What is a game? A standard definition is “a form of competitive activity or sport played according to rules” and this has been the working conception when I’ve encountered theoretical engagements with the notion of a game. But a recent symposium on eSports left me reflecting on how much more complex the ontology of games is when we consider contemporary video games, raising the question of whether digital games, particularly those played online, are something entirely different from their analogue predecessors.
Consider how a game like poker has developed over time. This family of card games has a contested history, with many potential predecessors being claimed. It has also has many variants, with rules that are stabilised through a range of artefacts, from ‘how to’ guides through to cultural representations and rule books in tournaments. As much as these artefacts exercise a normative influence over how poker is played, it’s predominant mode of transmission is interpersonal, with changes in the game liable to be piecemeal and taking place over long periods of time. In contrast, the rules of online digital games can be changed at a moment’s notice, with these being an important vector through which the relationship between the developer and the users unfolds. Every game has an infrastructure that supports it, even if it as minimal as conversations that have previously taken place between different groups that play the game. But the infrastructure of digital games played online allows for granular analysis of game events and immediate modification of the game rules. These might impede the reproduction of the game, for instance if too many rule changes alienate players, but the capacity to make these changes is something new and interesting.
There are also differences at the level of the virtual structure of the game: the latent order through which events unfold, driven by the rules of the game, but producing patterns which inevitably exceed what could be predicted from those rules alone. The complexity of digital games vastly exceeds that of analogue games, perhaps in a way which renders it impossible to render them formalistically in terms of branching probabilities. This isn’t always the case, particularly with older games which aren’t multiplayer. For instance I find it difficult to understand how something like this speed run of Super Mario 3 is possible unless there is, in principle, a ‘correct’ move to make at every point in the process, even if it doesn’t involve adherence to the formal rules of the game:
But more complex games, particularly those in which many players compete online, would seem to be a different phenomenon altogether. However is the challenge this poses ontological or epistemology? Is there no underlying (virtual) structure or is it simply too complex to be mapped? I find the former claim untenable because in principle it seems obvious to me that any particular instance of the game could be analysed, with sufficient data, in order to explain why it unfolded in the way they it did. This presupposes a structure in relation to which those outcomes become explicable. In which case, the problem is epistemic and perhaps suggests that other methods, perhaps data scientific ones, might be necessary. With enough data could the contours of such a virtual game structure be fallibly traced out, even if it resists analysis through other means?
Mark Carrigan 