What is a game?

Back from the Games Developers' Conference in San Francisco, Raph Koster weighs in on the perennial debate about what makes a game, starting with his own classic formulation, "Playing a game is the act of solving statistically varied challenge situations presented by an opponent who may or may not be algorithmic within a framework that is a defined systemic model."

Some see this as a “fundamentalist” approach to the definition. But I use it precisely because it is inclusive. It admits of me turning a toy into a game by imposing my own challenge on it (such as a ball being a toy, but trying to catch it after bouncing it against the wall becoming a game with simple rules that I myself define). It admits of sports. It admits of those who turn interpersonal relationships, or the stock market, or anything else, into “a game.”

Basically, I see this whole issue as this Venn diagram. I group sports, boardgames, and yes, videogames under “game” because all of them are susceptible to being broken down and analyzed with game grammar. They all have rules. They all have an “opponent” that presents varied challenges to the player. They all have a feedback loop. They all present verbs to the player. They all present goals. The fact that some use real-world physics as the opponent, some use the human body, some use dice or boards, and some use a computer is not even relevant when you break down the game atoms. I can group these easily because I’ve dug into them so much that I know there is something there in common.

“X” isn’t a game!


  1. I like Raph, but this definition just seems too limited – substitute ‘work’ for ‘playing a game’ in the statement above, and it makes as much sense. 

  2. Anybody remember that ’80’s staple “Games” magazine?  Along with all the simplistic, fun stuff, they also offered gaming theory and brilliant letters to the editor as well as some truly difficult brain benders.  Artistically tricked out envelopes for submissions also became a competition.

  3. In my social circle we have a long standing debate over whether Curling is a sport or just a game. It’s gotten to the point where each new member to the group is expect to take a side.

    After Sunday night’s ambulance ride to the emergency room, I feel confident that an injury of that degree is strong evidence of it being a sport but some may still disagree.

  4. What is a game? Well, based on this diagram I can tell you what it’s not: it’s definitely not interactive art (unless it’s a video game, which s the only sort of game that is also interactive art).

    1.  I think games in general are an art form. All games of which I’m aware are interactive. Therefore I think all games are interactive art.

  5. I still think games can be classed as storytelling/general entertainment without necessarily falling into the art category. I mean, CSI is not “art” but it does entertain certain swathes of the populace at times. There’s room for more high brow entertainment simultaneously of course. The latest Bonus Round over at GT discusses the role of action in games (specifically violence), and it put me in mind of all the varied genres we have in other forms of media that we don’t yet have in games. So it may not fall under the traditional guise of a game, but I think there’s scope for something like “Downton Abbey” which perhaps has only moderate to few action, and still call it a “game”…

      1. I suppose so, in the broadest sense.
        I just think of art as one thing and entertainment as another. Myself I see games more as entertainment, but it doesn’t have to be binary. There’s room for everything in between as the appeal widens and broadens.

  6. I wouldn’t define video games as the intersection between games and interactive art. There are video games that aren’t art, and interactive art that takes the form of board games- Brenda Braithwaite’s Train being a particularly powerful example of the latter…

    1.  See also Carcassone, where the whole point of the game is to create a beautiful, unique landscape with each play.

  7. The definition is far too open. Buying stocks or being a weatherman can certainly be considered “the act of solving statistically varied challenge situations presented by an opponent who may or may not be algorithmic within a framework that is a defined systemic model.” And while you can make buying stocks or predicting the weather into a game, it would be silly to say that those things are by definition a game.

    That definition basically comprises any act of solving challenges.

    It is also too narrow. Playing a game of Snakes and Ladders is in no way “the act of solving statistically varied challenge situations,” yet it is still a game.

    Basically defining a game is the equivalent of squaring the circle. People are always trying to come up with new definitions, but each new definition will always be found to be lacking.

  8. I think Jesse Schell has done a great job of thinking about defining activity vs toy vs game vs etc.  And I like his definition: “A game is a problem-solving activity, approached with a playful attitude.”  Of course, that simple definition comes only after working through no less than 10 properties shared by games, and defining what we mean by play. (And also the nuance of problem-solving! and activity!)  The Art of Game Design is a fantastic book.

  9. I’ve stated a few times in comments here that I think games are art, but I feel as though I should clarify my position a bit.

    It’s difficult to find a cogent game-theoretical definition of “a game” on the Web, so  I’ll make a stab at it from memory: A game is a series of one more events in which one or more players acts according to a ruleset. These events are called “moves”, and a set of moves which modifies some status attached to a player (called an “imputation” in Theory of Games and Economic Behavior) is called a “strategy”.

    Starting from this, I can guess that it might be possible to construct a program that would algorithmically generate rulesets (which one might call “games” even when they aren’t actively being played, just as a game of checkers is still a game while in the box). I wouldn’t call these “games” art, necessarily. The design of the constructing algorithm might be art.

    Games which are invented by human beings are art, even if they are crap art.

  10. I love how “board games” fall outside of “interactive art” even though there has been many artists who have already used board games in artistic ways. Further, if one took subjective sports like diving, ice skating, or ballroom dance, it should be relatively easy to see that sport can very well be included in a concept of “interactive art”. 

    Mr. Koster works on video games. Possibly with the great experience and expertise he has here it makes it easy to see the art. Where a lack of expertise in the histories and practices of art and sport might not give him the expertise in those areas to see the artistry. Further, many people don’t realize that interactivity does not require an electronic computer. 

    Mr. Koster’s “fundamentalist” approach is leads to an orthodoxy which actually limits the very subversive aspect of play and art. It is similar to fights over what is within any genre in fiction. It is a very misguided fight in all realms. 

    We live in an age of complex databases that allow for the ability to sort into groups based on many criteria. This chart regresses back into an age where only one taxonomy can be valuable. It is this desire to use stereotypic shorthand that causes prejudice in one’s thoughts against better solutions whether it be in games, art, fiction, or life. It is the myth of a platonic ideal especially when it comes to human creations.

    This is not useful. It is a procrustean bed that it is no fun to sleep in.

  11. I was not saying that board games could not be art. I should have labelled the red oval as “digital interactive art” (image updated in the original blog post now).

Comments are closed.