If you click on a link and make a purchase we may receive a small commission. Read our editorial policy.

Exclusive: Ken Levine on the making of Bioshock

When Ken Levine, the main man behind Bioshock and System Shock 2, drops you a line asking if you want to do an interview, you say "yes."

Levine's a fascinating figure – articulate, driven, passionate. And, no, I don't want to have sex with him. (Denial’s not pretty – Ed) It's worth stressing how this interview came about. Levine – a major developer – mailed me for no other reason than that he wanted to talk. No-one does that. He's played the PR machine on Bioshock enormously hard, clearly very aware of the enormous stakes he's playing for. And he is, in a real, fundamental way. Levine sold the company he co-founded in order to get this game done. Irrational no longer exist in name thanks to selling it to 2K, but without their money Bioshock wouldn't have been made in a recognisable way. It was only possible because of the Faustian deal, and he needs to make the best of it. It has to do what none of its peers and precursors (The Thiefs, The System Shocks, The Deus Exes) have done – become not just a hit, but a enormous HIT. If Bioshock does anything short of changing our world, he's failed.

So, yes, he likes to talk. As he should.

Anyway – Bits of the interview end up being cannibalised for features in PC Gamer UK, Wired and Edge. If they come online, clearly, I'll be linking to them – the PC Gamer one has lots of stuff on designers’ ethics and needs, while the Edge one is a making-of look at System Shock 2 (The Wired one's up now, and you've just wandered past its link. And I've edited the PCG one in now too - Ed). However, even with all that, there were still several thousand words of interesting material left spare. In the days leading up to Bioshock's release, Rock Paper Shotgun seems the perfect place to share them. I've included narrative bridges for the bits which have gone into the other pieces to give context. Oh – and this feature was written before I'd played the finished game, having only experienced the first couple of levels in preview.

We start at a fairly obvious point, but I was fishing for quotes for the more general-readership Wired feature. Bear with us, and read on for Ken's thoughts on the legacy of System Shock, how Little Sisters were formerly insects, the nature of superheroes, objectivism and, of course, much more.

Rock, Paper, Shotgun: So, how did you go about conceiving of Bioshock's setting, the underwater city of Rapture?

Ken Levine: It's a complicated question, as it evolved through a lot of thinking, starting with game design. I don't start with Story, because games are not story. Games are gameplay. Games are interactive. We'd done System Shock 2 and we'd discovered something that worked very effectively. You were in this isolated spot on this space-ship... so we could simulate that spaceship completely, at least for the time. We never ran into the situation where you're in New York and all of a sudden there's this Jersey barrier you can't go over, or there's this person coming down the street who you can't interact with in a way you would in real life.

Of course, in videogames we're making huge progress, but we're not at 100% reality yet. I wanted some place where the player could be cut off from the rest of the world which we could simulate incredibly deeply. I had this notion of an underwater city. From that, I thought if we wanted to make a game that was scary, it has to be believable... so why on earth would there be an underwater city? So I came up with this notion of this Utopia they didn't want anyone to find. From that, I wondered what sort of Utopia it would be, and came up with the character Andrew Ryan and his sort of philosophical background: pseudo-objectivism and extremely capitalistic view on the world. He'd be terrified the New Dealers in the US and the Stalinists in Russia would find his city, so – as he said – it wasn't impossible to build a city at the bottom of the sea – it was impossible to build it anywhere else.

RPS: This reminds me of something Doug Church said when I interviewed him about System Shock 1. It's better to simulate an enclosed environment in more detail rather than a larger environment with less fidelity. Is that something you'd agree with?

KL: I was very lucky to have Doug be my mentor when I first started out. It's something I definitely agree with him on. I think it's a piece of philosophy I inherited from him.

Talk turns to game design freedom and similar, before segueing back towards Ayn Rand Objectivism. Levine expresses more sympathy for Rand than I'd expect, and we debate the merits of Atlas Shrugged a little. He argues that Bioshock is more about the problems of unquestioningly following any philosophy.

RPS: Rapture kind of reminds me of throwing the muddy human elements into a pure philosophy...

KL: To me, Andrew Ryan is a combination of several historical figures, like Howard Hughes and Ayn Rand together. Unlike a character in a Rand book he's “a real person”. John Galt is a superman. He's not a normal person. He doesn't go to the bathroom. If you read the Fountainhead, the characters are these idealised supermen. They don't have doubts, they don't have fears – at least the hero characters – they don't make mistakes. And that's much like the Superheroes of the 40s. I think people are much more like the Superheroes of the 60s, Stan Lee's Superheroes who have real problems and make mistakes. I think Rapture is a place where there's a very powerful ideology put into play by actual people. And when people get into the mix, things get complicated.

RPS: One of the things I noticed when moving through Rapture was the vast numbers of art assets in it. There were things in the two and a half hours I played which I saw once, and I swear never again. It's a phenomenal amount of work – and it’s a different way of doing graphics. Instead of just pretty assets, there’s a variety of assets to make the world more vivid. It's consistent.

KL: One of the advantages of having an aesthetic which is consistent and a great art team is that, once they know what they're doing, they can create very beautiful things very quickly. All those questions you ask when you build an object into a world become a lot easier. When you're struggling with your aesthetic, it slows everything down. A talented art team with a clear aesthetic will go really far, really fast. That's not to underestimate the amount of work and talent. They're incredibly productive and – for the industry – small. Bioshock was made by a team of between fifty and eighty people, compared to ones you see at other publishers of 150-200.

"Of course, in videogames we're making huge progress, but we're not at 100% reality yet."

RPS: What was the internal team set up? How do you work and organise yourself at Irrational?

KL: We have separate departments. During the process of development, we added an animation department which we didn't originally have – it was originally part of the art department. We have a sound department, and a design department. Each one has a lead, and I interact primarily with them, telling them things I'd like to see done. I work with individuals on the team less. As the team gets bigger you need these hierarchical structures more so you don't get eighty people trying to talk to each other every day. The weird thing is we also have Australia, our other studio in Canberra. That's where my biz-partner Jon Chey is based. We sort of dragged them into the project more and more, as it got bigger and bigger. As the budget increased and Take 2 got more confident, we needed more people. We tend to be fairly slow to hire, so we didn't really have enough staff to make the game. We brought in more staff from Australia on a regular basis, which meant the product they were working on ground to a halt... but we really wanted Bioshock to be great. It evolved over time. I'd like to say we had it planned perfectly at the beginning, but that would make me a stinking liar.

RPS: How long has Bioshock been in development now, completely?

KL: It's a mix. We've been thinking about doing another first-person shooter game with a much more player-centric design since System Shock 2. The core design principles happened about four and a half years ago – the notion of the simulated environment and the three interacting AI types. The aggressor type. The resource gathering type. The protector type. That came from watching nature shows, where I wanted AIs in the world where people could watch them in the world and immediately understand their relation, in the same way they would if watching a nature film. The more that's intuitive, the richer the game can be. Things sort of evolved from there.

Conversation segues to the origins of the Little Sisters. I had heard that the Little Sisters had, in the original design, actually been insect creatures. Ken confirms it, explaining that – clearly – no-one had any empathy for bugs. As they're bugs.

"Rapture is a place where there's a very powerful ideology put into play by actual people. And when people get into the mix, things get complicated."

KL: But the game went through a fair amount of changes. Never the core design principles, but the aesthetic and story evolved over time. That was the later stuff. I had core story notions, but they kept being refined. I didn't sit down and put pen to paper and write the actual words until November or December of last year. Things change. And then when we have it working, we do a lot of external testing to see if people are getting it, if people are enjoying it. And not surprisingly, the first time we sat people down, they didn't – at least, not in the way we wanted them to. We would test, and then tweak, then test some more, and tweak. It was really powerful to make these small changes and get a huge result from it.

RPS: Could you give an example of that?

KL: I was really concerned. We'd made a mistake in previous games where we'd thrown every game system at the player in the first minute, in a pretty rich game. I think it can be alienating for some people to be overwhelmed by the wealth of options in the first moment. So we were trying to go in the other direction – the game really opens up as you go on. At the start, it's fairly linear without a large amount of player choice. But by the time you get to Arcadia or something, it explodes with all those choices. But we made a mistake by pushing it back a little further than it should be – the focus group were “Okay! We get it! We're psyched! Where's the rest of the stuff?”. Oh man! After all this worry, we'd gone too far, so went back and moved some of the complexity and the depth up. Move some plasmids earlier, move some weapons earlier, move some choice earlier. That was really interesting. I'd rather go too far with that and then have a test group tell us they get it, than create a game which overwhelms people. No matter how good your game is, if people don't get it... it's not good. It's not a good game.

We talk a little about games in this lineage and their underperforming sales. Ken argues that it's a case of them not concentrating on “the bread and butter” - they can be imaginative, but they need to scratch the basic itch too. We then get around to talking about the problems of being seen as a spiritual sequel to System Shock 2. Ken mainly laughs it off, as a “Rich Man's Problem” which has enriched his life. But games have to move on...

KL: We've seen a lot of games of that ilk come out and not make the impact we wanted them to, so slavishly following that formula 1 to 1. When you play the game you see there's a huge amount of resonance between the two [System Shock] games, but we needed to make Bioshock much bigger and grander, and we have the resources to do so.

RPS: The other thing about the Spiritual Sequel line is that I think it confuses people. I see lots of Shock 2 in there, sure... but I think it's closer to Shock 1, in many ways. Being based on the equipment rather than actual statistics - that it's more of a shooter with a lot of things piled on top...

KL: Absolutely. It's funny, as when we were making Shock 2, any time there was any deviation from Shock 1 there was a lot of very angry people. And now people are forgetting that Shock 2 really added all those RPG elements and character growth stuff that wasn't in Shock 1, which was more of a clean FPS. There may be some more similarities, in that regard. And hey! I'm cool with that.

RPS: I talked a little about the controversy issue earlier, but with Manhunt 2 being banned in the UK, there's clearly a controversy waiting with the Little Sisters. However, you've done a lot to try and be sensitive and careful in the game. Could you discuss your internal dialogue about what's too much and what isn't enough?

KL: For me it was always about... well, we got some heat from this from our audience [Ken is referencing some of the Shock fans being annoyed they couldn't actually kill the Little Sisters in random play, instead having to actually use a special device on them]. They viewed it as impacting the amount of choice they had in the game, but for me – and I speak for myself personally as a game developer – I'm a strong believer that artists should be able to make art. And the big difference between art and reality is... guess what? Art is not reality. I'm a huge believer in other artists making what they want to make and what they're comfortable in making. I can only speak for what I'm making and what I'm comfortable in making. And that's the best Bioshock we can make, which is the current incarnation, being a game which forces you to make the moral choice of whether to exploit the children or help them. And that's a very clear choice that’s right in front of your face.

All the other stuff – the ability to chase them down with a gun and hurt them that way – for me, it was personally distasteful and irrelevant to the story we were trying tell. It didn't enhance the experience. It took away from the experience. As you'll have seen, the combat gets so frenetic, they would get caught in the cross-fire and there would be no choice. It'd be a friendly fire incident. For me, it was meaningless and distasteful. We carefully went about it working out how it could be powerful and impactful and how the player could feel the impact of their decision... and I think the art team has succeeded in that, we've succeeded in that. I think people who are frustrated by that should play it, or at least view a video of that, and make their own decision.

RPS: The idea of the kids just dying in a Keystone Cops situation strikes me a waste of time, really.

KL: And to be frank, that happened all the time. You've seen how Big Daddy combats go. And I know people say, “Make the AI better”. But in real life, no matter how good people's AI is, they get killed in stupid situations which took away all the mediums.

RPS: They have to be vulnerable, or else they'll be able to look after themselves, by definition. An acrobatic Little Sister wouldn't need the big Daddy.

"The big difference between art and reality is... guess what? Art is not reality."

[Conversation moves onto System Shock 2, including the hilarious secret origin of the Psychic Monkeys. It starts to wander back to Bioshock when we talk about Eric Brossius' contribution to the original System Shock...]

KL: I remember even in Shock 1, the sounds the mutants and robots make, you can almost think you can make out English words, but you couldn't quite. To tantalise with meaning, without giving any meaning. Obviously only in a very limited way in Shock 1 days, as they had limited memory and a limited budget for that sort of stuff, but it's something I took to my writing for the AI in the world. In Bioshock, the AIs are sort of in a fugue state where they're half in the present and half in the past, and they're talking out loud to themselves as if they're still in the lives they lost so long ago. The insane rantings... I think those notions are quite powerful, and more meaningful as they exist on their own. There's a great experience in Shock 2 and Bioshock of hiding behind a cabinet and listening to them mumble to themselves, and you sometimes hear things which make them quite sympathetic. In Bioshock, the woman with the baby-carriage... she's obviously had a terrible loss in her life, and she doesn't know whether she's in the past and the present, and her mind slips between the two. It doesn't make her any less of a dangerous foe, but we really wanted to establish those themes pretty quickly.

RPS: Rapture's a fascinating place to be in. It's not just an original setting for a game, but an original setting per se.

KL: I think there are things which are too original. Really great games like... what's Tim Schafer’s game? Not Grim Fandango...

RPS: Psychonauts?

KL: Yes, Psychonauts had this problem. Where you have so much originality that people can't find a point of entry to it, in terms of the mass market. But people who spend some time with it, really fall in love with it. I wanted to make sure that Bioshock.... well, it's one way the shooter aspect works really clearly and people understand it as a shooter. I think if Psychonauts had been more clearly marketed as a platformer, and made sure its platforming was second to none, the mass-market would have gone with the really cool aesthetic a bit more. Some point of comfort to come in on.

Then we talk about the art-deco architectural style a little more, before rolling back towards how Bioshock does storytelling – that is, in-game, rather than cut-scenes.

RPS: It’s interesting that while you come from a screenwriting background, you don't want to turn games into films. Games are games, not cut-scenes.

KL: I think part of it was that I've been there. I've been in the film industry, so I don't have this desire to make minimum movies because I've been fortunate to work in that industry. I take games for what they are. I want to make great games. I don't have a second career I'm trying to experience. I'm trying to make a videogame, which allows me to concentrate on the storytelling capacity of videogames, not just aping another medium. I can't speak for other developers, but it seems like they want to recognised as great filmmakers. Which is fun – I got to write and direct that high-res trailer we did. That was awesome. I had a great time doing that. I got a chance to show my filmic and cinematic chops. But that was designed to be a non-interactive experience. I tried to bring some sense of interactivity, in putting the viewer in first person. I really enjoyed that. But that and the game are something else. I was sort of able to step away from doing that – controlling the experience. While in the game, I give control to the player, as they deserve to be in control, not me.

Bioshock is released 21 August in the US, and 24 August in Europe.

Rock Paper Shotgun is the home of PC gaming

Sign in and join us on our journey to discover strange and compelling PC games.

In this article
Follow a topic and we'll email you when we write an article about it.


iOS, PS3, Xbox 360, PlayStation Vita, PC, Mac

Related topics
About the Author
Kieron Gillen avatar

Kieron Gillen


Kieron Gillen is robo-crazy.