi’s investment in it. It also seems that in late 1965, sponsor Meiji threatened to drop from the show [124]. I can’t tell whose decision it was but, from episode 120 onwards, the studio started radicalizing its use of the bank system on Atom. At first, only cels had been kept and reused; then, it became character designs, especially for guests and crowd characters. Now, it was entire episodes. For example, #120 reuses footage from #11 and #68 with some new animation, while #138 does the same with bits from episodes #19, #38, #65, #94, #100, #103, #110, #119 and #122. Some others, like #142 or #181, would be remakes of earlier episodes (in this case, #10 and #8) with entirely new animation. Some episodes would also be rerun as is. What may seem like a cynical form of recycling made perfect sense: considering TV viewership at the time, there was no way that the viewers of the early episodes were still there 4 years later, or, had they been, that they’d notice the reused footage and stories. Moreover, it was a perfect way to save time and money, especially if the recycled episodes were paid the same rate as the properly-produced ones.

An interesting aspect in this development is that the majority of these “recycled” episodes were directed by the same person, Yoshiyuki Tomino. I don’t know whether the recycling was his idea in the first place or not. But he certainly was the man for the job: as a former production assistant and an in-house employee (unlike Dezaki, for instance), he probably had the best understanding of what resources Mushi had at its disposal and how to use them.

This didn’t mean either that Mushi entirely gave up on Atom. In the last year, three of the members of its original rotation came back for one last episode: Gisaburô Sugii on #148, a poetic SF retelling of Romeo and Juliet; Yûsaku Sakamoto on #189 with a meta episode in which Atom and Uran play their own roles in a film studio; and Tezuka for the finale, #193, full of references to earlier episodes and in which Atom sacrifices himself to save the Earth.

Conclusion: Mushi Pro and animation studios in Japan

When Tetsuwan Atom’s run ended on December 31st, 1966, Mushi Pro had become totally different from what it had been 4 years earlier. Initially a small group working in Tezuka’s house only meant to be a small “bug” in the animation industry, it now had hundreds of employees spread over 5 studios and multiple production lines, alongside an already complex network of partners and outsourcing companies. This was a radical change, just as significant, if not more, than Tôei Dôga’s installation in a brand-new building in Oizumi in 1956: both heralded the entry of Japanese animation, and especially of animation studios, in a new era.

Mushi, or rather Tezuka Animation, was initially anything but exceptional in the landscape of Japanese animation, which had changed little since the war. It was made up of small, grassroot companies counting at best a dozen members, working alternatively on indie projects or CM works. CM production had initiated an animation “boom”, but not fundamentally changed the structure of the industry. Similarly, the creation of Tôei had been a major change, but ultimately had no wide-ranging effects – studios like Iwao Ashida’s Cartoon Films Institute or Ryûichi Yokoyama’s Otogi Pro just kept going as they always had. In 1962, it may have seemed like Tezuka’s studio would be the same, just one more small company, with its only distinguishing feature being that its leader was a more famous mangaka than the others.

Tetsuwan Atom triggered a sudden shift, but it did not do it out of nowhere. As has often been pointed out, it is not even really the first animated TV series produced in Japan – Otogi Pro’s Instant History already held that title. Moreover, animation studios’ work regularly aired on TV through commercials, and US animation was frequently featured on Japanese airwaves. Rather than Atom as a singular work, the real change was rather all the planning and expectations that surrounded it.

What was unprecedented about Atom was the format – 30-minutes episodes. This allowed Tezuka and Mushi’s staff to present it as “story anime”, but more fundamentally, it aligned TV animation with live-action TV series in terms of length, budget, production and, soon, presence on airwaves. I say “production” because, even though the pipelines for live-action and animation are naturally very different, both had to be made industrially, requiring major organizational changes in Mushi and, soon, every other animation studio. It is at this point that the gap between “commercial” and “independent” production really appeared and suddenly widened, as all animation creators in the country had to make a choice: bank on the opportunities opened by Mushi, or keep doing as they had always done.

Moreover, Mushi seemingly succeeded at realizing what had been the dream of all animation creators and producers in Japan since virtually the birth of the animation industry: self-sufficiency. As I discussed elsewhere, a first attempt had been tried in the early 1950s by studio Nichiman, but failed; Tôei Dôga, entirely funded by its live-action counterpart, bypassed the issue. Mushi was confronted with it once again and, by adapting Tezuka’s works and profiting from its IPs, seemed like it had solved it.

Things were certainly not ideal. Mushi made profits, but these were extremely fragile – as we will see in later articles, the business model so carefully built by Kaoru Anami slowly crumbled once he was no longer there to oversee it. Moreover, it only worked insofar as Mushi never stopped producing, and producing more. Very soon, the studio found itself facing a vicious cycle out of which most anime studios today haven’t extracted themselves from: meeting the demands imposed by the increase in productions meant recruiting more people; but recruiting more people meant making more productions to pay them.

None of this, naturally, could have been foreseen by Tezuka and his colleagues. It would be vain to hold them responsible for the structural issues anime still faces today. But it remains vital to understand how these issues appeared, why the decisions that led to them were taken, and how simple artistic ambitions led Japan to become both an animation powerhouse and the source of so many labor inequalities. It is also vital to never forget the work of those who took these decisions or were subjected to their consequences – works, like Tetsuwan Atom, of incredible artistic diversity and fecundity.


[1] Tsugata 2007, pp.46-47

[2] Nakagawa 2020, p.43

[3] Shirakawa 2004; Tsugata 2007, p.56; Nakagawa 2020, p.74. Nakagawa notes that Tezuka moved in his new house in August 1960, which perfectly coincides with Saiyûki’s completion

[4] Cited in Tsugata 2007, p. 59. Otsuka, 2001, p.71 describes Tezuka’s work as “storyboards (ekonte) without any dialogue” looking “just like manga” – he seems to have been a bit confused by this process which wasn’t in use in Tôei

[5] Shirakawa 2004; Tsukioka 2013, p.323

[6] Shirakawa 2004, Nakagawa 2020, p.67. Nakagawa in particular attributes Tezuka’s idea to wanting to create “the first animated film where a main character dies at the end”. This would be consistent with Tezuka’s character, but I think the obsession with “firsts” comes more from Japanese historians than Tezuka himself. The fact is that Rin Rin’s sudden recovery at the end of the film does feel artificial

[7] Shirakawa 2004; Tsukioka 2013, p.323

[8] Cited in Nakagawa 2020, p.70

[9] Otsuka 2001, p.73; Shirakawa 2004

[10] Shirakawa 2004

[11] Tezuka 1996, vol.1, p.48; Nakagawa 2020, p.29; 

[12] Tsugata 2007, p.43

[13] Nakagawa 2020, p.37

[14] Tsugata 2007, p.44

[15] Shirakawa 2004; the date is given by Nakagawa 2020, p.74

[16] Tsugata 2007, p.82. Tezuka himself claims Kuri’s words were said at the third screening, in 1963, but Tsugata rightly points out that, by that time, they would have made no sense, and suggests this instead happened one year prior – if the entire anecdote is not apocryphal, at least

[17] Yamamoto 1989, p.9

[18] Yamamoto 1989, p.36; Nakagawa 2020, p.84. Regarding the multiplane camera, according to Yôji Kuri (Tezuka & Kuri 2013), one unit cost around 200,000 yen at the time, making it a significant investment  

[19] Yamamoto 1989, p.36; Tsugata, 2007, p.92

[20] Yamamoto 1989, p.40; Nakagawa 2020, p.87

[21] Yamamoto 1989, p.46

[22] Nakagawa 2020, p.83

[23] Otsuka 2001, p.84; Sugii 2015, p.49; Rintarô 2009, pp.30-31. Sugii claims that Otsuka was at the origin of the union, but Otsuka himself says nothing of it; he claims the union was created in June, but this goes against every other source, including Seiji Kanô’s research in The People Who Built Japanese Animation. According to Sugii, the lockout only lasted one day, whereas for Rintarô, it was “3 or 4 days long” and Otsuka says nothing of the length. Moreover, Rintarô says that the blacklisted members, among which he figured (possibly not for political reasons but simply because he had the reputation of a bad worker), had actually been designated by the union itself, as a precondition set by Tôei’s management for any negotiations

[24] Sugii 2015 p.48

[25] Sugii 2015 p.69. I couldn’t confirm whether Sugii was part of the union, but the fact that they supported him after his departure inclines me to think that he was

[26] Sugii 2015, pp.50-52; in Rintarô 2009, p.29, Rintarô himself claims that the nickname was coined by Tsukioka, but in the same book, p.168, Sugii insists that it’s him. In any case, it’s a pun on Shigeyuki Hayashi’s surname, “hayashi” being read “rin” and the name “tarô” being added because, according to Sugii, “calling him Rin-chan made it sound like he was Chinese”

[27] Cited in Itô & Nagao 2001, vol. 2, p.44

[28] Miyamoto 2009, p.291; Sugii 2015, p.167

[29] Itô & Nagao 2001, vol.1, p.4

[30] Cited in Itô & Nagao 2001, vol. 2, p.45

[31] Yamamoto 1989, pp.64-65; Nakagawa 2020, p.93

[32] Cited in Tsugata 2007, p.93

[33] Yamamoto 1989, p.64

[34] Cited in Tsugata 2007, p.72

[35] Nakagawa 2020, pp.103-104. Anami was originally an acquaintance of Daisaku Shirakawa, who even offered him and Tôei animator Daikichirô Kusube to join Mannensha and create an animation division there. They refused, but it’s through them that he met Kazuko Nakamura. In Nakamura 2007, p.284, Nakamura says she married Anami before entering Mushi, but this doesn’t explain why she is credited under “Nakamura” on Machikado and under “Anami” on Atom.

[36] Tsugata 2007, p.97

[37] Itô & Nagao 2001, vol.1, p.37. Yamamoto 1989, p.74 gives 38 people “excluding Tezuka and Imai”, but that’s still one person short – perhaps he excluded himself as well

[38] Itô & Nagao 2001, vol.1, p.6

[39] See Tezuka 1996, vol.1, p.135 or the citation from Boku ha Mangaka in Tsugata 2007, p.121

[40] Shirakawa 2004; Nakagawa 2020, p.110

[41] Nagakawa 2020, p.108 gives an amount of “around 300,000 yen”; Tsugata 2007, p.123, gives precisely 310,000

[42] For reference, in his novel, Yamamoto gives 750,000 yen; but he is the only one to do so against every other source, including Itô & Nagao, who directly consulted Mushi’s archives. This mistake on such an important point is one of the reasons why I consider it better to treat Yamamoto with extreme caution

[43] Itô & Nagao 2001, vol.1, p.40

[44] Tsugata 2007, p.126

[45] Yamamoto 1989, p.94; Tsugata 2007, p.123; Masuda 2009; Nagakawa 2020, p.108 

[46] Tsugata 2007, p.128. 

[47] In Hoshi 2018, p.66. Zero Sen Hayato is said to have had a ridiculously low budget and Ushio describes the 3 million number as “lower than Astro Boy” – to be honest, I can only think that Ushio is wrong here, but the fact that 3 million seems like a low budget to him is itself significant

[48] Tsugata 2007, p.129. Interviewed by Tsugata, Yamamoto mentions that the budget he had at his disposal for Jungle Taitei was 2,5 million per episode, but that “the actual figure must have been much higher” because of management costs

[49] Tsugata 2007, p.129; Itô & Nagao 2001, vol.1, p.40. It may be based on this reevaluation that Yamamoto gives the 750,000 yen figure – which lessens his mistake, but does not make it less wrong

[50] Tsugata 2007, pp.124-125

[51] Nakagawa 2020, p.121

[52] Nakagawa 2020, p.111

[53] Toyota 2020, p.51 gives 18,000, while Sudô, in Nakagawa 2021, p.47, gives 22,000

[54] Sugii 2015, p.51; Rintarô 2009, p.29

[55] The 12,000 yen figure is given by Rintarô 2009, p.29, Miyazaki’s salary by Tsugata 2007, p.131

[56] Yamamoto 1989, p.33; Tsugata 2007, p.132; Rintarô 2009, p.32; Kanayama 2006; Udagawa 2005, p.275

[57] Sudô, in Nakagawa 2021, p.47

[58] Tsugata 2007, p.132; Toyota 2020, p.51

[59] Tomino 2002, p.90; Masami 2008 

[60] Itô & Nagao 2001, p.8

[61] Nakagawa 2021, p.113

[62] Cited in Tsugata 2007, p.104

[63] Yamamoto 1989, pp.74-75; Cinema Novecento 2020, p.17

[64] Yamamoto 1989, p.82; Cinema Novecento 2020, p.17

[65] Yamamoto 1989, p.88; Itô & Nagao 2001, vol.3, p.44

[66] Itô & Nagao 2001, vol. 1, p.7

[67] Yamamoto 1989, p.121; Itô & Nagao 2001, vol.1, p.8 & 46; Nakagawa 2020, p.115

[68]  Itô & Nagao 2001, vol.1, p.46; Sugii 2015, p.90

[69] See notably Itô & Nagao 2001 and Tezuka 1999a & 2018

[70] Toyota 2014, p.227 & 2020, p.80

[71] Tezuka 1999, p.628

[72] Itô & Nagao 2001, vol.1, p.15

[73] Yamamoto 1989, p.17

[74] Yamamoto 1989, p.122

[75] Sugii 2015, p.90

[76] Cited in Itô & Nagao 2001, vol.3, p.44

[77] Shirakawa 2004

[78] Hachimura 2016, p.212

[79] Tomino 2002, p.60; Takarajima Editors Room 2018, p.77; in Shimotsuki & Tsuda 2003, pp. 111 & 114

[80] Yamamoto 1989, pp.98-101; Shimotsuki & Tsuda 2003, p.112

[81] Shimotsuki & Tsuda 2003, p.114; Cinema Novecento 2020, pp.104-105

[82] Shimotsuki & Tsuda 2003, p.114;  Cinema Novecento 2020, p.103

[83] in Shimotsuki & Tsuda 2003, p.115

[84] Nozaki 2011, p.271; Nakagawa 2021, p.53

[85] Tomioka 2015, p.242 & Kurokawa 2018, p.291

[86] Sugii 2015, p.79

[87] In Rintarô 2009, p.170

[88] Rintarô 2009, p.37

[89] Itô & Nagao 2001, vol.2, pp.9-10

[90] Itô & Nagao 2001, vol.2, p.11

[91] Itô & Nagao 2001, vol.3, p.8; Nakagawa 2021, p.122

[92] Itô & Nagao 2001, vol.2, p.8

[93] Itô & Nagao 2001, vol.2, p31; Nakagawa 2021, p.122

[94] Yamamoto 1989, p.123. I will discuss this in more detail in the next article, but Yamamoto’s version of the facts is very strange, as it strays off the chronology established by Itô & Nagao directly from Mushi’s documents; on the other hand, they confirm Yamamoto’s role as chief director from episode 61 onwards.

[95] Yamamoto 1989, p.135

[96] Itô & Nagao 2001, vol.2, p.5

[97] Yamamoto 1989, p.135

[98] Nakagawa 2020, p.159; Toyota 2020, p.54

[99] Nakagawa 2020, p.160

[100] Tsuji 2006; Mineshima 2012, 3; Toyota 2014, p.226 & 2020, p.53 

[101] Nakagawa 2020, p.79; Tsuji 2006. This project, called Fushigi na Shônen, was based on a simultaneous release of the TV show and manga; it was, in other words, Tezuka’s first real media-mix project

[102] Toyota 2000, p.96

[103] In Nakagawa 2021, p.18; Toyota 2020, pp.59-60

[104] Toyota 2014, p.227; Nakagawa 2020, p.160

[105] Toyota 2014, p.227

[106] Itô & Nagao 2001, vol.4, p.2

[107] Yamamoto 1989, p.170; Itô & Nagao 2001, vol.5, p.7

[108] Kurokawa 2018, p.291. Tomino’s exact function at this point is unclear: according to the organization chart discussed above, he served as chief of the directors’ division (enshutsu chief) in studio 1, a role which Itô and Nagao seem to identify as “chief director”. However, this contradicts multiple other testimonies on Atom, W3 and Jungle’s productions which separate the two functions. In Kurokawa’s case, he says that Tezuka acted as chief director (sou kantoku, a term not recorded in credits) of Atom in the last stage, but that the actual production was largely driven by Tomino and him (as producer-in-charge)

[109] Itô & Nagao 2001, vol. 2, p.12

[110] Sugii 2015, p.111; Takarajima Editors Room 2018, p.136; Okuda 2020, p.28. Yamamoto doesn’t narrate these events, but tells a similar story, with different dates and actors: in January 1964, Sugii told him about Mushi animators working for other studios. Yamamoto got angry and threatened to “beat up the bastards” but Sugii quieted him down – and the episode ends with Yamamoto’s melancholy meditation on the “weakness” of human beings who keep chasing after money (pp.126-127). Then, Sugii’s departure is attributed to his own impulse and said to have been done in good terms with Yamamoto [p.161]. Yamamoto obviously gives himself the good role here, so I think it’s fair to dismiss this anecdote, even if moonlighting might have started before the Big X case

[111] Sugii 2015, p.113

[112] Itô & Nagao 2001, vol.4, p.32

[113] Itô & Nagao 2001, vol.4, p.32. More precisely, Ushio says P Pro “wasn’t paid” for the inadequate parts. I suppose that a certain amount was subtracted from the fee Mushi paid to P Pro – but how was it calculated?

[114] In Shimotsuki and Tsuda 2003, p.113

[115] Okuda 2020, p.30

[116] Tomino 2019, p.25

[117] Shimotsuki & Tsuda 2003, p.111

[118] Takarajima Editors Room 2018, p.97; Okuda 2020, p.33

[119] Takarajima Editors Room 2018, p.121; Okuda 2020, p.33

[120] Dezaki himself claimed that Sugii had read his manga and this is the reason why he recommended him, in Oyama & Hayashi 2012, p.21 & Takarajima Editors Room 2018, p.98. However, Sugii himself denied this on at least two separate instances, in Sugii 2015, p.92 and in Takarajima Editors Room 2018, p.135. In the first of those interviews, Sugii claims that Dezaki brought his manga for the job interview, and that it is upon reading it on the spot that Sugii noticed his talent. Dezaki himself says nothing of bringing his manga to either the entrance exam or interview, but it’s possible that he brought a portfolio or equivalent

[121] Sugii 2015, p.111; Takarajima Editors Room 2018, p.99

[122] Okuda 2020, p.30

[123] Takarajima Editors Room 2018, p.100

[124] Kurokawa 2018, p.291

7 thoughts on “The History of Mushi Pro – 01 – The Road to TV Anime (1960-1965)

  1. Absolutely fascinating and thoroughly researched post. This draws in so many different little elements and origin stories. I love the bits about how there was no editing on the soundtracks, which I think definitely continues to dictate the hyper-efficient environment of seiyu dubbing today.

    Can’t wait to see more of this. One question I had that I didn’t see referenced in any of your other posts: When you say there was an “anti-Disney” faction at Toei, what were they objecting to exactly? The animation style? The production style? The storytelling? I would love to know a bit more about what those animators disliked about Toei’s direction.

    I’ve been slowly acquainting myself with animation history for a project covering theatrical films from all regions and this is now a cornerstone piece for my Japan knowledge. I’ll definitely check out some of your resources as well, thanks a million for the precise citations!


    1. Thanks for your kind comment!

      Regarding the “anti-Disney” thing, I believe it was mostly a style thing. It seems that the “Disney style” was associated with the realistic type of animation Tôei was going for at the time, and this is what the younger generation of animators was rebelling against. In following years (see The Little Prince and the Eight-Headed Dragon or Wan Wan Chûshingura for instance), they’d move towards something far more “modern”, especially in terms of design, and the animation style would become more varied and diverse.
      But fundamentally, as I discuss here, I think it was a generational and political issue: Tôei’s younger generation had other references (notably Soviet and French animation, and possibly non-Disney American works) and they wanted to integrate that into their works, whereas some of the older artists and the management remained close to a Disney-centric framework.


      1. Quite interesting! I had heard about Takahata and Miyazaki being part of the union activities, but I didn’t know it was tied so strongly to the collective influences of the animators.

        Would love to get your take on the different stylistic eras at some of these studios, as I’ve been attempting to compartmentalize some of that myself for my project. Any good way to get in touch?


  2. It’s interesting seeing your underlining of Tomino’s bank technique. The majority of other anime directors in general have had a much more formalized, non-radical approach to asset/footage reuse compared to Tomino. You can contrast Tomino with his contemporary – Nagahama, who seems to have laid the basic foundations for the transformation/combination sequence model, where the sequence is reused predictably, recognizably, and ritualistically. However Tomino had a very different approach, and in Gundam, seemed to use bank as a way to draw parallels between scenes and images. He doesn’t use bank footage for the purpose of showing off a recognizable and technically impressive scene, he uses them for mundane actions and situations, randomly, and with enough recontexualizations that the effect is subliminal.
    I think it’s a bit strange and unfortunate that Tomino distanced himself so much from using bank material after Gundam, as within Gundam, it was becoming remarkably avant-garde. I think if Tomino continued with this examination of the meaning elicited through reuse and recontextualization of imagery, it may have become a foundational technique in anime alongside Dezaki’s more easily adoptable techniques that prioritize production efficiency. For as much as it can be said that Dezaki’s technique is overcentralized around postcard memories as a cost saving measure, it is nevertheless, a technique that has been adopted by many to save resources, and remains Dezaki’s most recognizable technique. I wonder about a different path anime’s history could have taken where Tomino’s use of bank is as recognizable as Dezaki’s use of still images.
    Maybe there’s a degree to which Tomino was too interested in reinventing himself to more greatly appeal to broad audiences, and couldn’t embrace a strange technique like this. I was talking to Pilo about the “Suzumura Ikkou” credit used in Gundam for the Doan’s Island episode. There’s an article written by Kawahara Yoshie where he discusses that “Suzumura” was the name of a nearby restaurant, and that “Suzumura Ikkou” meant something like “let’s go to suzumura” among the staff, an ironic gesture used on the episodes with the worst schedules.
    The Suzumura credit does actually appear one other time in the series, on episode 41, which I think was also outsourced, but is much more interesting than Doan’s Island in so far as the reuse of footage is probably the most radical and experimental in the show. For as much as I think episode 41 is interesting in the lengths it goes to maintain the production in the unique ways only Tomino is capable of, I suppose Tomino may have been dissatisfied with it.

    Also, I didn’t have as much to say about this but; if Ootsuka was involved in writing up the union proposal, why was he not fired? Did he have more value to the company compared to other animators, and thus couldn’t be fired?
    And you might want to fix the link to your twitter account on this site. It still has the old ‘MatteoWatz’ URL.


    1. Thanks for the comment!
      (And the twitter link thing which I had absolutely not noticed lol)

      I’ve been trying to avoid teleology as much as possible but yes, the reason I insisted so much on Tomino’s use of bank (besides the fact that it’s arresting in itself) was because I had Gundam in mind. As you say, the use of bank on it is particularly intricate, both technically and, let’s say, aesthetically?
      In any case, such a use of bank would probably have been very demanding – either for the storyboarders (or Tomino himself) or the production staff. If there were more information on this specific aspect of Gundam’s production, maybe it might explain why it was used to such an exceptional degree on this show and not any other.

      As for Otsuka: if, as I wrote in the footnote, Rintarô is to be believed, the blacklist was actually not that political, it was mostly used as a way for Tôei’s management to get rid of people they wanted out – maybe partly for political reasons, but also perhaps for others. Rintarô says that, although he was active in the union, he thinks he was blacklisted mostly because he didn’t work and had a bad reputation because of that.
      For Otsuka specifically, I think Tôei’s management was well aware of who was an essential asset and who wasn’t – and Otsuka certainly was. Another indicative case is, in my opinion, the post-Hols situation: Takahata was put to insignificant TV series and probably pushed to leave whereas Otsuka or Miyazaki remained on feature films and were as safe as ever.
      This attitude makes sense to some degree, as animators would have been much harder to train than directors. Moreover, Otsuka wasn’t just any animator, and this is something everybody in the industry (not just Tôei) was probably aware of.

      Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s