What's a Game?

[Walt Freitag]

Contention over the definition of the word "game" (as a noun) has been implicit in a few recent threads. Allow me to share a definition that I've been using, and refining, for a couple years now.

Disclaimer: The way I define "game" is a purely intellectual matter and is not intended to advance anyone's political agenda, least of all my own. Though the purpose of the definition is to allow categorization of activities as being, or not being, games, no such categorization in any way condones or condemns the activity being categorized.

Game (n): An activity or process, involving at least one participant (hereinafter referred to as a "player"), that meets three criteria:

1. The player has a goal.

2. The decisions of the player affect whether or not the player's goal is accomplished.

3. The determination of whether or not, or how effectively, the goal is accomplished is more important to the player than the direct consequences of pursuing or accomplishing the goal.

This is the broadest definition I've been able to concoct that still draws meaningful distinctions between games and other forms of human endeavor. By focusing on the nature of its goal rather than e.g. the element of competition or the existence of established rules, this definition encompasses such activities as solitaire games, cooperative games, extemporaneous games (e.g. Calvinball), and games in which player(s) select their own goals or change their goals during play. A person who decides to throw a rock at a tree to see whether he will hit it is playing a game, albeit a brief and simple one.

Each of the criteria is, I believe, reasonable and necessary. Item 1 differentiates between games and other forms and aspects of play that are not games. Exploring is not a game in and of itself, but it becomes one if there is something to be found or achieved by exploring (as long as there's also the possibility that it will not be found or achieved). An interactive storytelling session is not a game unless (as is usually the case) the participants are given, or self-select, goals.

Item 2 constrains the nature of the goal that makes the activity a game. Since all volitional activities can be said to have goals, but not all activites are games, further conditions are necessary. The goal in question must be one whose success or failure can be affected by the player's decisions. The existence of a goal that cannot fail, for example, does not make the activity a game. In particular, if the activity itself is held to be the goal, that does not indicate the activity is a game because the activity takes place (and that goal therefore succeeds) regardless of the player's decisions. Similarly, "passing the time" as a goal does not indicate a game because time will indeed pass regardless of what the player does. (Note that the mere existence of such a goal also does not prove the activity is not a game, because there may be another simultaneous goal that does meet the criteria.) There must be a goal for which success or failure are both possible, and the player's decisions must be capable of influencing whether success or failure occurs.

Item 2 also differentiates between games and performance or spectatorship, where a goal may appear to exist (e.g. the hero of a story prevails, or the home team wins) but the stage performer or sports spectator cannot influence whether or not it's accomplished.

Item 3 distinguishes games from endeavors that for want of a better term are better classified as "work." A runner running a race may be said to be playing a game, but a runner running to catch a train is not, because in that case the desire to catch the train, rather than the revelation of whether or not the runner is able to do so, is the primary motivation for running.

This distinction can become tricky, but it seems to hold up when tested. For example, a lawsuit doesn't normally fit Item 3 of the definition, because the parties involved care more about what money changes hands or the public vindication of the tort (that is, the direct consequences of winning) than about determining whether they're able to accomplish a win. But occasionally, the reverse may be true, in which case calling it a game seems justifiable. (Tragically, it could be a game for one litigant but not the other.) Part 3 of the definition is actually a very restrictive condition that can exclude many pursuits often described as games: war games for training purposes, stock market investing, pro sports when the player cares primarily about the money, sports played mainly for exercise, salesmanship, espionage, gambling, diplomacy, etc.

However, such borderline cases must be carefully analyzed. A "war game" exercise, for example, might be conducted for the purpose of assessing force readiness or analyzing a military scenario. The decisions of the players do not affect whether this goal is accomplished, because the assessment will be made or the scenario analyzed regardless of which side wins. Furthermore, the assessment or analysis is probably more important to the organizers than finding out who wins per se. Thus, it appears that the exercise is not a game by both Item 2 and Item 3 of the definition. However, the perspective that's important in the definition is that of the player(s), not the organizer or observer. For the participants in the war game, the main goal is beating the enemy; their choices affect whether that goal is accomplished; and they care more about winning the exercise than about what the brass will do with the information learned. Therefore, for the actual players of the war game, it is a game, even if it's not for the top echelons.

In may cases, of course, multiple motivations for performing the same activity coexist and must be weighed. Does that pro ball player want to prove his skills against the opponents' challenge, or is he just playing for the money? Is that stock trader really so rich that for him the money is "just a way of keeping score" as he claims? In the spirit of GNS, it seems reasonable to assert that even though different motivations may coexist, in the end only one of them can be the most important to the player, even if it's not apparent from the outside which it is until a whole pattern of behavior is observed. So an activity always is or is not a game under this definition, but it won't always be easy to determine which. Which agrees with our real-world experience. Since it appears intuitively that the exact same activity performed in the exact same way with the exact same results (e.g. hunting for a rabbit using a shotgun) can be a game for some people and/or circumstances but not for others, some subjectivity based on the participants' inner motivation appears to be necessary.

What do people think? Is this definition viable? Comprehensible? Useful? How could it be improved?

- Walt

[xiombarg]

Bear with me, I'm thinking out loud here, so I may ramble.

I have a couple problems with this definition.

First, and most of, all, what about fun? I know Jared has said RPGs don't have to fun, but isn't that what makes it a game, and not work.

As an example, if someone comes up to me with a loaded pistol and forces me at gunpoint to dig ditches, telling me I must dig one 2 foot by 3 foot by 6 feet deep ditch every hour or he'll shoot me, well, what I'm doing is a game, right?

I'm participating in an activity or process. That makes me a player, right?

1. I have a goal. Dig a 2 foot by 3 foot by 6 feet deep ditch every hour, until the guy with the gun tells me to stop.
2. My decisions affect whether or not the goal is accomplished. I can choose to dig or not to. I decide how to pace myself, since I don't know how long I'll be doing this. And so on.
3. The determination of whether or not I accomplish the goal is more important than the consquence of persuing or accomplishing the goal... If I accomplish the goal, well, very well, I keep playing. It's the consequence of *losing* that I'm concerned about.

Okay, not the best example, perhaps, but I think point 3 is the slippery slope. If I play chess for money, and the direct consequence of the act -- getting money -- is more important to me than the act itself, does that mean chess isn't a game? I mean, how do you KNOW that 3 is the case? Isn't a lot to assume you can read the mind of the player to determine his intent? Shouldn't the criterion be more objective?

[Jared A. Sorensen]

A game is a game when there is a conscious effort/decision on the part of the player to MAKE it a game.

So your digging a ditch example might be a game, but probably only for the guy with the gun.

I can't define the word/concept of game by myself. But I do know in my heart (FWIW) that this conscious decision is required. This is why I don't think simulationist RPGs are games.

The first question is what do you do?
And when the exploration of character/setting is brought up as the answer, I have to ask...well how do you know when you're done?

[Mytholder]

I can't define the word/concept of game by myself. But I do know in my heart (FWIW) that this conscious decision is required. This is why I don't think simulationist RPGs are games.

The first question is what do you do?
And when the exploration of character/setting is brought up as the answer, I have to ask...well how do you know when you're done?

When you're bored, or the character is dead or otherwise unplayable, or when whatever aspect of the game world you were interested in is gone.

[Le Joueur]

A game is a game when there is a conscious effort/decision on the part of the player to MAKE it a game.

I can't define the word/concept of game by myself. But I do know in my heart (FWIW) that this conscious decision is required. This is why I don't think simulationist RPGs are games.

Even when those partaking of the Simulationism "MAKE it a game," even in the absense of competition or Premise?

Is the problem that what is given as the definition of Simulationism is incomplete?  This would mean that some Simulationism is game and some isn't and that the definition of Simulationism is simply too vague to exclude 'non-game' Simulationism?  That, I could buy; otherwise I get the 'baby (Simulationist games) is going out with the bathwater (non-game Simulationism)' feeling.

I'm more than a little curious about this 'Simulationism is not gaming' idea.

Fang Langford

[Mike Holmes]

I'm not sure what all the fuss is about. Let's say that Simulationism is not a game. OK, it's still an activity that I want to partake in. Reading a book or watching a movie is not a game, but I still like doing those. And I put way more copnscious decision making into a Sim RPG than those media. Might not be the types of decisions that some people are interested in, but "left or right" is interesting to me. This is so much a matter of preference that I wonder that we're still discussing it. Some like the Sim activity, some do not.

So who cares? Is it that Sim RPGs are labeled "Games" perhaps incorrectly? Seems like a pointless argument.

Mike

[joshua neff]

I have to say, I agree with Mike.

Let's say, for argument's sake, that Simulationist RPGs aren't, strictly speaking, "games". Let's say that Exploration of Character, as a goal in & of itself, isn't really "playing a roleplaying game".

...And then what? I mean, really, what's the goal here? What is there to gain?

On the other hand, let's say we definitively state that Simulationist RPGs are "games", & Exploration of Character is a completely valid form of roleplaying, right up there with "expanding on premise" & "use of resources to achieve victory". Again, what then? What do we gain by categorically deciding one way or the other that Sim is or isn't a "game"? I like theory as much as the next guy, but isn't this kind of verging on "vague intellectual exercise"? I thought the whole point of the GNS thing was "to make play better". Okay, so now there are people (like Mike & Mytholder) who can say, "Hey, I like that whole Exploration of Character thing. Woohoo! Lookee me, I'm a Simulationist!" Now we're going to turn around & say, "But that's not a game!" Um...okay...how is this making play better? How do my RPG experiences improve by deciding that a form of recreation I'm not even all that into (Simulationism) isn't a "game".

For the record, I think that Simulationism is a game. People predisposed to, say, Exploration of Character are entering into the activity with the set goal of "getting into the character's head & seeing what develops". Does the player have a concrete stopping point in mind? Maybe not. Not all games have concrete stopping points--sometimes, the goal of the game is to keep the game going as long as you can.

So, there. I've defined Simulationism as a game. One that I'm not really into, but a game nonetheless. And now what?

[Ron Edwards]

Hi folks,

In my essay, you'll see that the term "game" plays no substantive role. Although I have no particular objection to Walt's definition of the term, I still don't see any point to including it in the discussion in a formal way.

To me, the issues break down as:

Biggest category: Social leisure activity
Subset: shared Exploration of the five "things"
Sub-Subsets from there: individual and group goals (GNS)

I perceive (and use) the term "game" only for historical reasons. On a related note, the term "Gamist" is derived from "game" only etymologically, not substantively.

Best,
Ron

[contracycle]

First, and most of, all, what about fun? I know Jared has said RPGs don't have to fun, but isn't that what makes it a game, and not work.

I don't think thats relevant - not all games are fun, and some games are fun but have a decidedly useful purpose in the big picture.  

As an example, if someone comes up to me with a loaded pistol and forces me at gunpoint to dig ditches, telling me I must dig one 2 foot by 3 foot by 6 feet deep ditch every hour or he'll shoot me, well, what I'm doing is a game, right?

It is by my way of looking at it, yes.  It is simply not a fun game.

I think the important thing it is that it is BOUNDED - a game is a boxed off chunk of reality that operates under certain specific and understood rules.  It establishes certain forms of legitimised, expected or validated interactions between the participants, and the participants consciously use these within the bounds of the game.  To this extent, a game is at least partially predictable, in that cause and effect have been strictly delineated.

I don't even think that this is limited to creatures with human rationality - Fetch is a cross-species game played by humans and dogs.  Catch The String is a cross-species game played by humans and cats.  Whose Fault It All Was is a game played by divorcing couples.  The only distinction between game-as-behaviour, and game-as-entertainment, is the stakes.  Fetch The Stick is a low-stakes game which is itself training for a higher-stakes game of Fetch The Dead Bird From The Marsh - which might at times be a matter of life and death for the bird, the dog, and the human, and hence have very high stakes indeed.  In the meanwhile, Fetch The Stick can be fun and rewarding in its own right.

[AndyGuest]

Game (n): An activity or process, involving at least one participant (hereinafter referred to as a "player"), that meets three criteria:

1. The player has a goal.

Um, does the goal need to be significant ? Do I have to be actively pursuing the goal ? Does it need to be related directly to the game ?

If I am playing pool with my mates on a Saturday afternoon, and me reason for playing is to be with my mates and I don't care if I win or lose, am I still playing a game ?

2. The decisions of the player affect whether or not the player’s goal is accomplished.

So if it is totally random there is no game ? Snakes and ladders is not a game, not even a bad one ?

3. The determination of whether or not, or how effectively, the goal is accomplished is more important to the player than the direct consequences of pursuing or accomplishing the goal.[/color]

So a family playing 'I Spy' on a long car journey to kill time are only playing a game if they are keeping score ? Not if they are just doing it for fun or to kill time ?

I guess my problem with your definition is that it is simultaneously too broad and too narrow ;-). It includes things which can be considered games but are not inherently games ('I go to work to earn money' seems to fit all rules). Likewise some games fail under the description - Snakes and Ladders is a game (a bad one I'll grant you, but a game nonetheless).

[Walt Freitag]

"Game" is a complex concept with many borderline cases. The test I apply to a definition of "game" is not whether or not it magically resolves all borderline cases without controversy -- which I don't believe is possible -- but whether it provides the tools for analyzing such cases, including understanding why they are borderline.

For those who have suggested alternatives, I'm struck not by the contrast of the alternatives but how similar they are. Each includes some criteria for distinguishing games from activities that may involve the same challenges and procedures as games, but have a different motivation behind them. Thus, whether or not the game is "for fun" or whether or not it is "recreational" or whether or not the participants consciously decide it is a game. If you were to attempt to further define "fun" or "recreation," however, I believe you'd end up with something like what my definition already includes, a weighing of motivations between the intrinsic and the extrinsic rewards. (Including in a definition of "game" the criteria that the participants consciously regard the activity as a game is unfortunately circular. It implies that no objective definition is possible... which I agree with in part, but I think the areas of subjectivity can be usefully narrowed down.)

Does the goal need to be significant?

No. How significant is getting a small ball to go into a small hole, especially when you intend to immediately take it out again? Including a vague word like "significant" in the definition would be counterproductive. Significant to whom? In what way?

Does the player have to be actively pursuing the goal?

Not necessarily. But if a player is not actively pursuing the goal, the activity is unlikely to meet criteria 2 or 3.

Does it need to be directly related to the game?

Goals not related to the game will be irrelevant when criteria 2 and 3 are considered, so this is implicitly true.

If I am playing pool with my mates on a Saturday afternoon, and me reason for playing is to be with my mates and I don't care if I win or lose, am I still playing a game?

Yes. Look at it item by item. You have a goal, getting balls into pockets, whose success and failure are contingent on the decisions you make. The determination of how well you do so is more important than the direct consequences of doing so, assuming you're not motivated by a strong preference for spherical objects being safely contained in pockets rather than openly exposed on flat felt surfaces. The issue of being with your mates is not relevant either way to the question of whether or not the pool is a game. Being with your mates is not a direct consequence of your pool play, nor is it a goal whose success or failure your decisions will affect. You were already with your mates when you started playing pool, and they won't disappear or reject you if you fail to sink balls.

However, in the seemingly similar case of a businessman who plays golf in order to be able to meet with golf-playing business contacts, who has little or no concern with the game itself, I would conclude that it is not a game for that person. Being able to meet with contacts is a direct consequence of actively pursuing the game of golf, attempting to get balls into holes, because if the businessman doesn't at least make an honest attempt, he will not be permitted to meet with the contacts in that venue in the future. For him, golf is part of his work. (And even the IRS agrees that the expenses incurred in playing golf to meet business contacts are legitimate business expenses. I defer to them as the ultimate authority in distinguishing between work and recreation.)

Is it a game if it's totally random? Is Snakes and Ladders a game?

Again, let me repeat that the important question to me is not whether or not Snakes and Ladders is technically a game, but understanding why the issue is in question. The reason in this case is clear: if the decisions of the player do not affect whether or not the goal is accomplished, then it's not a game, but it's difficult to perceive any manner in which a player even makes decisions in Snakes and Ladders.

The question is even more important in gambling games. In one sense one could say that my decision to put my money on red or black makes all the difference in whether I win or lose. But in another, one could say that it's irrelevant because the decision has no effect on my chance of winning or losing. Similarly, how hard or how high you throw the dice during Snakes and Ladders does determine what number is rolled and hence whether you win or lose. But on the other hand, it's still totally random regardless of the decision.

I'm inclined to give the gambler and the toddler playing Snakes and Ladders the benefit of the doubt. Their decisions are determining the outcome, even though a random mechanism intervenes and makes it impossible for them to truly direct those decisions toward a particular outcome. Of course, most adults do not play Snakes and Ladders, nor would they play roulette very often if money weren't involved. But there's no reason not to call them games. (Note that this may disagree with the game theory definition of games, which is also highly technical and not very useful for our purposes).

So a family playing 'I Spy' on a long car journey to kill time are only playing a game if they are keeping score ? Not if they are just doing it for fun or to kill time?

No. This would be true if the only goal you consider is winning by getting the most points, but there is also the goal of guessing what object the person spied (or, for the "it" player, choosing a spied object that the others will have difficulty guessing).

I think I see what you're suggesting: don't the direct consequences "having fun" or "killing time" outweigh "the determination of whether or not, or how effectively, the goal is accomplished" in a vast number of cases as a motivation for play? That just twists the definition. "Having fun" or "killing time" are likely reasons why the determination of whether or how effectively the goal is accomplished is important to the player in the first place. So no, those "goals" cannot be more important than themselves.

'I go to work to earn money' seems to fit all rules.

No, it fails miserably at rule #3. An easy way to test whether or not the pursuit of the goal motivationally outweighs the direct consequences of pursuing or achieving it is to take away those direct consequences and see if the player would still be interested. Would you still play Monopoly if at the end of the game, all your houses and hotels and deeds and hard-earned cash are just taken off the board and put back into the box? Of course, since that's the normal event. Would you still play an evening of poker with your pals if at the end of the evening, all the money was restored to its original owners? Probably, though you might wonder why you bothered with the poker and suggest just watching videos next time. Would you still go to work if you didn't get paid? I'll leave that one up to you.

- Walt

[joe_llama]

I totally agree with Walt on this one. His definiton of 'game' is the best I've seen so far. My game design theory uses very similar ideas in explaining what games really are. (Side note: I promise to unveil my theory as soon as possible).

I've been waiting a long time for someone to bring up the subject naturally. Our communication environment in the world of game design is still too vague, IMO. I mean, they don't tell you what to paint in art school but they teach you many tools to express yourself. So why not do the same with games? And to start developing these tools, we need to know what the hell games are in the first place.  
 
I feel it is imperative that we seperate the term 'game' from other words such as 'fun' 'activity' and 'play'. No matter how related, they mean different things. Somewhere along the way they got jumbled up. This leads to terrible miscommunication which I believe slows down designer creativity and blocks many paths of development. Whether the GNS model deals with the term 'game' or not is irrelevant.

It's true that the difference between a game and a project, a work, or even real life is at best semantic. But that's the whole point: a game is a simulation of a dynamic problem with a reachable solution through challenge. The whole point about games is problem solving under specific circumstances and parameters.

Games are very human. Our mind is designed to be a problem solver. Games help test and train the mind and society in hypothetical dangers and successes. This is also true for dogs, cats and other similar 'lower' lifeforms.  

As for Simulation, it is merely a design pattern. It is the designer's choice to simulate a specific environment but it serves the desginer's goal and not the player's goal. The player's goal provides the focus to the game. Otherwise, the game goes wild in all directions. It could be an 'interesting' and 'fun' activity but it will not be a game. I believe that some of the 'Player Dissatisfaction Syndrome' we see in RPG's is because these "games" not only provide rules which fail to express the experience desired but also that these rules lead to nowhere!

In the same notion, I liked 'Mulholland Drive' because it provoked my thoughts but I'm not surprised that 80% of the people watching it hated it as hellspawn. These people aren't dumb or prejudiced - they were given a problem and they want a solution. Same thing with all other art forms. What we like about 'abstract' or 'symbolic' art is the ability to interpret it in many ways. Our mind tries to solve the problem but can't. This is why it is also hated by many - it leads to confusion and frustration, which may also be unhealthy to some of us.

Simulation is legit but it's not a game, so please don't write on the box that it is.

With respect,

Joe Llama

[xiombarg]

Simulation is legit but it's not a game, so please don't write on the box that it is.

Except by the English definition of a game, it is a game. It's only not a game according to the definition above, which is a highly-specialized subset of the normal English definiton of the word.

I can create a definition of "TV Show" that excludes sitcoms, and decide not to call them TV shows, while respecting them as "fine broadcast material, just not TV shows". However, I don't think you should get upset when other people call it a TV show.

I mean, you're not really defining "game" here, per se. You're defining another word you want to use in the place of the word "game", because "game" is too vague. I understand this. This is why I don't use the word "nice". It's too vague. But if you're going to do this, you need it to be a distinct word, just like "helpful" is distinct from "nice" but a helpful person is generally considered nice.

[AndyGuest]

Would you still play an evening of poker with your pals if at the end of the evening, all the money was restored to its original owners? Probably, though you might wonder why you bothered with the poker and suggest just watching videos next time.

Leaving aside that this is the only way I play poker, I hate the idea of losing money in a game ;-).

Would you still go to work if you didn't get paid? I'll leave that one up to you.
- Walt

Hmm, that to me suggests that there is no such thing as an entity/activity that is inherently a game, there are merely activities which may or may not be a game depending on the motivations of those involved ?

By your description golf isn't inherently a game, likewise if I choose to go to work for fun, without being paid, does it then become a game ?

Also can the goal of a game not be the pleasure of playing ? When I take part in sport (a rare occurence) I am just there for fun, I don't care if I win or lose, taking part is what matters (bleagh that sounds sickening but its true), the only thing that spoils my fun is people taking the sport too seriously and getting upset about it. So for me a game doesn't require a goal in terms of an end-point, the game itself can be the goal.

Dragging this back to the original source, I consider sim RPGs games where the goal is the process of playing, the endpoint isn't that important.

[contracycle]

Hmm, that to me suggests that there is no such thing as an entity/activity that is inherently a game, there are merely activities which may or may not be a game depending on the motivations of those involved ?

Thats pretty much how I see it.  IMO, there is a subset of games which exist for entertainment; this is what we conventionally refer to as "a game", something fairly trivial.  But that, quite literally, is just the tip of the iceberg.