Games don’t have goals, people have goals.

Click image to view. ~1hr sketch. Responds to clicks and taps. Best in webkit browsers (Chrome/Safari).


The term ‘game’ is slippery. It has many ‘senses‘ or meanings that depend on the context of word’s use, e.g. “Hand me that game,” “Did you watch the game last night,” “I’m programming a game,” “We’re playing a game.” Two important senses are game-as-activity and game-as-artifact (or, more specifically, game-as-system). However, the notion of goal—frequently noted as a characteristic feature of a game in whatever sense—becomes problematic in the case of game-as-artifact/system.

In the first sense a game is a certain kind of goal directed activity involving certain kinds of equipment. So, to say that Settlers of Catan “is” “a game” is to say that Settlers of Catan is a certain kind of goal directed group activity involving conventional interpretations of various actions, a certain set of equipment,  and  contextual needs or regularities (e.g. sitting around a table for an uninterrupted period of time while protected from the elements). In such a game, people are trying to achieve a condition they judge to be success and avoid conditions judged to be failures. In this case “playing Catan” means participating (or perpetuating) the Settlers of Catan activity.

In the second sense, Settlers of Catan is a game in that it is an artifact (of sorts) with a certain set cause and effect relationships and their entailed behaviors. That is to say it is system. Obviously, none of the behaviors in a board game hold without the people moving the bits around the table, but we could argue that the operation of the game’s causal relationships is irrelevant, the behaviors—such as paying X resources to gain Y building—can be just as easily performed by a computer while maintaining a kind of formal or logical consistency between. Thus  Settlers of Catan ‘exists’ by virtue of these specific systemic relationships, however they manifest.

I take it for granted that the systems perspective is useful for both design and analysis, but there’s an interesting catch: however specifically we might point to the mechanical relationships in, say, a pinball table, there is no way to locate a goal state. There is nothing about any specific configuration of the artifact-system that qualifies it as being a goal state. Implicit in the systems perspective is that the game is a game by virtue of its internal or structural or mechanical qualities in a way that’s independent from people’s thoughts on the matter. But goals are normative; they rest on judgements about the rightness or wrongness of some act, event or condition. Yet there seems to be nothing in a system’s behavior that is sufficient to distinguish a configuration of that system as a goal state; we must resort to a judgement (either the designer’s or the players’).

This may seem weird because the goals in most video games appear plainly obvious in some way: score displays, win/loss screens, literal statements about goals etc. But in each of these cases, the item is only representative of a goal in the case that they are ‘correct’—they are only representative of a goal if they are judged to correctly represent a goalIf the computer reverses the win/lose screens, saying a player lost when they won and that they won when they lost, it would only be reversed with respect to what the player judges to be a win or a loss. There’s nothing to say, for example, that a version of Tic-Tac-Toe could be such that getting three in a row is a loss condition. There’s nothing to say that both players win if neither gets three in a row. There’s nothing to say that any condition can’t be a win or loss condition for that matter. Nothing about the brute mechanics of the system makes the reversed messaging incorrect because nothing about the brute mechanics of the system ground a goal state in the first place.

What then are these indications? If we open up our analysis to include human judgement, these items become small persuasive devices, suggestions or arguments for how the system should be operated.

With this notion in hand, we can consider alternative ways for making these goal-arguments, and we can consider how features of existing game make these arguments (intentionally or not).  Moreover, we might consider every game a ‘sandbox game’, but with distinctions amongst the force or form of their goal arguments.

3 Replies to “Games don’t have goals, people have goals.”

Leave a Reply

Your email address will not be published. Required fields are marked *