Hideo Kojima’s development diary for Metal Gear Solid 2: Sons of Liberty was penned between 1998 and 2001 and published in the book METAL GEAR SOLID 2 THE MAKING in 2002. It was never officially translated into English; however, after I discovered the original scans online, I dedicated an unhealthy amount of time to resource and puzzle together a verified good translation. More on that here
Going in, I wanted to understand if there was anything in the development that sets his process apart from what I’m familiar with. How does Kojima write and what inspires him, how are the playful mechanics prototyped, and how does he handle production, scope, and being a leader?
I’m also interested in understanding how he sees and grows the “Kojima” brand. Is that something he’s conscious of and fosters, or is it all about the games? And finally, how does he handle the stress and pressure that comes with the almost impossible task of shipping a game on time.
Kojima’s Inspirations and Writing Process 🔗
The Metal Gear series mixes surreal with hyperreal, espionage with the supernatural, philosophical depth, and shameless fanservice, all delivered with complete sincerity. Kojima really wants to entertain.
It’s well known that he watches a lot of movies (and reads plenty of thriller novels). In his diary, he often leaves entries about what he watched and what worked or didn’t:
Went to Kyobashi to see Run Lola Run… The first half was interesting! The German techno synced with Lola’s running was fresh, even visually. However, the twist is revealed in the first 30 minutes.
— June 8, 1999 (火)
I get the sense that he really celebrates the media he consumes, breaking down why something is engaging, and then… thinking about how to use it.
In MGS2’s grand finale, the entire sequence where secrets are revealed will be portrayed in real-time, with characters pointing guns at each other the whole time.
It’s the kind of tense situation we’ve seen in Quentin Tarantino or Ringo Lam movies, where more than three characters are all pointing guns at each other, unable to move.
— January 20, 1999 (水)
In his diary, Kojima refers to his writing work as “scriptments,” a term originally coined by filmmaker James Cameron. A scriptment combines elements of both a script and a treatment, making it more structural than a traditional screenplay. Cameron used this method when writing Terminator 2.
By studying filmmakers he admired, Kojima learned about Cameron’s process and adapted it as his own writing style for games. His co-writer, Tomokazu Fukushima, defines a Kojima scriptment as “an overly detailed game design document that covers all sorts of aspects such as plot outline, game features, level designs, specifications, etc.”
Handling the vision for so many elements—and doing it well—requires a lot of expertise. These responsibilities are typically divided among several disciplines, but I’ve come to learn that Kojima is surprisingly multidisciplinary. More on that ahead.
On Themes and Tonal Flexibility 🔗
Kojima tends to pick big themes about humanity, consciousness, and the nature of reality. The stakes are extreme, usually involving the threat of nuclear holocaust. But then you have a knife-wielding vampire who dances flamenco, and design briefs like this:
I added an item called the ‘Hair Removal Device.’ It’s a rare item that you can obtain by putting Emma to sleep and shaking her. If you have it, Raiden’s armpit hair disappears.
— January 16, 2001 (火)
It’s a playfulness that’s characteristic of Japanese media and common in manga. I might be generalizing, but I find that Japanese storytellers more widely embrace tonal flexibility and absurdity in a way that a western writer might fear could undermine the narrative.
I think this flexibility translates well into games and game making. When your audience controls pace and camera in a virtual world that’s barely held together, things will get goofy. So let them be a goof and let the story match that goofiness, (then occasionally hit them with something raw and provoking).
The Luden Mask is a wearable item in Kojima Productions’ Death Stranding. Its description has a nod to the player (or maybe to the game’s creators); it reads: Designed with Homo Ludens in mind, their form encourages the wearer to never forget their playful side.
To contrast, a game like The Last of Us goes to extreme lengths grounding the player and controlling the experience. Objects and actors don’t clip into each other, animations blend seamlessly, and the camera faces the action. It’s all extremely skillfully done, and I admire them both, but maybe with a preference for the goof.
On Creating Characters 🔗
When I told my peers about this research project and asked them to define what they liked about Kojima’s writing, “the charismatic characters” came up a few times, which I thought was pretty apt. The characters are all quite charming, even the bad ones. Maybe especially the bad ones.
Dead Cell, the villains of MGS2, From the left: Vamp, Fatman, Fortune and Solidus.
One of the intended villains for Metal Gear Solid 2, Old Man (or Old Boy), was cut late in production. He would later resurface as fan favorite The End in the sequel. The diary captures the moment of his conceptualization and gives us some great insight into Kojima’s character creation process:
“I’m thinking of introducing a seasoned veteran who prevents aging with telomerase enzymes. A legendary hero from the second World War. Someone who showed even Big Boss the ropes in various strategies. Physically close to 100 years old, and boasts over half a century of combat experience. The codename ‘Old Man’ is said to be a walking piece of military history. He calls all of the soldiers of the world ‘youngsters.’ ‘No one has seen more battlefields than me. And no one alive has come closer to death than I have.’ He attacks using a Panzerfaust. I’ll leave the details of the weapon to Hirano-kun. As for the voice, Saka-san would be great.”
— November 10, 1998 (火)
We get a backstory and a name, some ideas of his mannerisms, and ultimately a vision of what he sounds like. It gives a vivid image and something a concept artist can run with. The format is very similar to what I’m familiar with; it’s the “comic-bookness” and I think, willingness to commit to such a super extra character for a game that’s unique.
On Research and Authenticity 🔗
Throughout development, Kojima is reading a lot of fiction; Stephen King, Nancy Collins, and Paul Auster to mention a few. He appears to be taking inspiration from Stephen Hunter whose thriller/conspiracy fiction would blend well into the world of Metal Gear. There’s also a lot of nonfiction research into U.S. military projects and history.
The team bought gun replicas, studied military magazines, visited museums, and took guided tours of tankers. They even had a military consultant teach the whole development team about room clearing, soldier behaviors, and movements:
“The first day of Mouri-san’s special seminar. We started with the basics, like how to grip a gun and stance. It’s completely different from looking at books or videos—extremely clear and useful. And way too fun! I almost forgot about work.”
— June 16, 1999 (水)
While every studio gathers references and many do tours, I think there’s a noteworthy dedication to authenticity here. Military paraphernalia is almost fetishized, and rivals most modern shooters in realism—it creates a hyper-real presentation that’s then juxtaposed with the hyper-extra plot and characters. This hyper-real vs. hyper-extra dynamic, I believe, plays a major part in creating the enigmatic imagery Kojima games are known for.
Keeping Things Contained and Shippable 🔗
The diary entries don’t provide much insight into workflows, and Kojima mostly talks about the timeline as something that causes him great mental and physical unwellness. However, I can point to two decisions that would have had a major impact on streamlining the production and ultimately getting the game shipped.
First, early in the entries, Kojima details plot and themes that closely resemble what will appear in the final version of the game:
“The idea of an armed group taking over an offshore plant (artificial island) built to clean up an oil spill feels a bit weak. So, we’ll have them take the visiting president hostage as well.”
— November 4, 1998 (水)
“This time, the focus is on things that cannot be passed down through genetic information alone — such as soul, ideas, history, circumstances — and how to pass these on to the next generation.”
— July 13, 1999 (火)
While a large part of the team was still wrapping up the EU versions of MGS1, these narrative pillars were already drafted in the Metal Gear Solid 2 Grand Game Plan Translation and uploaded to the internal server. see the original document.
What stands out is how impressively close this Grand Game Plan is to the final product. That’s not something I typically see with early Game Design Documents as development tends to drift from that initial vision. I’m also a big fan of the document’s format and structure—it’s a template I’ll definitely use for future design and pitch documents.
The Big Shell - The location where the majority of Metal Gear Solid 2 takes place.
Secondly, MGS2 is designed and laid out in two contained locations: the boat (The Tanker) and the offshore platform (Big Shell). Contained locations provide grounding and serve as anchors for the spectacular events to take place. Big Shell, in particular, with its hexagonal layout, is a level designer’s dream—a great space to lay out story events and gameplay.
It might seem obvious, but the benefits of contained locations really solidified for me on my current project. For instance, if your game is set on a boat, there are clear, established rules around layout and architecture unique to that environment, making it easier to define the boundaries of the space. By contrast, a game that spans multiple locations, connects through transitional spaces, or features a shifting structure demands more time and resources to establish those rules.
By not allowing MGS2 to balloon out with complex locations, the team could save time, reduce production complexity, and focus their creativity and problem-solving on the new hardware (PS2), new gameplay ideas, and cinematic storytelling.
What is Kojima Like as a Developer and Director? 🔗
My preconceived notion was that Kojima is a bit of an armchair director/writer who thinks a lot about marketing. That was (mostly) wrong. He is very much a hands-on developer with vast knowledge in tech, design, and art. He understands rendering techniques, level design, and texture work. He writes both large-scale story structures and detailed design briefs.
Predictably, he comes across as a perfectionist, offering very detailed feedback on specifics. In particular, for MGS2, the rain and underwater effects are frequently mentioned and discussed.
“I shared some ideas about the underwater effects with Okajii and Shibata-kun. Personally, I hope we can convey a sense of being underwater and floating without using underwater fog. First, let’s see how far we can go with bubbles and floating particles. The key is the refraction of light…. Worst case, we might have to use some raster-like processing.”
— January 16, 2001 (火)
As a leader, Kojima comes across as blunt but fair. He sets high expectations and sees grit during uncertain phases as a sign of a true professional. Yet, I also notice a lot of care for his team; he mentions them frequently, gives each lead a nickname, and considers their creative growth.
“Shin-chan’s drawings have improved significantly since he joined the company. However, looking closely, while they’re well-crafted, his pictures don’t have the same raw energy as when he started. I’m not looking for a polished drawing from him. I want his unique ideas and taste. I remember his doodles of Captain Underpants.”
— February 9, 1999 (火)
I didn’t find much reflection on his own personal brand, but it’s clear that he thinks deeply about what it means to stand out, achieve success, and meet the expectations of eager fans. He pushes very hard for novel gameplay and ‘world firsts,’ understanding that being different from the norm—doing what others don’t—is worthy of attention.
This following entry is a telling reflection he shared after watching an interview between the Ghibli directors:
“After I got home, by chance I ended up watching a making-of program for My Neighbors the Yamadas. From the position I’m in now, I wasn’t convinced by Takahata-san and Miyazaki-san’s words. I couldn’t accept Takahata-san’s declaration, ‘It’s fine if this movie doesn’t sell. Even if it doesn’t sell, it’s still a success as an animated film!’”
— July 26, 1999 (月)
The Toll on the Body 🔗
What really blows my mind, though, is the absurd number of projects he juggles simultaneously. While developing MGS2, he’s also working on the European version of MGS1 and MGS1: Integral. On top of that, he’s producing an entirely different game, Zone of the Enders, and writing movie reviews for a magazine. And if that wasn’t enough, he’s out drinking with colleagues and has a seven-year-old child at home! Somehow, he also finds time to write this dev diary, watch movies, and read books.
Predictably, his work-life balance seems busted. Many work entries are written on weekends. The toll the project has on his mental and physical well-being is apparent with vivid descriptions of health issues and frequent visits to doctors and hospitals. It all seems stress-induced:
“During meals and breaks, the MGS2 will suddenly come crashing down on me, like it’s going to crush me. I have no appetite. I know I need to eat, but it’s hopeless. Somehow I manage to swallow something.”
— January 15, 2001 (月)
Stress, anxiety, and overwork are common in game development, and I know this “pressing down” feeling very well. Thankfully, for me, it has gotten better in the last few years. Kojima was in his late 30s when this was written, and there’s no indication of any healthy shift. In one of his last entries, long after development had ended, he still hadn’t recovered, and the diary concludes on a somber note where he bemoans his bad health.
“The Year of the Snake is almost over. I feel drained. And my health still hasn’t returned.”
— December 28, 2001 (金)
A few months later, Kojima announced that he did not plan to return as the director for the Metal Gear series. Perhaps concerns for his health played a role in that decision? After shipping a game, I’ve often asked myself, “Why am I doing this again?” It’s usually not the joyful period one might expect. But then, slowly, creative curiosity and motivation start to return, and the cycle continues. I imagine something similar happened to Kojima, as he ultimately returned to direct several more games in the series.
Personal Takeaway and Reflections 🔗
It’s easy as creatives to become a bit narrow in our vision and celebrate talent and craft more than the product’s entertainment value for an audience. I believe Kojima keeps perspective of that and pays a lot of attention to other directors he sees doing the same.
He also works with an awareness of the medium, despite worshiping and talking about cinema almost daily. He sees his own creation very much as a game with an audience that will put their own final touches on his work.
I didn’t find much insight into the development of emergent gameplay and systems interplay that are such defining elements of the series. Similarly, I would have liked to learn more about the decision-making hierarchy and how the skeletal structure was laid out in the engine. In his final entry, Kojima admits that he’s cleaned out half of the diary’s content to remove “sensitive information,” which would have made for much juicier reading, I’m sure.
Something that did surprise me was how enjoyable it was to peek back at the turn of the century—the insane hype for the PlayStation 2. Classic movies hit theaters, like The Matrix, Austin Powers, and Eyes Wide Shut. It was particularly satisfying to read about Kojima’s jealousy of Zelda and his envy of Western productions like Half-Life.
My first time playing MGS2 in Sweden somewhere 2002
I often wonder if doing these studies and writing these essays actually help me as a creative… or if I should put all that effort into actually… making stuff. It gives me comfort then to see that enjoying and understanding entertainment—and the people who create it—is something Kojima so openly embraces.
I think my main takeaway is to let the things that inspire me guide me and, maybe, worry less about everyone’s approval. If I feel confident in what I like and I can work out why I like it, I should chase that energy rather than trying to impress. I’ll probably have more fun too.