Many thanks to ehoba, whose priceless contributions to online anime history discourse have made this article possible, and to Alia Demnati, whose exemplary work was a major inspiration.
The two Patlabor movies directed by Mamoru Oshii are landmarks in anime history. Even without considering their intrinsic artistic qualities, their importance for the development of the aesthetics and techniques of Japanese animation cannot be overstated. Major works in the post-Akira landscape, they played a key role in the maturation of realist animation and contributed to introduce capital players such as studio Production IG, director Satoshi Kon, or animators Kazuchika Kise and Hiroyuki Okiura. And that is not all, as they were also central for the development of 3DCG animation and the spread of the modern layout system, which thoroughly changed animation craft. In other words, they are essential works in order to understand Japanese animation of the last 30 years.
In this article and the next, I am therefore going to retrace in detail the production of each Patlabor movie. Although I want to focus just on those two, they of course do not come out of nowhere: the multiple contexts of mecha in the 1980s, the OVA boom and the development of theatrical anime all need to be taken in consideration. In other words, this will not just be a presentation of each movie, but also a more general examination of late-80s anime and the most important artistic and business dynamics behind it.
Headgear, IG and the road to Patlabor
As is well-known, the Patlabor franchise is the creation of a collective named Headgear, made up of 5 artists: manga artist Masami Yûki, mecha designer Yutaka Izubuchi, character designer Akemi Takada, writer Kazunori Itô and director Mamoru Oshii. Such collectives between completely different creators are rather rare in the Japanese animation industry, usually fragmented in small studios between which navigate individual freelancers. 3 of the 5 artists in Headgear, Takada, Itô and Oshii, knew each other since at least 1981: they had all been members of studio Pierrot and their first collaboration was on Urusei Yatsura. However, the formation of Headgear was the result of a very different context, only partly related to its members’ previous work in Pierrot.
The members of Headgear in 1989. Going clockwise from the top: Mamoru Oshii, Yutaka Izubuchi, Kazunori Itô, Akemi Takada and Masami Yûki
The origins of Headgear and Patlabor are instead to be found in the duo formed by Yûki and Izubuchi. Between 1980 and 1983, Yûki was part of a group of manga artists named Doronto Theater (or alternatively Doronto Empire) which, among other things, pitched original mecha ideas to studio Sunrise, but also collaborated in radio dramas and wrote for anime and manga magazines. Around 1983, Yûki and his assistants came up with a project named Bidol, a mecha series which would take place in a post-apocalyptic Japan – their idea of Tokyo needing to be rebuilt after a devastating earthquake would be the origin of Patlabor’s Babylon Project. The project seems to have been progressing quite well, and to have been planned as a 26-episodes anime series – but never went past that point. It is at this time that Yûki thought of reorienting his original idea towards a police series, and that he came into contact with another group close to Sunrise, Parallel Creation. One of the members of Parallel Creation was mecha designer Yutaka Izubuchi, who quickly became interested in Yûki’s drafts. There, both men’s stories diverge: Yûki says that he shared his ideas with Izubuchi as a friend and colleague in 1983, while Izubuchi claimed that he first heard about the project 2 years later, through an editor of the magazine Animec. He said:
I met an editor of the anime magazine Animec and he showed me Yûki’s story book, which contained sketches of characters and scenes, and images of robots used by the police. I thought this could become something interesting!”
Regardless of the exact date, Animec seems to have played an important role, since Akemi Takada also mentions it in her own account of Headgear’s origins. In an interview, she explained that the magazine organized various promotional events and talks between creators. It is there, she said, that Izubuchi came in contact with her and Kazunori Itô and then introduced them to Yûki. The group quickly hit it off, and Yûki shared his drafts with his new friends. The project reached a new stage in December 1985, when Itô invited Bandai producer Shin Unozawa to a Christmas party the group organized. As their “Christmas present” (Izubuchi’s words), they presented him with the Patlabor drafts, which Unozawa immediately agreed to produce.
Unozawa was not just any producer: he was probably one of the best people you could get in touch with to create a new series. Originally in charge of plastic model sales and planning, he was one of the founding members of the “Bandai Frontier” division in 1983. Bandai Frontier represented the company’s earnest entry in the anime market: another member of the division was Shigeru Watanabe, the creator of Bandai’s home video label Emotion and the producer of the very first OVA – Mamoru Oshii’s Dallos. As Patlabor developed, perhaps there was some rivalry between Bandai producers – especially between Unozawa and Watanabe, who had his own group of protégés in the form of Gainax and their OVA-turned-movie project, The Wings of Honneamise.
It’s not clear exactly what Unozawa’s role was at first – once again, accounts diverge. According to Yûki, Izubuchi had pitched the Patlabor project to Sunrise before they met Unozawa, and what the producer did was actually to take back the project and handle it himself. According to Unozawa himself, however, he was the one to send the idea to Sunrise. This was in 1986, just as the Gundam franchise was back with Zeta. Unozawa’s idea was to create a parallel franchise with a completely different setting, that would not compete directly with Gundam. However, probably because they believed that it would, indeed, create competition, Sunrise’s executives rejected the proposal.
By June 1986, the 4 artists held a “training camp” in a hot springs resort, where Takada produced her first character drafts based on Yûki’s designs. Their original idea was to make an omnibus OVA series in which each episode, although it would take place in the same universe, would be handled by a different team. The impetus for that was most probably the omnibus OVA Twilight Q: the first episode, released on February 28, 1987, was written by Itô and featured designs by Takada. However, Unozawa had different plans.
Probably realizing that he could not compete with either Sunrise or Gainax, and witnessing that the OVA market was quickly becoming crowded, Unozawa decided to adopt a different business strategy. He radically lowered the budget, settling on 10 million yen per episode – that is, less than half the average budget for an OVA episode at the time, which was around 25 million yen. He also shortened the project’s planned run by half – from 13 to 6 episodes – and finally asked that a single director be put in charge of overseeing the series. It’s not clear whether bringing in Mamoru Oshii was Unozawa or Itô’s idea – in any case, it was the latter who went to convince Oshii to join. It was apparently an uphill battle, as Oshii was initially not interested in doing a mecha series.
By mid-1987, almost all pieces were in place. Although Oshii was a freelance director, he had been working with studio Deen since 1985, and it was probably thanks to him that the company agreed to produce Patlabor. Unozawa asked the 5 artists to officialize their status as a collective, apparently for copyrights management – and thus, Headgear was born.
Deen had been brought on board for production – that is general management – but it would not be the one to actually create, in the very material sense, Patlabor. This task would be outsourced to multiple studios, among which the most important one was a young company created in December 1987: IG Tatsunoko. As the name indicates, IG was created by two Tatsunoko-related people: producer Mitsuhisa Ishikawa and character animator Takayuki Gotô – their initials made up the “IG” in the studio’s name. They decided to set up their own place during the production of Red Photon Zillion in 1987. Ishikawa was a freelance producer who had always worked under one-time contracts with Tatsunoko, while Gotô was the leader of a freelancer collective called Studio Chime. At the time, Tatsunoko seems to have been in a particularly difficult financial position, which led Ishikawa to reportedly mortgage his parents’ house to raise funds for the show. The situation was dire, and Ishikawa decided to break off from Tatsunoko: with Gotô, Zillion’s character designer, and support from most of the artists who had worked on the show, they created their own studio.
The investment mostly came from two companies: Tatsunoko itself, and Kyoto Animation. The latter, which was still only doing cel painting at the time, had a long relationship with Tatsunoko and had extensively contributed to Zillion. Its president Hideaki Hatta was probably convinced by Ishikawa’s enthusiasm and decided to help him. This was in line with Hatta and his company’s creators-first philosophy – under Gotô’s lead, IG was explicitly set up as an animator-centered studio – but also a smart investment: Kyoto Animation had been set up as a real company in 1985, and establishing a close relationship with IG meant more partnerships were to come.
At the time of its creation, IG only counted 5 members, but all the artists who had worked on Zillion represented an important network of other freelancers and studios, such as Kyoto Animation, Anime R… and Mamoru Oshii. Indeed, it is also on Zillion that the roots of the relationship between Oshii and IG were established: Oshii was asked to help out by the series’ director, his friend from Tatsunoko Mizuho Nishikubo, and therefore storyboarded episodes 2 and 15 under the pseudonym of Rei Maruwa. It is apparently because of that experience that Oshii asked IG to work on the Patlabor OVA.
The Patlabor franchise was then born in earnest in April 1988, with Yûki’s manga and the OVA starting in parallel. The conditions were tough, and Kazunori Itô later explained how they managed to make do with the ridiculous budget they were given:
We wouldn’t just spend 10 million yen all at once… so we spent 8 million in episodes 4 and 5, saved 2 million each time, and then spent 14 million at the end. We were trying to find a way to make it work”
Thanks to this careful resource management and the great work of the staff, the OVA was an immediate success: a 7th episode was suddenly added to the 6 that were originally planned, and a movie was quickly greenlighted. This was Patlabor: The Movie, released on July 15th, 1989.
Patlabor: The Movie, production and staff
Thanks to a production report by IG employee Kôji Sawai, episode director on the movie, it is possible to make out a rather detailed timeline of Patlabor 1’s production. Sawai was integrated in the production “near the end of Summer 1988” – which may situate it in August, that is between the release of episodes 3 and 4 of the OVA. The project must naturally have been greenlighted earlier, which in any case was very early given the OVA’s run – producers Shin Unozawa and Tarô Maki reacted very quickly to the series’ success. Then, we know from the movie’s making-of that location scouting took place on November 8 and 9, 1988. With art director Hiromasa Ogura mentioning that it took him and his team 6 months to complete the backgrounds, we can conclude they were finished in April or May 1989 at the latest. Considering that IG’s last episode on the OVA, #05, came out in November 1988, the animators probably followed a schedule similar to Ogura’s. All this coincides with Sawai’s testimony, which mentions that editing happened in “early summer” of 1989.
In other words, the production was very tight; it was even more so the case because of the movie’s low budget – between 100 and 150 million yen, ten times the cost of an OVA episode, but still low for a movie at the time, especially when budgets for theatrical productions were rising extremely fast and getting close to a billion yen for the most ambitious ones such as Akira and Kiki’s Delivery Service. But there was no room for negotiation, especially for the indebted IG, which had after all been born out of similarly dire conditions. And, in any case, the result was very strong: the movie didn’t move much, but when it did, it did so superbly, and is carried by superb storyboarding, art direction and general visuals and sound effects.
As it was their first or second work together, Oshii was not acquainted with most of the animation staff from IG. However, many other people on the movie had already collaborated with him for some time. There was, of course, the group of voice actors Oshii had almost systematically worked with since Urusei Yatsura or even, for some, The Fantastic Adventures of Nils: Shigeru Chiba, Toshio Furukawa or Yô Inoue. They were all coordinated by recording director Shigeharu Shiba, a collaborator of Oshii’s since Nils, who then worked on every one of his subsequent animated works. Then, we find two artists who had started working with Oshii on his episode of Twilight Q: art director Hiromasa Ogura and color designer Sayuri Ike. Similarly, the two production managers on the movie, Masayuki Hanzawa and Makoto Kubo, had been on Twilight Q #02. As for music composer Kenji Kawai, the long collaboration he would establish with Oshii had begun in 1987, on the latter’s first live-action film The Red Spectacles.
Finally, while this was their first collaboration with the director, it’s worth mentioning that at least 4 key artists on Patlabor had worked on Honneamise: these are art director Hiromasa Ogura, photography director Mitsunobu Yoshida and layout artists Takashi Watabe and Kiyomi Tanaka. The Honneamise connection I will dive in more detail when discussing Patlabor 2, but it may have been established, at least for the two layout artists, as follows. On Patlabor 1, Shôji Kawamori was present on mechanical design, possibly invited by Izubuchi. Kawamori was then probably the one to get Kiyomi Tanaka on board: a mechanical designer and illustrator that may or may not have been from studio Nue, Tanaka had at least had some involvement with Macross. And then, being offered the position of layout artist, he would have invited Watabe, his colleague on Honneamise who stated that he was invited by someone who had worked with him on the movie – although he couldn’t remember who. In any case, given Bandai’s heavy involvement in both movies, that channel may have also played a role.
Oshii, Kise, Utsunomiya: anime realism in 1989
Despite all potential production difficulties, Patlabor’s animation maintained a strong level of consistency in the drawings and quality in the movement. In all their testimonies, the artists involved (notably character designer Akemi Takada, director Mamoru Oshii and animation director Kazuchika Kise) all explain that they were aiming for some sort of “realism” in the designs, models and movement. Such statements are very important given the situation of the movie in anime history and Oshii’s career. For this reason, it appears necessary to make a small recap on the development of realist animation until 1989.
Throughout most of the decade, almost all realist animators regardless of their origin were aiming for what I call “slow acting”: animation on 1s or 2s that used a lot of drawings to create movement that would be as fluid, detailed and nuanced as possible. This agenda was largely formulated by Takashi Nakamura on Gold Lightan in 1981 and Genma Taisen in 1983. This goal was cleared by the three great realist movies of 1988: Char’s Counterattack, Grave of the Fireflies and especially Akira.
As soon as Akira was completed and released, all the artists adjacent to the many realist groups therefore faced one question: which way to go now? The most visionary among them (notably Mitsuo Iso, Satoru Utsunomiya and Kazuchika Kise) came to separately formulate a new objective: it would be to create movement as detailed and convincing as Akira’s, but on much lower framerates. In other words, it meant doing in “limited” animation what Akira had done in “full” animation.
Things went very fast, as the three men’s experiments in that new direction all came out in 1989: these were Mobile Suit Gundam: War in the Pocket (Iso), Patlabor 1 (Kise) and Gosenzosama Banbanzai (Utsunomiya). It so happened that two of these three works were directed by Mamoru Oshii and produced in exact parallel: the first episode of Gosenzosama came out just a month after Patlabor 1. Even if Oshii didn’t closely control Utsunomiya and Kise’s work as animation directors (he quickly stopped checking the animation for Gosenzo, leaving it all to Utsunomiya – things were probably similar for Patlabor), he therefore played a major role in the evolution of animation aesthetics, just as he had in his earlier works that championed the “slow acting” philosophy.
Indeed, Oshii’s two studio Deen OVAs, Angel’s Egg in 1985 and Twilight Q #02 in 1987, featured in their staff many students or members of studio Telecom, one of the main origins of the realist movement. On Angel’s Egg, we find Masaaki Endô, Makiko Futaki, Shinji Otsuka and Yoshiyuki Sadamoto. On the other hand, on Twilight Q, character designs would be handled by Katsuya Kondô and the animation entirely handled by Ghibli-related animators. Besides their staff, both works shared a similar atmosphere and animation philosophy, focused on extremely slow, deliberate and melancholy movement. Here, the “slow acting” wasn’t just an aesthetic or technical challenge, but played a direct role in the dreamlike mood of both OVAs, as it seems that characters are barely conscious and act as if they were sleepwalking.
In 1989, as I explained, things were rather different. Gosenzosama was a comedy imitating puppet theater, and Utsunomiya’s approach was to make the most out of a difficult schedule and low framerates: the animation would be constantly exaggerated and theatrical. Utsunomiya therefore staked the “realism” of the animation not on its adherence to strict photorealism, but rather on the ability of the movement to convey physical forces and psychological energies. This was a perfect fit for Utsunomiya himself, a Telecom alumnus, and the many Telecom/Ghibli-related animators on Gosenzosama.
On Patlabor 1, animation director Kazuchika Kise and the artists of production IG took completely different options. Notably, they kept using slow acting in certain scenes, notably what is arguably one of the most important in the movie, the investigation on Eiichi Hoba. There, the function of slow acting was the same as in Angel’s Egg and Twilight Q: use realism to create a dreamlike, uncanny atmosphere – something that was also expressed in the backgrounds, as we will see.
More radically, however, Kise decided, largely on his own, to make radical changes to the character models. At first, he had been given some basic instructions: as Oshii explained, he “told him to make [the characters] realistic, to consider bone structure, to draw heads bigger and eyes smaller”. However, Kise went even further by adding detailed shadows on the faces to highlight physical features such as dimples or small folds of the skin, or more generally to emphasize the volume of objects. This surprised even Oshii, who said “to tell the truth, I didn’t think Kise would go this far. Every time facial expressions changed, he would even animate the shadows”.
This may just sound like Kise being a particularly thorough and zealous animation director. But there was more to it as, by doing so, Kise completely reworked Akemi Takada’s original models and rid them of much of their appeal. The simplicity and beauty of Noa’s bishôjo character was, in a way, “spoiled” by Kise’s corrections which create an off putting, sometimes truly unattractive, impression. Takada herself seems to have been unhappy with Kise’s changes, and the movie’s producers protested; but Oshii defended his animation director until the bitter end so that his work would remain untouched.
Animating shadows was becoming a staple of realist aesthetics in 1989. Indeed, one of Satoru Utsunomiya’s most important responsibilities on Gosenzosama Banbanzai also consisted of animating realistic shadows – not those appearing on characters’ faces, but rather those projected by the characters’ bodies on surfaces such as floor and walls.
Besides this parallel with Utsunomiya, it is possible to cite 2 major precedents for Kise’s drawing of shadows. One is chronologically distant and the other extremely close. The first is 60s and 70s animator Eiji Tanaka, arguably one of the fathers of what would go on to be called “Tatsunoko realism”. Originally a shôjo manga artist, Tanaka put an extreme amount of care into the textures of clothes, skins and makeup. One of the ways he found to create such a sense of texture was by drawing undulating, free-flowing shadows on faces and bodies. For Tanaka, however, the use of shadows was not just an element of realism: it was also part of the appeal, a way to make the characters “prettier” and more sculptural.
The second precedent dates from 1988 and was therefore extremely close to Kise and Patlabor: it is Yoshifumi Kondô’s work as character designer and animation director on Grave of the Fireflies. Kondô’s approach was more holistic, as it consisted of adding multiple lines on the characters’ faces in addition to shadows, and it was also less aestheticizing. Quite the opposite, in fact: following from the movie’s strictly realist philosophy, characters were not prettified in any way. For that reason, facial features such as dimples and skinfolds were present all throughout.
Eiji Tanaka and Yoshifumi Kondô’s respective shadows in Kurenai Sanshirô (left) and Grave of the Fireflies (right)
Speaking in purely aesthetic terms, Kise’s work on Patlabor 1 can therefore be read as the meeting of those two lineages and/or tendencies: the anatomically accurate, but idealized bodies of Tatsunoko realism, and the Takahata-Kondô duo’s increasingly intense rejection of anime’s conventional expressions and designs. However, it is possible to theorize that Kise could only do what he did thanks to his special position as animation director. This can be understood in two manners.
First, acting as he did seems to have been the only way Kise had to impart his own style to the movie. Indeed, he was constrained on at least two sides: on one hand, by Takada’s designs, on which he had no control; on the other, by the layouts, made not by the animators or Kise himself but by two specialized artists, Takashi Watabe and Kiyomi Tanaka. With unflinching support from the director, Kise probably felt that he could “hijack” the movie’s look by overriding Takada’s characters and completely making them his own.
But it is also precisely because Kise wasn’t the character designer that what he did was possible in the first place. Indeed, extremely realistic models would have been significantly harder to animate, and having to use them as a base would have put a heavy strain on the animators. But because Kise added the realistic elements after the fact, that is when he was applying corrections to the already-finished key animation, he took the charge entirely upon himself without losing the movement that had already been created. In a way, it could be said that Kise took the other artists’ animation as a rough template upon which he built to develop his own individual sensibility. This, too, was somewhat similar to what Utsunomiya seems to have been doing on Gosenzosama Banbanzai: although he was character designer as well, he took complete control as animation director so that the animation would perfectly fit what he had in mind. To completely entrust such vital final touches to the animation director was quickly becoming one of Mamoru Oshii’s characteristics, and would remain a key aspect of his and IG’s future productions.
“Painterly realism”, special effects and the uncanny
If realism was the overall goal that the entire team was aiming for, it didn’t just take shape in the character animation. The background art, under Hiromasa Ogura’s direction, also played a major role. Here, too, it is necessary to go back a bit – more specifically, to Twilight Q #02, which was definitely where most of Patlabor 1’s aesthetic, atmosphere and ideas originated.
Ogura was a student of the single most important art director in anime history, Shichirô Kobayashi: he joined Kobayashi’s studio in 1977, and did his first major work on the epoch-making Treasure Island. He remained there until 1983: then, with fellow Kobayashi alumni Hiroshi Ono and Toshiharu Mizutani, they created their own place, Studio Fûga. It is from there that Ogura would make his first masterpiece as art director, The Wings of Honneamise, in 1987. As Honneamise’s production was wrapping up, Ogura heard about Twilight Q and wanted to work alongside Oshii; he therefore left Fûga and became the art director of Oshii’s episode as a freelancer.
The background art process on Twilight Q #02 was a unique one. It was the first time in Oshii’s career that he proceeded to do thorough location scouting around Tokyo to take photos of rundown buildings. Then, Ogura and his team proceeded to directly trace their background art from the photos to reproduce the exact buildings, while altering the texture so that they would look like badly-made photocopies. Patlabor 1 used a similar process, although it wasn’t quite as experimental.
What was essential there was, once again, location scouting. As I mentioned earlier, it took place over two days, mostly to prepare for the Hoba investigation scenes: one day in a boat to get perspectives from within the Kanda river, and one day inside town and on the banks. A large amount of photos were taken, in black-and-white on Oshii’s specific orders. Aside from this material, Oshii also used many books from photographers famous for their work on Tokyo, which he referred to extensively: in his storyboards, he meticulously reproduced the pictures and indicated in his annotations the exact page and title of the images he referenced. This was capital for Oshii, who stated in the movie’s making-of that his work as director was “entirely dependent on the background art”.
This statement is easy to understand: especially in the more meditative scenes when characters don’t move much or only do so very slowly, the storyboarding and layouts emphasize wide shot compositions in which the backgrounds are the stars of the show. This was further supported by a specific choice by Oshii: that the movie would not only be shown, but also made in a wide-screen format. Indeed, while most anime movies in the 80s were shown in a wide format, the background art, animation and photography were made on square (4:3) paper, and then cropped for theater showings. This of course meant that some visual information was lost, and that the initial way things were made did not correspond to the way they were shown.
As an answer to that, Oshii decided to make Patlabor 1 (as well as Patlabor 2) on wide-screen paper. This was a way for him to claim more creative control, and to materially support the elements he believed central for the movie to work: compositions and backgrounds. But this posed a certain number of challenges: not only did the paper have to be specially ordered, this also changed the way artists had to work. For example, we can refer to the words of Toshiyuki Inoue (who wasn’t on Patlabor) on working in a wide-screen format:
The range of the movement is also determined by the size of the screen: the larger the screen, the wider is the range of movement that the spectator has to follow. So if you animate a fast movement on 3s, it may become hard to follow. For this reason, unlike with TV, we try to animate on 2s.”
This remark illuminates what I previously said about the movement and animation in Patlabor 1: Kise and Oshii’s decision to stick to relatively slow movement and direct most of their effort to carefully crafted compositions and detailed shading work may have been a product of their trying to adapt to the wide-screen format and its specific constraints.
Going back to the background art, according to Ogura, the balance between photorealism and “painterly realism” was quite hard to maintain, but he managed to tilt things in the latter’s direction to avoid something that would be too cold or too close to Twilight Q #02. The way he managed to do so was by resorting to a solution somewhat similar to what Kise did on animation direction: it was to use light contrasts and shadows to create volume and depth.
Ultimately, both Kise and Ogura’s work served the movie’s wider aesthetic purpose, which had probably been Oshii’s agenda since Angel’s Egg and his first experiments with slow acting: to use verisimilitude in order to create a feeling of uncanny. By depicting Tokyo in such detail, Oshii and Ogura made it the real main character of the movie, and invited the viewer to reflect upon it and its existence both as a real city and as an image in the artists’ – or in the antagonist Hoba’s – mind.
Besides the backgrounds, there was another tool essential to realize this vision: that was CGI. Patlabor 1 is largely about computers, and visualizing their interfaces was essential. In line with the series’ setting, technology would not appear too futuristic, and the computers of 1999 would remain close to those from 1989. This, in addition to Oshii’s dissatisfaction at the usual look of computer interfaces and modelizations in SF anime, led him to try to integrate CG animation for one scene in particular: the 3D modelization of the Ark as it collapses at the end of the film. Oshii himself explained it as follows:
Considering the digital techniques of the time, it was hard to animate robots and even harder with anime characters, so I started to use real digital images to produce the digital images that appear in the story. Until then, digital imagery in science-fiction anime was hand-drawn by the animators, which I didn’t find realistic at all and left me dissatisfied.”
It is there, however, that the movie’s low budget proved to be a hindrance: the 60 seconds-long sequence that Oshii planned would have cost 20 million yen, a sum that the production could not afford. Determined to have CGI anyways, the director turned to one of his friends, Tokumitsu Kifune, who agreed to make a 30 seconds-long sequence for only 2 million yen. Kifune was a teacher at the Tokyo Arts and Design University, and member of the collective IKIF alongside Sonoko Ishida. IKIF had started out making experimental animation, but by the turn of the 80s, they started working on CGI and taught it at university.
IKIF wasn’t the only one to contribute CGI: on the movie’s credits, we also find the post-processing company Terebi Technica. It’s hard to tell what exactly they did on the movie, but according to French researcher Alia Demnati, it must have been either to transfer IKIF’s footage onto film, or the promotional clip for the Babylon Project. As for computer monitors and screens, they appear to have been made analogically by using various optical effects.
This may have been done by the photography staff, but there is one last element that is worth mentioning: the work of the two special effects artists, the couple formed by Masahiro and Sumie Murakami. Indeed, Patlabor 1 makes a rather heavy use of multiple cel processing techniques, notably airbrushes. They were used in a variety of contexts: most often, airbrush was applied to the Labors when they stand in the background of a shot or just stand immobile, to integrate them better with the background and perhaps emphasize their unique presence. This was a rather common use of the technique.
Where it became truly original was when it was applied to characters, which is especially the case of two of the movie’s most iconic scenes. The first is the opening part, Eiichi Hoba’s suicide: in an extreme close-up, Hoba’s face is slowly panned from the right of the screen and progressively invades the entire frame. The image has a specific sense of texture thanks to the brushwork, one very different from the one usually used for the skin of human characters. Here, the use of the airbrush indicates Hoba’s abnormality – his madness, inhumanity, or perhaps the fact that he has already renounced life. Alongside the beautiful lights of the setting sun portrayed by Ogura and the background artists, it gives this scene a unique horror film atmosphere.
The “horror” aspect is also present in the other prevalent use of airbrush: that is when Asuma, after having tried to hack inside Hoba’s HOS, provokes a major incident inside the Shinohara Heavy Industries factory. As all screens are invaded by the same message, we get a shot of Asuma from above with a fisheye lens. In what is quite a technical feat, only Asuma’s face (and the reflections of screens on it) receives the airbrush treatment. Once again, the very specific texture tends to separate him from “normal” animated characters, illustrating in this case the bewilderment and terror Asuma goes through.
All of the above remarks allow us to understand what was Patlabor 1’s objective as a film: a very specific sense of atmosphere best summed up by the word “uncanny”. As computers and machines around Tokyo stop functioning or do so in unexpected, troubling ways, the characters of the movie go through a feeling of estrangement and can’t quite understand what is happening. It is that sentiment that the movie tries to convey to the viewers. We can also see why realism was such an important element: depicting characters, places and movements familiar to the viewer invites them to identify and share the same experience. But at the same time, this familiarity creates distance: characters that move like real bodies and not animated drawings, a city that is both Tokyo and not Tokyo…
Such feelings and strategies are generally associated with Oshii’s body of work, and their recurrence in his career is why he is often considered to be a singular, exceptional auteur. However, as this article has shown, all this isn’t the direct product of Oshii’s individual creativity. It was the result of specific techniques practiced by specific people in specific conditions. If Patlabor 1 is such an important moment in Oshii’s career, it is therefore not simply because it perfectly establishes the themes and style associated with his work: it is also and mainly because it is there, with Headgear and IG, that he managed to truly create a group with the artists who would carry and further his vision for the years to come.
Books and articles
Demnati, Alia. L’industrie de l’animation japonaise aux prises avec l’image numérique : la réponse d’Oshii Mamoru (1995-2004) (The Japanese Animation Industry and the Digital Image: Oshii Mamoru’s Answer (1995-2004)). Art et histoire de l’art. Université de Paris. 2020.
Inoue Toshiyuki. “第２回 アニメの「コマ打ち」とは何か――井上俊之が語る「コマ打ち」の特性 | かみのたね” (What Is “Timing” in Animation? Toshiyuki Inoue Talks about the Characteristics of “Timing”). かみのたね, March 5, 2019. http://www.kaminotane.com/2019/03/05/4984/.
Oguro Yuichiro. “アニメ様七転八倒 第31回 劇場版『パトレイバー』と脳内アニメ” (The Patlabor Movie and Animation Inside the Brain). WEBアニメStyle, March 27, 2006. http://www.style.fm/as/05_column/animesama31.shtml.
O’Mara, Sean. “Bandai, The Post-Gundam Wave, and the Year 1985.” ZIMMERIT – Anime | Manga | Garage Kits | Doujin, December 24, 2021. https://www.zimmerit.moe/bandai-1985-gundam-honneamise-gunpla-emotion/.
Ruh, Brian. The Stray Dog of Anime : The Films of Mamoru Oshii. Palgrave Macmillan US. 2013.
Kise Kazuchika. “それ以前”の世界に『新しい攻殻』を作ること – 黄瀬和哉総監督の考える『攻殻機動隊ARISE』とは? (To create a “new Ghost in the Shell” in the world “before” – What does Kazuchika Kise, General Director, think of “Ghost in the Shell ARISE”?). マイナビニュース, June 21, 2013.
———. 「攻殻機動隊」25周年リレーインタビュー 黄瀬和哉. (Relay interview with Kazuchika Kise for the 25th anniversary of “Ghost in the Shell”). アニメハック, October 30, 2015.
Ogura Hiromasa. 美術監督 小倉宏昌さん (Interview with Art Director Hiromasa Ogura). ちびねこトムの大冒険 公式webサイト.
Oshii Mamoru. 押井守監督インタビュー 「『ミニパト』こそ究極の押井版『パトレイバー』」 (Mamoru Oshii Interview “Mini-Patlabor is the ultimate Oshii version of Patlabor”). 高畑勲宮崎駿作品研究所 Production IG. February 8, 2002.
———. 押井守監督が自身の作品について語る！ 3週連続単独インタビュー 第一弾 ～パトレイバーシリーズ～—シネピック (Director Mamoru Oshii talks about his works! Three consecutive weeks of independent interviews. Part 1 – Patlabor series). WOWOWOビジネス, March 14, 2018.
Oshii Mamoru, Izubuchi Yutaka, and Unozawa Shin. 押井守、絶縁状態だった出渕裕と「パトレイバー」草創期についてトーク！ クッション役にはプロデューサー・鵜之澤伸 (Mamoru Oshii talks about the early days of “Patlabor” with Yutaka Izubuchi, who had been isolated from him! Producer Shin Unozawa in-between them!). アキバ総研 25/01/2015.
Shin Akiko. “制作スタッフが語るぶっちゃけ制作秘話！機動警察パトレイバー トークイベント『暴走トークショー』レポート | SPICE – エンタメ特化型情報メディア スパイス” (The Production Staff Talks about the Secret Story of the Production! Mobile Police Patlabor Talk Event “Runaway Talk Show” Report). SPICE（スパイス), November 25, 2020. https://spice.eplus.jp/articles/279137/amp.
Kasai Osamu, ed. 機動警察パトレイバー 完全設定資料集 Vol.1 TV編 (Patlabor: Perfect Establishment Data Vol.1 – TV). 一迅社. 2007.
———. 機動警察パトレイバー 完全設定資料集 Vol.2 – OVA編 (Patlabor: Perfect Establishment Data Vol.2 – OVA). 一迅社. 2008.
———, ed. 機動警察パトレイバー 完全設定資料集 Vol.3 – 劇場映画編 1 (Patlabor: Perfect Establishment Data Vol.3 – The Movie 1). 一迅社. 2008.
Oguro Yuichiro, ed. 機動警察パトレイバー 映画版 ガイドブック (アニメージュ 1989年08月号付録) (Patlabor the Movie – Guide Book (Animage 1989-08)). 徳間書店. 1989.
Uejô Tarô, ed. THIS IS ANIMATION 機動警察パトレイバー 設定資料全集 (This Is Animation: Patlabor the Movie & OVA Series). 小学館. 1994.
4 thoughts on “Mobile Police Patlabor The Movie”
Fantastic article. Always felt like the Patlabors made special use of backgrounds and establishing shots to reinforce the tone of the scene, and it sounds like those moves were done extremely deliberately.
Amazing material! Can you specify source of Inoue’s quote please?
Is it the one about the size of the paper and timing?
Inoue Toshiyuki. “第２回 アニメの「コマ打ち」とは何か――井上俊之が語る「コマ打ち」の特性 | かみのたね” (What Is “Timing” in Animation? Toshiyuki Inoue Talks about the Characteristics of “Timing”). かみのたね, March 5, 2019. http://www.kaminotane.com/2019/03/05/4984/.
Oh sorry, didn’t notice that was in source list. Thank you for the answer.