Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
October 24, 2014
arrowPress Releases
October 24, 2014
PR Newswire
View All
View All     Submit Event

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

What is a game and why do we play?
by Nils Pettersson on 01/16/13 02:44:00 am   Featured Blogs

The following blog post, unless otherwise noted, was written by a member of Gamasutra’s community.
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.


Starting by nailing down such an elusive concept as what a game is can be a daunting project, especially if you aim at creating an all-encompassing definition fit for all situations and people. I believe that I have found a definition that helps me immensely during my design sessions. So bear with me as I attempt to explain this definition.

All creatures play when they are young, this much is apparent through simple observation. The question then becomes:

Why do we play and what purpose does it fill?

We need to figure out the answer to these questions in order to define what a game is. Most if not all designers I know are driven towards creating the most immersive, fun and interesting game possible. A game that will be remembered for years and years for its brilliance. This is certainly very ambitious but if we do not have the answers to the above questions we are just swinging blindly.

Most people can agree that we play because it is fun. But why is it fun? Fun is, after all, just an emotional response to something we are doing. It is a combination of different chemicals released in our brains in order to promote positive activities.

I am a great fan of Raph Koster and especially his book A Theory of Fun. If you haven't read it yet I highly suggest you do it now. It is, in my mind, essential reading for any game designer. Koster argues that:

"Fun arises out of mastery. It arises out of comprehension. It is the act of solving puzzles that make games fun. With games, learning is the drug." - Raph Koster

Mastering a skill might take a lifetime but it is time well spent. 


This is by no means a new idea, and we are supported by several other great thinkers who have reached the same conclusion:

  • "The most effective kind of education is that a child should play amongst lovely things" - Plato
  • "Play is the child's most useful tool for preparing for the future and its tasks" - Bruno Bettelheim
  • "Play is the highest form of research" - Albert Einstein

Our brains are hardwired learning machines, this is what drives us into playing. Learning gives us the tools we need in order to overcome the challenges that are inevitably part of life. Overcoming challenges gives us a sense of achievement and makes us feel good about ourselves, which creates the feeling of fun. A good game is a game that constantly challenges the player and forces her to learn.

Our definition of a game must therefore include, in some manner, that learning is at the core of why we play.

But just figuring out why we play games is not enough. If a player is not guided by the game it is inevitable that she takes the path of least resistance since that is human nature. How can we as designers prevent the players from bypassing the challenges we create and in doing so bypass the entire learning process that is central to playing? If we let the player ignore the challenges she will not learn anything and the game will not be considered fun.

In order safeguard against this it is important to be able to control what challenges are presented and what methods are available to overcome them. To achieve this we need to consider rules. Since I come from an academic background this is the time to use a couple of terms that I find incredibly useful when discussing games.

These terms are widely used but my first acquaintance with them came from the book Man, Play and Games written by Roger Caillois. This is another book I cannot recommend highly enough. These terms are:

  • Paidia - Unstructured and spontanteous activites
  • Ludus - Structured activity and explicit rules

It is easy to see where these two terms fit into our society at large. Paidia is what we associate with childrens' games and playing. It is about improvised, creative playing were rules are implicit and changes rapidly as the game progresses. 

Ludus on the other hand is perhaps most associated with sports, but all organised games fall into this category. Ludus is about playing within explicit rules towards a clear goal. It has the advantage that the creator can quite easily control the play experience since all players must abide by the same rules. 

Based on the facts that:

  1. We play games to learn
  2. It is human nature to take the path of least resistance
  3. Ludus has explicit rules at its core

We can draw the conclusion that in order to guide the player within the game, to make sure that she encounters the challenges we present and does not simply bypass them, we need to stick with ludus. Our game must have enough rules to limit the player's actions while still allowing for a learning process. As such we cannot simple create rules that are so strict that the player will always act in the exact same manner. There must be room for error and improvement.

Sports is a typically associated with ludus.


We are supported in this conclusion by child psychologist Bruno Bettelheim who clearly states the difference between play and games:

  • Play - "Freedom from all but personally imposed rules, no goals."
  • Games - "Externally imposed rules, goals." 

Our definition must include ludus in some way since we are defining game and not play (which would be paidia).

A regular mistake when creating a definition is to focus solely on a factual specification of what something is from a purely techincal standpoint. This completely ignores that all things exist in a context, and in my opinion that context must be addressed by the definition. In my opinion a game is not so much an artifact as it is an activity. A game has no meaning until it is played. When it is played it serves as a medium for the player to learn. The act of playing and learning is an activity. This is what I mean with a game being an activity rather than an artifact.

Now that we have the technical specification, the context and the purpose we can create our definition. I suggest the following:

A game has explicit rules and goals, and is played with the primary purpose of learning.

This definition neatly ties together what we have discovered about games and the purpose behind playing. In order to test this definition to see if it actually works on a real example we need to look at an existing game. A definition must be useful in a practical context after all.

Let us look at Skyrim developed by Bethesda. A game widely praised for the amount of freedom granted to the player to creatively explore and generally do as she wishes.

  • Skyrim is set in a huge world where the player is free to wander as she wishes as long as she stays inside established boundaries. - This explicitly states the arena of the play area.
  • Characters can be developed and improved in a variety of ways in order to meet challenges and customise the way the player plays the game. - This gives the player choices about how to tackle said challenges while still staying within the confines of the rules.
  • The player has freedom within the rules to seek out and overcome challenges via creative thinking and trial and error.

The illusion of freedom.


Skyrim gives the player the illusion of choice that feeds her creativity while still guiding her along the intended experience. This illusion of choice is somewhat of a holy grail to game designers and something that is increadibly hard to nail down in a game. Everything in the game is controlled by explicit rules that cannot be broken since it is a digital game (computers are notoriously bad at independent thinking after all).

We can see that Skyrim fits into our definition neatly. If you choose to test the definition on other games you will find that what you typically refer to as games fits into the definition while other activities leaning more towards paidia do not.

If I were to give any practical advice from this largely abstract discussion it would be that if you wish to trap a player inside the game you create, if you wish to keep them playing, you must:

  • Present her with a problem that require her to learn in order to overcome a challenge.
  • Create rules that are explicit enough that she is unable to simply bypass the problem.

Related Jobs

Digital Extremes
Digital Extremes — LONDON, Ontario, Canada

University of Central Florida, School of Visual Arts and Design
University of Central Florida, School of Visual Arts and Design — Orlando, Florida, United States

Assistant Professor in Digital Media (Game Design)
The College of New Jersey
The College of New Jersey — Ewing, New Jersey, United States

Assistant Professor - Interactive Multi Media - Tenure Track
Bohemia Interactive Simulations
Bohemia Interactive Simulations — Prague, Czech Republic

Game Designer


Andreas Ahlborn
profile image
While an interesting read, my problems with this article start early on:

"All creatures play when they are young, this much is apparent through simple observation."

Even if you restrict your definition of Creatures only to the Animal Kingdom the sentence should say:
"Only a very small number of spezies play..."
Reptiles don`t play, Fishes don`t play and arthropodes (80% of all known spezies) certainly never play.

"Playing" certainly requires higher brain functions. I couldn`t even guess if all mammals have the capacity to play during their "childhood". I had to research this or make wild assumptions like "this much is apparent through simple observation".

Then it goes on - you are presenting such a generous "definition" of what games are, that it`s easy to come up with a dozen other things that easily fit into your definition (I´ll put that argument to the test later)

But the best part is here:
" A game has explicit rules and goals, and is played with the primary purpose of learning.

This definition neatly ties together what we have discovered about games and the purpose behind playing. In order to test this definition to see if it actually works on a real example we need to look at an existing game. A definition must be useful in a practical context after all."

So let me sum up the core statements of this article that should -in your opinion- answer the question you put in front of this article: "Why do we play and what purpose does it fill?"
1.Our brains are hardwired learning machines, this is what drives us into playing.
2.Our definition of a game must therefore include, in some manner, that learning is at the core of why we play.
3.A game has no meaning until it is played.
4.A game has explicit rules and goals, and is played with the primary purpose of learning.

let me rephrase that for you

1.Our brains are hardwired learning machines, this is what drives us into reading.
2.Our definition of a book must therefore include, in some manner, that learning is at the core of why we read.
3.A book has no meaning until it is read.
4.A book has explicit rules [grammar, syntax] and goals [the end], and is read with the primary purpose of learning.

Therefore (following your definition)-> a book is a game.

Nice try, though ;-)

Don`t understand me wrong, I don´t have any problems with speculations, but I have with sloppy definitions that mask themselves as "largely abstract discussion" (your words) which "must be useful in a practical context".

I couldn`t find any "practical context" here.

Next time, when "nailing down such an elusive concept", you`d better take a hammer and avoid the sponge. Defining what games are is not considered a "hard" problem in 2000 years of philosophy for nothing.

Nils Pettersson
profile image
Thank you for the comment and feedback. Allow me to retort.

I have actually never observed a young animal or human that does not play or participate in behavior aimed towards learning useful skills in an organized and protected manner. Though I must admit that my knowledge of arthropodes is limited at best. This sentence might indeed be sloppily written, I will concede that point. Have you ever observed, for instance, kittens or cubs. Or this video for that matter:

Playing, as in ludus, possible requires higher brain functions. Playing, as in paidia, possibly requires less due to its lack of exlpicit rules. Both are valid ways of playing and fulfills the same purpose.

The sentence might be sloppy but the point still stands. We (mammals atleast) play when we are young. I would very much like to see the source data supporting your statement that most animals do not play.

As far as my definition goes and the fact that you aim to disprove it. Any definition can be disproven by changing the words that make it up, which you did here. Such as changing played to read. Each word in a definition is important and contributes to the whole. By changing that word you do not test the defintion, you change it to be about something completely different (reading a book) effectively destroying the definition. If you wish to test the defintion I suggest that you work with it as written.

Let us move on to the practical applications of this defintion.

This defintions will assist you in the following manner when actually designing a game:

- It lets you remember that the core of a game is learning, thus guiding you to consider the essence of your game rather than more shallow concepts such as graphics, audio or even the drama. By beginning with the learning experience you get a pure core concept that can be solidly built and iterated upon.
- It lets you remember that games are made up by explicit rules and goals. This forces the designer to formulate those rules and goals and make them work together towards the learning experience. Game design is, after all, creating rule sets for other people to implement.
- Finally it lets you remember that the game is supposed to be played, as such it stops you from designing a book or a film.

The purpose of a defintion is to assist in everyday work. I have just shown how this defintion points towards the core of what a game is and as such, if you adhere to it, lets you begin at the right place in the design process. If this is not a practical context I do not know what is.

It is certainly true that defining a game is an almost impossible task, hence 2000 years of attempts. There have been some rather useful definitions created though and this personal attempt was not created in a vacuum. I have based my thoughts on Huizinga, Caillois, Koster and a few other philosophers. Effectively drawing upon that philosophical history.

In the end it is about finding a definition that works for you and that helps you in your work.

And that concludes my retort :)

I think I managed to answer all your feedback, please let me know if I did not. Thank you again for the comment.

Luis Guimaraes
profile image
Something is a game if, and only if, it can be gamed.

Nils Pettersson
profile image
Could you elaborate on that statement please?

Darren Tomlyn
profile image
You need to wait for me to finish my blog, since, unfortunately, (again, as usual), the OP is built upon an inconsistent foundation, (which is further based upon an inconsistent understanding and perception of the English language).

If you want to know what the word game represents, then you need to study it as a matter of linguistics, because that is the TRUE nature of the problem we have - understanding our language. Without such a foundation, everything is built on sand, which is why it keeps going round in circles without end - (probably for millennia).

Harold Myles
profile image
Nice article.

One thing I would mention and you might want to think about deals with your last definition:

A game has explicit rules and goals, and is played with the primary purpose of learning.

While I agree that much of our drive and instinct for playing and playing games comes from physiological mechanisms that promote learning, I wouldn't say that a games purpose is for learning. Or that an individuals primary purpose to play a game is to learn. At least not consciously.

Learning may be one of the end products of playing a game. And may very well be the underlying motivator by releasing dopamine for successful learning, triggering mechanisms that have evolved to push our species to learn and therefore play.

However, I wouldn't be so quick to add it to the definition of what a game is. The purpose of a game and an individuals motivation for wanting to play it may not have anything to do with learning.

By weird analogy I might throw a ball through a hoop. What makes that action work is alot of physical laws, things like gravity. But I wouldn't say gravity was the purpose for throwing the ball. Gravity is what made it work.

Similarly, playing a game may result in learning and triggering physical and psychological reinforcements, and that may be what makes the game work. It may be the reason a player will continue to play, even if they don't know it. But that doesn't really have anything to do with the purpose of the game or the individuals purpose for playing it.

Think of slot machines. We play those. And we play those largely due to them tapping into those same mechanisms that get triggered through learning. Designers of slot machines have hacked into those genetic mechanisms to keep people playing. And I think we can agree there isn't much learning going on. The slot player's purpose for playing isn't to receive stimulus, or even lose money, even though that will be the end product. They want to win money. The slot machines purpose isn't to trigger dopamine, or give out money. It's purpose is to extract money.

But you can kinda view modern video games doing the same thing. We hack into people, pushing buttons inside their brains. Those buttons may have evolved there because they promoted learning. But we aren't pressing them to have people learn. We are pressing them to keep them engaged.

I suppose what I am trying to say is that including learning in the definition might be to low level. It very well may be the evolutionary purpose for playing, but the purpose for a game doesn't have to be the same as the evolutionary roots of playing.

Hopefully that makes sense or I may even be looking at it wrong. But I thought I would bring it up.

Darren Tomlyn
profile image
The problems the OP suffers from are far too broad and deep - (which should tell you a lot about our entire perception of the matter in general atm) - to make it worthwhile picking out little issues here and there, but here's my take:

Games involve three main elements:

Something a person does (for themselves).

Play (something done that is non-productive) has nothing to do with any of these, which is a major problem with the OP. Likewise, something a person does TO compete, is not the same as competition itself - they are two different elements (as above) that should not be confused for each other.

However, play (something done that is non-productive) is NOT the same as playing a game or music etc.. (This is a REALLY big problem people suffer from.) (Games and music can both be played(verb) for work (noun).)

Game != play

And yes - that people can learn by doing something (anything) does not mean everything we do is DEFINED as and by the act of learning, itself.


But all of this is just a symptom of not understanding how and why the English language functions, and how the information it is used to represent defines and affects such functionality. (Specifically the relationship between what concept any information belongs to/represents and how and why it's representation (a word) is then used - the basic rules of grammar.)

Since we don't know what concept the word game belongs to, we don't understand it in relation to the rest of the language, which is why it's being pulled all over the place - (puzzle, competition, art and play are affected in a similar manner, (for exactly the same reason (the same basic concept)), too). And without such a foundation, no understanding of the information the word game is used to represent will ever be complete or consistent - hence all the posts such as these that try and answer this problem in such an inconsistent manner, without realising it.

Nils Pettersson
profile image
Thank you for the comment.

Interesting feedback and I definitely see your point. But I believe that if you want to connect with the player on a deeper level you need to understand what makes them tick, so to speak.

Slot machines and other games of chance falls neatly into Caillois defintions of Alea and I highly recommend his book "Man, Play and Games" if you are interested in the subject.

With regards as to modern video games. We as designers might not aim towards making our players learn but it is what our players do. They will break apart the rules of the game, find the optimal path through it and finally master it. They will do this regardless of whether we intended them to or not. Therefore I believe that it is important that game designers consider what and how they are teaching their players in order to create a truly compelling game.

Nils Pettersson
profile image
@Darren Tomlyn

I do agree that the word game is problematic. But I dispute your definition that games involve three main elements:
- Something a person does (for themselves).
- Competition.
- Rules.

I believe that six parts make up a game and agree with Caillois in this regard:
- It is free, or not obligatory
- It is separate (from the routine of life) occupying its own time and space
- It is uncertain, so that the results of play cannot be pre-determined and so that the player's initiative is involved
- It is unproductive in that it creates no wealth and ends as it begins
- It is governed by rules that suspend ordinary laws and behaviours and that must be followed by players
- It involves make-believe that confirms for players the existence of imagined realities that may be set against 'real life'.

When it comes to defining play there are basically to different definitions of the word:
- Ludus, structured activities with explicit rules (games).
- Paidia, unstructured and spontaneous activities (playfulness).

Ludus is perhaps the one most often assiciated with modern games (both analogue and digital).

And I do agree that Game != play. I never state this in my post however.

Learning is at the core of being alive since if we are not learning we are stagnating or forgetting. Humans only got to the top of the foodchain by being able to learn more than all the other species on this planet. As such I would very much argue that humans are defined by our capacity for learning.

I look forward to reading you post on this subject. :)

Darren Tomlyn
profile image
@nils - sorry for the time it took me to reply - been busy working on my blog post.

Unfortunately, your replies have two main, fundamental problems:

1) They're not consistent with the basic rules of English grammar.
2) They're still based on game=play (being non-productive).

Denying that game=play, even though you say they must be non-productive, betrays any understanding you have of this matter.

play as doing something non-productive != play as taking part in an activity/using something in such an activity. (Both when used as verbs, and when applied and used as nouns).

(Good play in a game does not mean it's non-productive, just that it was good behaviour when taking part.)

Another problem is because I didn't describe the three main elements in a manner that was fully consistent with what they are in relation to everything else - and deliberately so. The fact that my description was the only way we currently have of describing such behaviour, however, should tell you something... (Though you need to understand what I meant by 'for themselves' (as opposed to 'for others') which is something you seemed to have missed).


The biggest problem we have with language at this time, is understanding everything in relation to everything else. Most of the information we use the word game to represent, (along with others, such as art, puzzle and competition) are symptoms of this.

People play (take part in) games for WORK (productive reasons) - and they ALWAYS HAVE, for as long as they have existed (millennia upon millennia).

However, we have problems with our current understanding of what we use the word game to represent, currently.

One of the reasons is that what some, (this isn't directly related to your post btw), consider the word game to represent is now out-of-date - (game (in general) = gambling). Of course, when you have a large and profitable industry trying to convince people that they are the same thing, even though that is no longer true (propaganda) - (the word game (in general) changed, probably because of the split with the new word gamble/gambling) - it shouldn't be surprising that many people get confused.

Unfortunately all those in the best position to educate people about this matter - (those that make games (activities) that are separate from gambling) - are not doing so, and are often getting confused themselves, making what they think are games, but are instead, competitions.

But game != gambling, otherwise every single game would involve such behaviour, which they do not - and when it is present, it is often in addition to what we call A game in the first place. This means the the act of gambling, and the use of game as a verb to represent such behaviour, is not consistent with game when representing an activity - a game, as oppose to I/we game.

The rest of the problems we have with the word game, are a matter of understanding cause and effect.

The problem with the word game, is that the use of the word to describe an activity, is the CAUSE of the word used to represent everything else - (even if it's now based on an out-of-date definition, such as game (gambling) and game (small animals)).

This is a problem, because most of the information that other words use are related in the OPPOSITE direction - the cause is the thing or behaviour, that then gets applied - (such as with compete/competition).

Which is why understanding the information about the activity we use the word game to represent, then allows us to understand all of the other pieces of information it also represents in relation to such a thing.

The basic rules of grammar defines and determines WHAT types of information the language contains and uses, (the basic concepts such information belongs to) - (because this then determines how their representation is used) - and this is where we have problems.

Such information the words game, art, puzzle and competition represent is rarely recognised and understood in a manner that is consistent with such rules of the language that govern such use, and therefore understanding how and why it is related to each other and the rest of the language, because of such rules, is therefore a problem.

If games were non-productive (play), then we would NEVER play (take part in) games for productive reasons (work).

The existence of ALL professional sports that involve games means that this CANNOT be true. The fact that games are also used by the military and police (among many, many others) for training and selection purposes - (productive reasons) - also betrays this understanding and perception of games.

I put it to you that games have probably ALWAYS been used, and have existed, for such basic, productive, reasons and purposes, (especially selection).

(By selection, I mean finding out who is best at doing something - (i.e. playing a game to find out who can run the fastest/throw the furthest etc. - but of course, that means understanding what the activity is that the word game is used to currently represent in the first place.)

The word game has not always existed, and has not always been used to represent such an activity. But NOW it does.


The word game is used to represent the information of:

1) A specific activity
2) A thing/collection things that are used, usually by themselves, to enable such an activity.
3) The behaviour of gambling (usually for money).
4) Wild mammals/birds that used to be hunted for sport or food.

Both 3&4 are consistent with previous out-of-date definitions and uses of 1 - which is why they can cause problems for people that do not understand and recognise how and why they are out-of-date.

So the problems we have are with understanding the use and definition of 1. The question that needs to be answered that is currently causing problems, however, is:

What basic concept does such information belong to, that helps determine what it is in relation to the rest of the language, (including puzzle, art, competition, work and play).

Which is something we cannot currently answer consistently - (which is where my blog comes in) - and without which such information cannot be correctly described.