[Industry veteran Richard Fine looks at the benefits of conveying information via gameplay compared to other modes of exposition, in this #altdevblogaday-reprinted opinion piece.
There's a well-worn mantra in fiction writing: "Show, don't tell
." The Janet Evanovich quotation sums it up nicely:
"It is the difference between actors acting out an event, and the lone playwright standing on a bare stage recounting the event to the audience
From an information-content point of view, the latter still gets everything across -- but the former would be more fun to watch. Why is that, though? What's so fun about seeing the events acted out?
In many cases, if not all, fun comes from learning (as Raph Koster writes
). The process of understanding -- of gaining a new explanation of how something is done or why something works, and thereby solving a problem that was posed by the lack of that explanation -- is enjoyable.
So maybe that's why showing is more fun than telling -- because it offers more to learn, more opportunities to undergo that process of understanding.
"The Colonel was very brave
When somebody tells you something, you've got two options: take it or leave it. You can't do more than that, because you've got nothing else to go on. It's their way -- their summary of events -- or the highway. If you've got questions or objections, there is no further evidence available with which to reason. (We're talking about a broadcast medium here, like a book or a film or a game. Obviously in a normal conversation, you can ask questions -- but increasingly it stops being what I'd call telling).
So the amount you can learn is very limited -- the speaker, usually the narrator, has done most of the inference for you, so what you see is what you get. It's efficient for the author to write, and efficient for the audience to consume, but if there are any doubts or questions lingering in the audience's mind afterwards, it does nothing about that. Telling tells you the way things are
, and hopes that that's good enough for you.
Older games made heavy use of telling for things like the game's storyline; for example, the 'mission briefing' format, where the game's story would be presented as screens of text between each level, is textbook telling. At the same time, many games also explained their game mechanics by telling the player how they worked in the manual.
"Lying prone in the mud amidst the gunfire, the Colonel gritted his teeth, and reached down past his blood-stained uniform to pull a cigarette out from the packet. The wound hurt like hell, but he was damned if he was going back now -- not while his boys were still in the enemy camp
On the other hand, when somebody shows you something, you are being given more detailed information, and being allowed to decide for yourself which things are significant, and which things feature into your ongoing understanding of how the characters think and how the world of the story works. You have (or make) guesses about how things work and why people do things, and as the Show gives you more and more information, you reject and refine those guesses.
A show can take a told idea -- such as 'the Colonel is brave' -- and let you see exactly why
the narrator considers him brave, and when
he is brave, and what his motivation is for being brave, and so on. If you doubted that he was brave -- perhaps because he did other things that it didn't seem so brave -- then the more detailed account given by the show might help you see him in a new light, and understand how his brave nature could fit with those other things. Showing allows you to not only discover the way things are, but also how and why they are that way
, which is a much more substantial piece of knowledge that can be used to answer many questions and assuage many doubts.
The original Half-Life
was greatly praised for its innovative approach to storytelling: it was almost all show and no tell. There was no opening screen full of backstory text, no long cutscenes with characters 'filling you in' on what had happened with no regard for what a person would actually
say in their situation. The shocking moment when you first meet the U.S. Marines and discover that they're out to get you too isn't highlighted with someone saying "The Marines are under orders to kill you too" (at least not immediately). So there's a wonderful few moments of confusion -- was it an accident? Did they mistake you for an alien? No, these guys are too good for that. If they were shooting at you it must have been on purpose. Why would they do that? And then the truth dawns on you...
Many games, now, follow in Half-Life
's footsteps. Cutscenes and flashbacks replace the briefing screens of old. The location design includes information that story elements can be deduced from, like computer terminals and newspaper clippings -- or, more subtly, big welcoming banners displayed on the outskirts of ruined cities, human skeletons on the tables of alien medical labs, and so on.
Showing is still limited. While the author isn't overtly giving you their own interpretation of events as with telling, it still influences things, as they're still in control of which details you are shown and which you are not. So, they'll anticipate the questions you'll have, and try to include enough details that you can work out the answers; but they'll often fail to anticipate your questions entirely correctly, and they might not provide satisfactory answers -- or you might miss the answer that they intended you to find. If that happens, then there's no recourse, as with Telling.
It's the learning equivalent of a rail shooter: the author sets up the questions and answers for you in the hope that it will satisfy you, and they usually cover most of what you wanted to know, but there's often a few unresolved issues -- plot holes, actions that seemed out of character, and so on. Not to mention the "Don't go in there!" effect: Who hasn't watched a movie, seen the characters do something stupid, and thought, "Aargh! Why they hell didn't they just do it this other way?"
That's where games come in. Because games, as an interactive medium, add a third mode of exposition: Play.
When a creator lets you play with something, lets you engage with a mechanic and poke it and turn it over in your hands, then you gain the freedom to explore counterfactuals ("what if" questions). In a film or a book, if the characters take the left-hand door, then they will always
take the left-hand door, no matter how many times you rewatch or reread; but in a game, you can take the right-hand door instead. Instead of being nice, you can be nasty; instead of being cautious, you can be gung-ho; and you can see what happens.
For the most part, you can decide which questions you want to explore, and which ideas you want to put to the test, at your convenience. Through play, you can discover not just the way it is, and how it comes to be that way, but also why it couldn't be any other way
-- why the ideas that are simply shown in other media are hard to vary
without making them worse. (Or, sometimes, that it should
be another way, as characters in films and books often don't make the best decisions).
Every game, as you'd expect, contains some element of play. The most remarkable ones are the ones that let you play with unusual things: Facade
, for example, lets you play with human relationships; Dead Rising
includes a mechanic that lets you play a little with photography and capturing interesting and well-framed shots; and a game like Mirror's Edge
lets you play with free-running and acrobatics. There are, of course, a huge number of games that let you Play with things like combat tactics, too.
Just as with showing, however, there are limits. Often in a game you can't
take the right-hand door instead of the left-hand door, because the developers set the game world up that way, barring the door and not modeling the room behind it -- but really, this just means that "which door gets taken" isn't something that the developers are letting you play with in that game. They let you play with other things instead, like whether you throw a flashbang grenade into a room before you enter it, or whether you use a health pack now or later, or what happens if you press the button marked 'self destruct,' and so on.
Play is also subject to the developer's interpretation, just as much as showing or telling, because the developer is free to distort and focus their modeling of real-world phenomena. A game about running for a political office, for example, might base its mechanics around making rousing speeches and designing good policy -- or it might base it around bribing electoral officials and leaking scandalous stories about one's rivals...
OK, so play can be used for exposition purposes. So what? Isn't showing and telling good enough already?
Firstly, consider that none of these exposition methods apply solely to game stories: exposition can be used to convey any
idea, from "he is Luke's father" to "you should hit the weak point for massive damage." Any idea that you want the player to acquire will appear somewhere in this framework.
Secondly, while showing and telling are often perfectly fine and easy, play allows you to convey deeper ideas, and it allows you to convey them more persuasively. For a person to be persuaded that an idea is true (or at least, the closest thing to true they've seen so far), any alternative ideas they can think of need to be dismissed as inferior -- inferior because they don't work, or they don't fit the evidence, or they're pointlessly complicated, or they're too arbitrary, or whatever.
While showing and telling can convey an idea, they're one-shot weapons: the creator decides what idea he wants to get across, and commits it to a medium that is ultimately read-only, in that, once published, his work will only ever
communicate that idea. The audience consumes that idea, perhaps with varying interpretations (i.e. my idea of 'brave' might be a bit different to your idea of 'brave'), but if they don't buy it, the author has no second chance to convince them.
When the audience is invited to play with an idea, however, the author's got a much more populated arsenal: the game can accept an entire range of approaches that the player might try out, developing them to their conclusions. If the player is playing the game a particular way, and they think of a better way, then they can try that out. They can put all of their alternative ideas to the test, and discover what the actual problems are with each one in turn. When they've finished playing the game, not only do they understand the idea that you set out to communicate, but they've also dismissed many of the alternative ideas they thought of, too.
Sid Meier famously said that "a good game is a series of interesting choices." For something to be a choice, there must be multiple alternatives available to the player -- and for the choice to be interesting, the alternatives must all be viable, to the extent that it could be worth trying any one of them. Considered against the role of play in allowing the player to learn what not to do, it's clear to see why Meier is right: if you don't offer choices then you are merely telling or showing, rather than playing; and if the choices are not interesting, it is because the player doesn't consider the presented alternatives to be superior and in need of testing.
I invite you to consider which forms of exposition you're using for which aspects of your game, and to think: are there things that you're presently showing or telling that you could be playing instead?
If you're making a game about a demon hunter, for example, how are you communicating that the player has killed a demon? I'm going to guess that you're showing it -- the player swings their sword, the demon spits up blood, and so on. The player will probably buy that -- after all, it's a simple show that conforms to all the standard gaming imagery -- but, if you wanted to let them play with it, why not let them choose exactly where to stab and slice the defeated demon, based on a simple model of demon anatomy?
The best strategy is the one that leaves no living demons, while still taking the least time (because other demons are around and the player shouldn't linger). As the player works their way towards that strategy, they'll make mistakes, sometimes wasting time chopping up a thoroughly dead demon, other times finding themselves face to face with a pissed-off and wounded but not dead
demon... but when they succeed, they won't just be taking your word for it that they killed the demon, they'll feel like they did it with their own two hands, because they've developed expert knowledge on the subject. It helps transform the player from a generic fighter into a real butcher.
Let the writing of your game -- not just the story, but the themes and concepts you want to convey -- inform not just the art and sound direction of your game, but the core gameplay design itself. Let it inform your balancing and tuning. Bake the message you want to send into the player's own actions.
Play with play. You'll have more fun that way.
[This piece was reprinted from #AltDevBlogADay, a shared blog initiative started by @mike_acton devoted to giving game developers of all disciplines a place to motivate each other to write regularly about their personal game development passions.