Today’s issue of The Escapist features an article about the illusion of videogame interactivity. To quickly summarise it, the author begins by saying that games are differentiated from other art forms by their interactive nature, but then discusses how many emotionally moving games (the examples used are Half-Life 2: Episode 2 and Metal Gear Solid 4 – there are spoilers!) reach their emotional peaks during scenes that are barely interactive at all. Players are thrown into a particularly difficult situation, but heavy scripting ensures that enemies will always drop dead before they can finish the player off, or that bombs will always be defused with one second left on the clock. I think the implication is that interactivity has a time and a place, but game designers shouldn’t be afraid to take it away when it adds to the emotional force of the game. I don’t like the sound of this at all, but it’s worth talking about.

I think I can come across like a total purist when it comes to defining ‘games’ as an art form. I was always taught during my Economics degree that a game was a situation created by multiple agents, each with their own objectives and rules of action, and it’s something I’ve taken to heart when it comes to videogame design. The most contentious part of this definition is that single-player games cannot truly be considered ‘games’. This seems to upset a lot of people, but I do think it’s true – I would describe all single-player games as toys and puzzles; sandboxes to play around in and enjoy the results, or specific sets of challenges waiting to be overcome. The boundaries are not at all fixed, but they never involve truly strategic gameplay simply because computer-controlled agents cannot exhibit strategic behaviour. They are not intuitive, they cannot learn anything other than what they are told to learn, they are easily stumped by AI glitches that no human player would ever fall for. Defeating a computer-controlled opponent doesn’t require a true strategy, but simply a kind of ‘behavioural solution’ – learning to manipulate their predefined reactions as a kind of meta-game puzzle.

Of course, I would still call Monkey Island a computer game even though it has no strategic element, it’s just a different use of the same word. A computer game can be a game or a toy or a puzzle or something else – I like to think that the only essential features are that they must be played on some kind of computer and they must involve at least some kind of player action. Even in the case of a game like Sentient, where players can start a new game, put the control pad down and enjoy a (limited) gameplay experience while waiting for a solar flare to rip through the shuttle bay and kill them, inaction represents a concious choice of action on the part of the player. So by these standards, I think the gaming medium is very broad, and I’m always interested in new interpretations of the form (as anyone who’s heard me talk about Rhythm Tengoku can attest).

So anyway, I recently played through Metal Gear Solid 2 after many years of failed attempts at learning to sneak properly. It’s a game that is famous for being full of long cut-scenes – a combination of game engine cinematics, codec conversations and the occasional scripted gameplay sequence (such as using the directional microphone to listen in on conversations). It’s always tempting to go a bit Jon Blow and say that games like this are betraying the medium by dragging players out of the game proper, but I think there’s more to it than that. The thing I realised while playing MGS2 was that Kojima’s goal isn’t to create artistic games, but to create meaningful experiences, and he does it by combining different media. Some parts are interactive, others are cinematic, others are simply dialogue.

The storyline focuses on a plot to manipulate information and distort public opinion, with the new Metal Gear model functioning as an all-seeing, all-powerful internet filter. But not only must players confront the physical apparatus of the machine, they themselves must also cope with being manpulated by the political motivations of their information sources. Every character encountered in the game has their own personal agenda, often at odds with their affiliated organisation, and they all spoon-feed the player conflicting information about their identity, the game environment, the events taking place… almost everything the player thinks that they know. The strange thing is, the interactive portions of the game – in which the player strives to rescue hostages and destroy Metal Gear – generally serve to further the linear story, while the cut-scenes and codec dialogue – in which the player simply stares at the TV screen – seed the player with misinformation to manipulate their behaviour during gameplay.

It reminded me of Watchmen, which I saw at the cinema recently. I’m not going to talk about the film here, but one of the notable elements that had been removed were the chapter epilogues – the extracts from Under the Hood, the papers from Rorschach’s psych file, and so on. Most of the people I’ve lent the comic to tell me they skipped over those parts, but I think they add a lot of flavour to the story. Yes, they pull you out of the comic book format that Watchmen is supposed to be exploiting, but in doing so they give you fresh perspectives on the characters. They add to the story in ways that comic panels alone cannot, supporting them instead of diluting them.

I’ve come to see Metal Gear Solid’s cut-scenes in a similar light. Even those endless relationship chats with Rose have a purpose, to slowly and steadily establish your impression of Raiden as a useless, whining coward. It’s all pulled back into significance towards the end of the game, but it also serves simply to get on the player’s nerves. When the game was first released, everyone was – naturally – expecting to play as Solid Snake again, and Konami even relased doctored trailers full of cut-scenes with Snake in Raiden’s place. So when players completed the prologue Tanker mission and suddenly found themselves playing as this reclusive rookie called Raiden, they were shocked. It was another intentional plan to mislead players, feed them false information and manipulate their behaviour (such as manipulating them into buying more copies of the game?)

But the Escapist article describes a different phenomenon. I find these heavily controlled sequences of ‘fake interaction’ to be really dishonest. I can see why they would be very effective – just watching a video of the MGS4 example made me come over all funny, and I can see how it could be electrifying to play – but once the illusion is broken (as it has been, now) it just becomes a chore. Grand Theft Auto 4 made a total hash of this as the game went on. One of the later missions involves raiding an abandoned Sprunk bottling plant and killing a rival gang boss inside before he can escape to the roof. I threw everything I had into targetting him and taking him out as quickly as possible, but he always seemed to escape to the next room and I would always be killed by his surrounding goons. At one point, I shot him twice in the back of the head with a sniper rifle – enough to kill anyone under the established game rules – but the cinematic scripting had secretly made him invincible. Despite being repeatedly told to chase and kill him as soon as possible, the optimum solution is – perversely – to advance slowly and surely, killing his bodyguards from behind cover to preserve your health for the rooftop showdown.

Another example could be the player’s apparent invincibility in the new Prince of Persia game. Whenever the player suffers a fatal fall or is defeated in combat, their NPC partner Elika steps in to save them, and play continues. The idea was that, in this post-arcade world, players should be ‘beyond failure‘ – that with no need to eat up the player’s credits, there’s no real reason to punish them. Which is total rubbish of course, but I guess it was a pretty ballsy move. But what is the point of combat if you can’t lose? What is the point of a bottomless pit if you can’t fall in?

This is something that is going to bug me throughout HL2: Episode 3, I am sure. How much of my success is down to me, and how much is due to the game’s scripted influence? Considering how much work Valve have put into the Left 4 Dead‘s AI director, I wouldn’t be surprised if the whole game was designed to adapt to your playing style, dropping in as many enemies as it takes to slow you down and giving you extra supplies when you need a boost. I find the prospect a little depressing. If the hardcore gamer’s sense of pleasure comes from mastering the rules of the game, it seems unfair to let the game change its own rules during play.

Like most lies, these moments of ‘fake interaction’ might be appealing at first glance, but can wreak catastrophic damage once the truth is revealed. For an interesting counterpoint, there is a scene near the end of Shadow of the Colossus where the player is sucked into a portal, and can use their full range of moves to try – and fail – to resist being sucked in. But unlike the examples mentioned in the Escapist article, the player is expected to realise their destiny while the scene is still taking place. As they cling desperately to a pillar or a flight of steps, players watch their grip gauge – the most used, most important tool in the game – ebb away with sober familiarity, and they realise that they are simply delaying the inevitable. Despite all their adventures, all the colossi they have slain, they are still powerless in the face of the gods and wizards that oversee the game world, and Ueda expresses this using the same rules and interface that the player has grown familiar with.

This is more like the kind of interactive storytelling that I would like to see! Give players a sucker punch, pull the rug out from under their feet, but don’t lie to them! It trivialises the player’s actions and makes the whole game seem phoney. People rarely look back fondly on a time when they fell for a cheap trick.