Posted by Cumulomentis under
Warning: call_user_func_array() [function.call-user-func-array]: First argument is expected to be a valid callback, '_config_wp_home' was given in /home/.marla/cumulomentis/cumulomentis/wp-includes/plugin.php on line 48
Warning: call_user_func_array() [function.call-user-func-array]: First argument is expected to be a valid callback, '_config_wp_home' was given in /home/.marla/cumulomentis/cumulomentis/wp-includes/plugin.php on line 48
Design, Theory
[8] Comments
Emergence is something that seems to be getting talked about more and more frequently these days. It’s a movement towards unpredictable gameplay, where the systems, environments, and components that define the game react to how the player chooses to behave rather than force a type of behaviour on the player. The gameplay, in other words, ‘emerges’ as a consequence of the player’s actions.
That definition, however, makes me question whether games have achieved anything even remotely close to emergence, or if, considering the traditional nature of game design, they ever could. I call this problem ‘design intervention’ because I am too grandiose for my own good, but also because I believe that it sums up the antithesis of emergence: games cannot be unpredictable if they are designed.
What is commonly referred to as emergence these days is in fact non-linearity; it hinges on the degree of player choice and freedom, but the two should not be confused. If I, as a player, am in a locked room with a crowbar and a key to the door, that gives me two ways to get out, but can two designed choices be called emergent? I wouldn’t have said so. Now say I also have a screwdriver, and there’s a window leading out to a drainpipe, and maybe an air duct in the wall (we all know how prolific those things can be), would that be any more emergent? Some might say yes; I wouldn’t.
There doesn’t seem to be any difference between being asked whether I want to make a choice between options 1 and 2 or options 1 to 10. If each of those options is designed, the gameplay cannot be unpredictable and therefore emergent. It’s not reacting to my choices, I’m reacting to its options.
It’s perfectly possible for those options to get more and more complex, more deeply embedded within the mechanics of the game (using the broken door to smash the window, or two pieces of the splintered wood to make fire by friction) but even then, the depth doesn’t hide the designed intent; the physical properties of each component have still been selected at a high level to allow for those choices, and there are still many things the player will not be able to do if they have not been allowed for in the design.
This thinking also applies to artificial intelligence, which is sometimes held as being an excellent example of emergence. However, the logic still stands. The AI entities in the game have been programmed to react to the player and environment in a certain way, and as such can be predicted. Sometimes the AI is simple and barely duck to avoid being shot, and sometimes the AI is more complex and will ‘communicate’ with its buddies or have some rudimentary learning system, but even then it has been designed at a high level to do those things. Clever perhaps, but we will always know just how clever.
What this means is that having a creative influence at a high level (design intervention) seems to be completely at odds with the idea of emergence as defined at the start of this post. Emergence and storytelling can’t coexist in the traditional sense. In order for a game to have emergent behaviour it needs to be unpredictable by anyone, and so the designer would need to relinquish control of it. This is where it all begins to get a bit weird and theoretical.
Say we go back to the example of the room. As a piece of geometry designed for the sole benefit of the player, it cannot be called emergent because every aspect of it is contrived, but as a piece of architecture designed for use as an actual room (which the player happens to be in) the emergence can finally come through. Now if the player breaks a door, or a window, or a wall, it would be in spite of the designer, not because of him.
Will Wright and Introversion have already shown that with procedural generation, randomised but believable organisms, cities, and societies can be created with no high-level input at all if desired. Suppose those organisms were placed in those cities with those communities. Suppose those organisms themselves built the cities and formed the communities because of the depth of their generated personalities and goal systems. Suppose the organisms were not generated directly into their modern state at all, but were procedurally ‘evolved’ for the equivalent of a few billion years based on the environment they lived in during that time. And so on and so on until designers are creating the beginning of a procedurally generated universe which is simulated through the aeons to finally produce a planet and a continent and a city and a building and a room. Which is where you start.
Obviously all that is way way out in the realms of science fiction right now, but I was just using it to illustrate a point (also it makes my brain squeal with joy). By moving down through the levels of design into deeper and more fundamental principles (and leaving the higher levels to be generated by those principles) the designer is creating more and more possibilities for more emergent gameplay.
Of course, as I said before, this would completely throw traditionally crafted games out of the window. A story couldn’t be written because the designer wouldn’t be able to direct the player or create important high-level elements like characters, locations, and events. But who says that games need to be crafted in that traditional way? Imagine the possibilities of exploring and interacting with environments and situations that haven’t even been conceived of. With the procedural techniques that would be necessary to achieve high levels of emergence, more content and depth could be generated than it would be possible to create in a lifetime.