If you want a great example of why a good story isn’t necessary for a game to be fun, take a look at Tom Cross’ opinion piece on Dead Space. I haven’t gotten around to playing Dead Space, but according to Tom, just about everything non-game is “carefully and stylishly unoriginal”. The characters are flat and uninteresting, the plot is completely predictable, and overall the game fails at establishing a truly frightening experience. That being said, Tom loved playing the game.

I think Dead Space is a good argument against laissez-faire story design: the point of view that believes that games don’t need a story, just interesting game mechanics, and that the goal is to have the game act as a vehicle for players to put themselves into a world where they can make their own story. It seems that the designers of Dead Space attempted to create a blank slate protagonist that anyone could relate to, hopefully facilitating the player’s desire to insert themselves into a fantasy world where they can create their own story. The problem is that from a plot perspective, a flat protagonist is uninteresting.

Isaac never speaks, and you never get any indication of his mood, other than that he doesn’t like dying. He wears a mask throughout the game and reacts to little. Apparently, this makes him relatable, because so many of us are demure, voiceless, deep space mechanics who constantly wear masks.

And because the main character is flat, the story that revolves around him lacks any emotional attachment or depth. But the game is fun, it has a good set of core game mechanics. What get Cross worked up is that the gamer in him enjoys the pure game, but because Dead Space makes an attempt at telling a story, his basic human desire for interesting and compelling plot drives him to feel annoyed at the same time. A game doesn’t need story to be fun, but if a game does try to tell a story, the lack of a coherent and interesting plot generates the slightest bit of friction during gameplay. It rubs us the wrong way. A game doesn’t need story to be fun, but if you do want to have a game that tells a story, make sure you implement that story’s plot properly.

The problem with Dead Space, aside from the mediocre effort placed in crafting a story, is that there is no care to implement game mechanics that best tell this story. The story and game mechanics in Dead Space are for the most part completely unrelated. On top of that, the reason the game isn’t as frightening as it could be stems from the fact that the space of possibilities afforded by the game’s mechanics simply don’t have a scare factor to them; they don’t deliver.

I’m not suggesting that every flaw Cross points out is directly related to a flaw in game mechanics and nothing else. But I do want to emphasize that every flaw, in part, does relate to poorly executed game mechanics. I find this last statement odd when considering that those very same game mechanics provide Cross with an “amazingly fun” experience. However, it does makes sense when you consider that games and stories are two completely different things. What makes a game fun and what makes a game tell a good story aren’t necessarily the same. But imagine for a second what would happen if they were the same. This is what coherence is all about.

Overall, though, I think the most fascinating part of Cross’ article is the following chunk of text:

[...] most people are focusing on how the tempo of that movie [Aliens] is similar to Dead Space’s gameplay. They say that this game is like Aliens, with its frantic action and small scares, and less like Alien‘s slow creeping dread.

What “most people” are doing here, without even knowing that they’re doing it, is focusing on how the dynamic plot (telling) created by Dead Space‘s game mechanics (and experienced through gameplay) is much more like the plot (telling) of James Cameron’s Aliens than Ridley Scott’s Alien. Fantastic. It’s good to see people intuitively catching on to this concept.

I was reading an analysis by Daniel Cook over at Gamasutra that at first made me nod in agreement. Games have repetitious themes and often implement them in disjunction with or with disregard for their game mechanics. It’s good to see yet another member of the games industry catch on to this endemic problem and suggest a solution. But then a curious thing happens, Cook goes from talking about literary themes to what I’d rather call genre, but he keeps on calling it “theme”. Piracy in a work isn’t, as Cook assumes, a literary theme like “redemption” or “estrangement”. It’s a theme in the same sense that you can have a costume party with a pirate theme… but that definition doesn’t do us any good here. A literary theme arises from the interplay of plot, setting, character, conflict, and tone. According to Cook:

The theme you select directly influences how you present your initial skills to the user. By saying “pirates,” I turn on a particular schema in the player’s brain and a network of possible behaviors and likely outcomes instantaneously lights up.

“Pirates” here is more of a genre, a mental model for what is  expected of the structure and content of the work. It isn’t the same as a story about redemption, something much more intangible and difficult to communicate. But whatever you want to call it, mixing up this “theme” with literary themes only leads to a confused analysis of how themes and games interact. To implement a literary theme, let’s say redemption again, would require well thought out and coherent game mechanics that convey the essence of what it is to be a protagonist experiencing or delivering redemption. There isn’t a simple mental model for the representation of a literary theme, since these are generally aspects of the human condition that have confounded us since the dawn of recorded history.

I mean, okay, this kind of stuff is generally dismissed as quibbling. Getting into arguments about terms and definitions is usually the quickest way to say a whole lot of nothing while annoying the hell out of everyone. But, clearly defined terms are the only way we can have productive conversations. It seems everyone in this industry has their own definition of theme,  story, plot and narrative; and we wonder why no one can agree on what it means to have a game that tells a story.

The “eye” or  “camera” in video games (essentially the player’s viewpoint) has always been designed pragmatically to allow the player to play the game. Rarely has there been any consideration of what the viewpoint is telling the player about the experience, other than purely game-related facts. If I could make an analogy to film, the camera in video games today is  used like it was by filmmakers at the dawn of the 20th century. They say, “here is the scene!”, and really nothing else. Similarly, in games the camera says “here is the game!”, and really nothing else. It wasn’t until the 1950s that different camera angles and the introduction of the jump cut started to be used to tell the story rather than just present the scene.

Corvus has a series of posts on the topic over at Man Bytes Blog. He believes that “by limiting our video games to this presentation, we’re limiting our ability to use the camera as an effective storytelling tool”.

Absolutely. We’ve seen attempts at adding interesting camera angles and even dynamic cut-scenes to games, and where these didn’t destroy gameplay by making the game impossible to play, they certainly did make the presentation of the game more lively, but ultimately their goal is still to show you the game. The only exception is probably the “establishing shots” that some games offer you when entering a new level or being faced with a new foe or puzzle. I believe looking to film for inspiration is a good starting point, like early film looked to theatre. However, it’s important to realize that the presentation of story in a game happens differently. Cut-scenes, dynamic or otherwise, can certainly take advantage of film theory to better tell what they hope to tell, but these aren’t intrinsic parts of a game.

If we want to consider how to revolutionize the use of camera in video games, we must consider them from a systems point of view. How do they function within the game’s system of play, what game mechanics govern their behaviour? By considering cameras as a functional unit, with rules and behaviours that, if coherently designed, can impart mood, atmosphere, character, emotion—that is to say: plot—then, it might be possible to discover our very own version of the  “jump cut”, something that revolutionizes the way cameras tell stories in games.

Rubes over at The Monk’s Brew has a few thoughts on the tricky problem of trying to have a realistic protagonist that fits into the story… who is also controlled by an unpredictable player. The problem is of course that all too often a game’s mechanics will allow the playful player to do something ridiculous, something the protagonist would never really do. Should we just make a game and, as Jimmy Maher once said, expect the player  “to accept the premise and situation of the story she is in, and to behave in a reasonable manner”?

I don’t think we should ever expect anything of the sort. Let’s face it, there shouldn’t be a right and a wrong way to play a game. If it’s within the limits of the rules, then why shouldn’t the player be expected to do something that conflicts with the story? The answer isn’t, I don’t think, in having sophisticated enough AI to respond to a player’s inanity and keep the fiction going. The work required is just too astronomical.

It may some day be possible, but until then there is a solution that can be implemented in games today. The answer is to have game mechanics that are coherent with the story being told. Instead of having “all player actions [...]  interpreted by the game within the context of the character performing the action (his or her personality and relationships) and the situation within the narrative”, the game mechanics can be designed to only allow player actions that are coherent with the story. If every game mechanic is coherent with the story, then any version of the dynamic plot generated when playing the game will be coherent with the story being told. That’s the key.

How do we do this? Well, I’ve mentioned character creation before. Instead of first coming up with the character’s history, personality, or even their name… craft the character based on the function you want them to have in the story. Once you’ve established that, create game mechanics that coherently express this function. It’s easier said than done, which is why you rarely see it. But it isn’t by any means impossible, it just requires some forethought.

Welcome to part III of The Flaws of Narrative, Manifested, a look at Michael Abbott’s Narrative manifesto. Check out part I and part II, to see what I think about the rest of the manifesto. Keeping the best (and most difficult) for last, part III is dedicated to Jonathan Blow.

I don’t think there’s any other single person in the games industry today that’s more in line with my feelings on story and games than Jonathan Blow. That being said, somehow I feel that if ever we got together to talk, we’d end up disagreeing more than agreeing.

Jonathan Blow – Conflicted games

Well, let’s start with what we’d agree on, because ultimately that’s what’s most important.

First of all, game mechanics that are “disharmonious” with the story being told create conflict in games, preventing the game from really resonating with players. This is exactly in line with what I’ve said before on coherence in games. He gives the examples of how in BioShock the story tries to establish a ideological conflict between radical individualism and altruism by having the player chose between killing the Little Sister for personal gain, or saving her for…well, here’s the problem: saving her gets you half the personal gain, and every third Little Sister you save you get a bonus. In the end, the difference between killing or saving the Little Sister is negligible. The story wanted to say one thing, and for obvious game balancing reasons, the mechanics subverted that meaning. BioShock’s game mechanics establish a “dynamical meaning” (I’d go with “procedural meaning”) that conflicts with the meaning the story is trying to tell. Jonathan argues that every game mechanic has a meaning, whether intended by the designer or not, due to our natural inclination to attribute meaning to everything we encounter. Since we can’t avoid it, we need to start looking for it and training ourselves to design games with it in mind.

All of that aside, I figure where Jonathan and I will disagree is with small things like the meaning of “story”. I see story as an abstract choronology outside of any medium that can be any possible narrative about any possible thing, whereas Jonathan sees stories as those narratives that are worth telling. But really this isn’t an impasse, we’re talking about the same thing. I chose not to narrow what should be considered a story because I don’t want to inadvertently limit the power of what we’re establishing here. I think that even games that don’t try to tell stories can still benefit from the notions of “harmony” and “dynamic meaning”. Take what I’ve said recently about Team Fortress 2 as a good indication of a mainstream game without a “story” worth telling that still benefits from these concepts.

I expect that this is just one of the small quibbles we’d have because of our different backgrounds. I mean, he agrees with Gaynor’s panic over the inherent chaos and unpredictability of the player, and I don’t. But I figure ultimately we’d agree more often than we disagree.

This is part two of the Flaws of Narrative, a look at Michael Abbott’s Narrative manifesto. Click here to read part I.

If I were to follow Michael’s order, I would next comment on Braid‘s visionary designer Jonathan Blow, but since I basically agree with everything he says, that’s going to be a mountain I’ll climb in part III.

Steve Gaynor – The game designer’s role

Steve’s point of view is similar to Redding and Hocking’s (part I) in that he believes the designer should be largely “hands-off” and simply give players the tools they need to create their own stories.

I believe this is again an example of a viewpoint that sees the inherent chaos and unpredictability of the player, and without being able to see beyond traditional linear storytelling just gives up on story in games by pointing to the popularity of games like The Sims, or online worlds, where players create their own stories. My reply is basically the same in part I. Just because there is an infinite number of prime numbers, we shouldn’t avoid looking for the Riemann zeta function; and just because genetic mutation is inherently unpredictable and chaotic doesn’t mean we shouldn’t pursue the limits of the theory of evolution. Again, Gaynor’s point of view isn’t wrong, we should definitely endeavour to allow players to experience richer stories that they feel are unique to their actions. It’s just that this view keeps us from fully considering the larger picture of storytelling in games.

L.B. Jeffries – Non linear reactive stories

Jeffries is definitely a wild-card, but I like that. I think it’s definitely by getting away from our preconceived notions of games and stories that we’ll make progress in this field. The application of tarot card reading, an ancient “storytelling” art, to storytelling in games is really fascinating, but isn’t the answer to the overall problem.

Jeffries suggests that just like tarot cards take advantage of our mind’s natural inclination to create meaning, to take chaos and enforce order on it, so too could these principles take the chaos of player agency and allow the player to infer their own meanings:

A series of reactions like someone crying for help if you shoot them or a dog following you if you feed it could be created in response to the player. Rather than worry about how these relate to some grand linear story, simply leave them as short vignettes that connect and relate to one another through A.I.

This, I believe, is a fantastic approach to storytelling in games, but it doesn’t need to come at the expense of a grander story. With enough forethought, the vignettes could easily combine to form a larger dynamic plot, or else help reinforce a dynamic plot told through other game mechanics, art, environments, sounds and haptics (yes, I will hammer you with this string of potential storytelling vehicles every chance I get).

Over at Gamasutra they have a detailed and well researched piece on crafting compelling characters in video games. But I think something’s missing. It’s not that what Tychsen is saying is wrong, technically he’s just reporting his observations as he sees them. His honest goal is give people the knowledge they need to design better characters in games, that you can see. Unfortunately, the way he approaches what a character is dooms him to repeat the same mistakes that have lead game after game to implement flat, boring characters.

From a systems’ point of view, a character is just a unit within the game, with attributes, relations, statistics, and behaviours. This is Tychsen’s viewpoint, and it’s fine when creating a game for gameplay’s sake. But as soon as you introduce story, you need to look at characters in a different way.

[…]if the character is not interesting to play, the gaming experience will not be of a sufficient quality to motivate the player to continue player.

Fascinating characters can make a game and create lasting relationships with the player that keep them coming back for more – as is evidenced in the game series featuring characters such as Lara Croft, April Ryan, Max Payne, Crash Bandicoot and Sonic the Hedgehog.

First off, you could have the most interesting character in the world and it won’t save you from plain old bad gameplay. Consider the wildly popular and completely unsuccessful Sonic the Hedgehog franchise. I love that little guy, and I have fond memories playing Sonic games on my Sega Genesis, but I really haven’t enjoyed the 3D versions that have been coming out over the years. Sonic the Hedgehog is a character, yes, but moreso, he’s the embodiment of a specific kind of gameplay that I really enjoy. Take away the gameplay, and I don’t enjoy Sonic anymore.

And this leads me to my point: If experiencing gameplay is to experience a game’s procedural plot, and Sonic is the embodiment of that gameplay, then Sonic isn’t just a character but an actantial component of the plot.

Actants aren’t units in a story; they don’t exist and affect the fiction as agents, “causing” things to happen. Instead, actants only exist because you want a certain action in your plot to occur. It’s a subtle change of perspective that has you constantly acknowledging the fictionality of the plot, the fact that the story is a construction. Uncle Ben doesn’t die at the hands of some burglar because one rolled lawful good and the other chaotic evil, or because the burglar had a higher agility score, or because the burglar had a bad day, really needed the money and got careless. The reason Uncle Ben dies at the hands of some burglar is to provide the realistic motivation needed by the plot to fulfill Peter Parker’s transformation into the friendly neighbourhood Spiderman we all know. As far as Spiderman’s origin story goes, that’s the only reason why the actants of Ben and the burglar exist in the plot.

That’s interesting, you might think, but what does that have to do with game design? Well, like I said, Tychsen’s systematic approach isn’t wrong, it’s just lacking the depth needed to implement characters that are well rounded and feel like they belong in the game’s procedural plot. For that, you need to consider what that character’s actantial role in the plot will be. And this is where the two meet: once you know exactly what an actant’s purpose is in the plot, then you can start thinking about what game mechanics, attributes, statistics and behaviours best serve to coherently fulfill that purpose. The end result is a character that’s both structurally and actantially sound, the kind of character that really emphasizes the plot and brings your game to life.

Ian Bogost has a new column over at Gamasutra called Texture where he makes the case for a tactile appreciation of games. From Go‘s game pieces to bump mapping, to force feedback to games that “touch”. These touchy-feely, hard to quantify aspect of the games “are pleasures more subtle and confounding then the anonymous fun of solving a problem in a game”. But the concept of texture can be taken further than these simple tactile pleasures; a game’s texture can be used to relay plot.

–

Yes, the “P” word. Allow me the benefit of the doubt.

Bogost briefly talks about how different meanings can be conveyed through force feedback: The tremor that accompanies gunshots in Call of Duty 4 alert the player to unseen dangers, the rumble of lose dirt beneath Link’s feet in Legend of Zelda: Orcania of Time signals buried treasure, and the pulse of your player character’s heartbeat in Silent Hill is an indicator of his current health. Interestingly, Bogost also adds to this last example that the heartbeat’s quickening pace as you get closer to death “instills fear”. This mechanic isn’t just an innovative, non-visual way to provide feedback to the player. It also adds something completely irrelevant to the game’s system of play, but quite remarkable to the game experience.

To take these kinds of textural effects even further, let’s take a look at Ico. For those unfamiliar with it, Ico is the story of a young warrior, Ico, who as part of his village’s tradition is taken to a mysterious castle. There, he finds a young woman named Yorda who he must save by helping her escape the castle without having her dragged away by shadow-like creatures. Ico also uses force feedback, but not for any game related reason. Ico can take Yorda by the hand to drag her along at a quicker pace. When they join hands a quick burst of force feedback reinforces this connection. The startling rumble of the controller mimics the startling physical sensation of a first touch, and the more you play, the more it comes to illustrate the tactile, sensual nature of their relationship, the trust and safety of holding hands. It can even come to signify the emotional connection of the two characters, reinforcing the emotional attachment the player starts to feel for Yorda.

This textural effect is pure plot. It comes to affect the way you feel, and is irrelevant to the game itself. But what if we wanted a textural effect that was more like the one in Silent Hill? Useful and evocative. Then let’s take the rumble effect in Ico a step further. Suppose that the controller also rumbled distinctly, sharply, when Yorda is attacked by shadow creatures, or if she is about to fall. The rumble would serve the useful purpose of giving the player feedback on Yorda’s well being, while the startling vibration would mimic the pang of fear and worry that Ico feels over her safety, emphasizing their relationship and again helping to reinforce the player’s emotional attachment.

What’s important to realize here is that all of these actual or suggested textural effects in Ico are basically irrelevant to the game’s system of play. Within the game’s rules, you have no choice but to care about Yorda’s well being since the player fails if she dies. Rumbling when Ico and Yorda join hands (or when she’s in danger) isn’t necessary to the player caring about Yorda, the player has no choice; but it nonetheless does add something remarkable to the game experience. These “added” textural effects are valuable because they enhance the way the game tells you its story. To borrow from Barthes, texture acts as a kind of indicial function, a part of the plot that establishes mood, atmosphere or gives character. Like a jump shot as opposed to a slow pan in a horror movie, the way you tell (show) the story adds to how the story is experienced. The jump cut will make the scene startling and ultimately more satisfying than the slow pan, although it doesn’t actually change the story in any way. Texture too is a way for games to establish mood, atmosphere or give character. These are tiny, aesthetic aspects of the way the story is told, yes, but ones that make the story richer for them.