A Comprehensive Game Design Methodology
From First Ideas to Spectacular Pitches and Proposals

The content of this website is licensed under a Creative Commons CC BY-NC-SA 4.0 Attribution–NonCommercial– ShareAlike 4.0 International License. You can freely use and share everything non-commercially and under the same license as long as you attribute it to and link to:

J. Martin | |

However, you can also buy the Ludotronics PDF edition
for an unreasonably moderate price at DriveThruRPG.
Learn here about the five excellent reasons to do so!

Why DriveThruRPG? It’s the largest download store for role-playing stuff in existence and you’ll probably end up buying much more than just your copy of Ludotronics. Which would benefit game designers everywhere!

Why not Amazon? For one, illustrated non-fiction isn’t well-suited for the Kindle format. Also, at a €14.99 price point, Amazon’s cut amounts to €9.75. Well, no.

Level Five: Architectonics

Process Phase Level Five

Beat 6. Personnel

Personalities & Personifications

In this final beat from the Architectonics territory, we will take a comprehensive look at characters against the background of function.

Counting among the first things you learn as a writer, usually, is that stories are about characters and not about things. Star Wars is not about the Death Star but about Luke, The Lord of the Rings is not about the One Ring but about Frodo, and The Long Goodbye is not about the portrait of Madison but about Marlowe. That’s true, but it’s not the whole truth. What stories are about are these characters’ wants and needs and the emotions attached to these wants and needs that drive Luke, Frodo, and Marlowe forward. But it’s still not the whole truth! That Luke feels frustrated, Frodo desperate, and Marlowe disillusioned isn’t meaningful in itself. What stories are about are the emotions of the audience, not the emotions of the characters. When, and only when, your audience reliably feels your characters’ emotions by empathizing with them, then your story and your set of characters have justified their existence.

How your characters’ emotions can be represented most effectively, in turn, depends on the medium. In a slightly different context in Beat 3. Paradigms Lost, we differentiated between words, visual media, and games. For our current context, we need the additional distinction between the spoken word and the written text. In stage plays like tragedies or comedies, emotions are primarily represented through dramatic dialogue. In written media like short stories or novels, emotions are primarily represented through dramatic description. In movies, emotions are primarily represented through dramatic visualization. Stage plays also convey emotions through visualization and written texts and movies also through dialogue. But these are not their primary modes of representation, respectively.

Interactive media like games have three primary modes of representation, each attached to one of the emotional classes and their corresponding dramatic arcs. From our emotional landscape that we developed in Level One: Integral Perspectives I, the fiero class (associated with the player development arc) is primarily represented through player action. The fear class (associated with the story development arc as the situation the player can see themselves in) is primarily represented through dramatic visualization. The pity class (associated with the character development arc through empathizing with characters in the game world) is primarily represented through dramatic dialogue. Each, in turn, can utilize the other modes of representation as secondary or tertiary modes, respectively, but should do so only for excellent reasons, not as a default mode. (We will get back to this topic in the context of goals in Level Six: Integral Perspectives II.)

Mapping these aspects to game characters, we can now define their principal functions and how these functions are represented.

  • The function of the player character is to want something the pursuit and accomplishment of which can be primarily represented through player action (fiero) and secondarily through dramatic visualization (fear).
  • The function of non-player characters is to help or hinder the player character and to elicit emotions in ways that can be primarily represented through dramatic visualization (fear) and through dramatic dialogue (pity).
Fig.4.44 Game Character Model I: Meta-Function
Fig.4.44 Game Character Model I: Meta-Function

Within this framework, we can proceed to explore the properties and qualities of functional characters in detail. Many of these properties and qualities will be informed by Melanie Anne Phillips and Chris Huntley’s Dramatica: A Theory of Story, which warrants a brief preamble.

The Dramatica theory has a reputation for pulling the uninitiated into byzantine mazes of narrative theory, a reputation that is mostly undeserved and brought about more by its technical vernacular than by its practical concepts. The aspects from this theory that have been incorporated in our Ludotronics game character model in this beat are the composition of the character set with its role of the contagonist; the differentiation between main character and protagonist; and the notion of the steadfast character. These aspects from the Dramatica theory enrich the palette of narrative techniques and are especially valuable for the purpose of designing dramatically complete games.

Before you can begin to build the functional character set for your game, two important concepts need to be introduced. The first is the difference between functional characters and interesting characters. Roughly, functional characters deepen the story’s plot and the player character’s journey (motivation), while interesting characters deepen the player’s involvement (emotion). In more theoretical terms, functional characters are attached to the structural mode of a narrative, similar to plot points and journey stages, while interesting characters are attached to the rhetorical mode of a narrative, similar to style or form or word choice. What exactly is structured by functional characters will engage us in a minute. While this difference between functional and interesting characters should always inform your design, it should be perfectly invisible to the player. The functional character who keeps spurring the player/player character on and into action by providing designated perspectives and arguments, and the interesting character who is an impressively sarcastic herald who enjoys fast horses, handsome knights, and the opportunities of diplomatic immunity, is one and the same! What connects the functional and the interesting aspects of a character is what motivates that character, but that, alas, is a game writing topic that would lead us too far astray.

The second concept that needs to be understood—beyond a passing familiarity, that is—is the concept of archetypes. If you consult your books or your favorite search engine for story character archetypes, you will most likely be confronted with neat and tidy lists that contain a range of archetypal characters, along with their respective properties and explanations, “based on C. G. Jung.” None of this is remotely true. Even if you manage to filter out lists of stock characters, which have nothing to do with archetypes in the first place, these lists neither reflect Jung’s concept of archetypes, nor his archetypes! What happened here? It’s not unlike a telephone game, only that the message, instead of degrading, was step by step transformed into ever more streamlined listicles. Considering the significance and prominence of the concept of archetypes for storytelling in general and our purposes in this beat in particular, let’s saddle up for an excursion into its colorful history.

Everything begins with Jung’s concept of a deeper layer of the unconscious, where images or motifs reside that are not personal memories. These images or motifs he calls “archetypes,” a term adapted from philosophical traditions. One example of such an archetype is the Shadow as the unrecognized dark half of the personality of which the “devil” is a variant. Another example is the Magic Demon, endowed with mysterious powers, as one of the most ancient conceptions of “god.” This might sound familiar. But there are other archetypes, like the Anima as the all-merciful Great Mother and the Animus as the Father Figure, who together form the archetypal idea of the Syzygy. Then, there’s the Child archetype, the Maiden, the Dwarf, the Old Man whose appearances include the king of the forest and animal manifestations, or the archetype of the Spirit whose workings for good or evil depend on conscious decisions. And so on. Jung’s archetypes certainly don’t compile into a list, and particularly in his essay “Archetypes of the Collective Unconscious,” Jung explicitly excludes the possibility of a list of archetypes that one could then learn by heart. Also, in the same essay, he points out that the figures who appear in myths and fairy tales are not archetypes, but “stamped” and “handed down” forms that are derived from archetypes.

The second distinguished node in our telephone game is Joseph Campbell, who calls Jungian archetypes elementary ideas or “ground” ideas that manifest themselves not from Freud’s personal, biographical unconscious, but from a “biological” unconscious. These ideas, in turn, appear in different historical environments and in different costumes as “bounded” forms, as Campbell calls it—as, e.g., the archetypal “teacher of mankind,” bringing great boons or fire, who has been “bound” into the Trickster, or Shadow figure. This paints a subtly different picture. It’s these “bounded” forms that Campbell attaches to the hero’s journey, giving them names like the “supernatural helper,” or the “herald” whose appearance triggers the crisis and the journey’s Call to Adventure, or the “ogre” father who also appears as the Proteus figure of the Shapeshifter. All this is more familiar ground already, but not all of the items are familiar or bear their familiar names, and we’re still far from having a coherent list.

Enter Christopher Vogler. In his seminal The Writer’s Journey, he is the one who takes all these concepts and contexts and bakes them into the form most writers and storytellers are familiar with. He greatly simplifies, and simultaneously sharpens, the idea of archetypes toward practical usefulness. He does this by casting them as different aspects of the human mind, personified and endowed with human qualities, that “play out the drama of our lives.” He further enhances their functionality for the purpose of storytelling by making them more flexible through abstraction and by curating their numbers, until we finally get our familiar list that contains only those archetypes Vogler deems most useful for writers—the shapeshifter, the threshold guardian, the trickster, the shadow, the herald, the ally, the mentor, and what Vogler calls the “higher self.”

Yet, for the writer, this scheme poses a structural headache as soon as the story has a broader cast of characters instead of a small, ragtag group of adventurers who stick together throughout. Some archetypes might be integral to specific plot points but not to others, not least through their “interesting” characteristics, and become unavailable when they’re needed. Vogler defuses this by going a step further. He transforms these archetypal characters, who, as described, have already been substantially transformed from their Jungian beginnings, into masks that can be switched between characters. That way, different characters can play different archetypal roles at different times.

Which is a clever solution, but not a free lunch. While certainly streamlined, Vogler’s archetypes—shapeshifter, guardian, trickster, shadow, herald, ally, mentor, higher self—are still very specific and not easily transplanted from one interesting character to another interesting character without friction. One casualty of which, as touched upon briefly, is character motivation because it can no longer tie a character’s functional and interesting characteristics together coherently and conclusively.

That’s already a formidable headache, but it’s not the only one. The hero’s journey is certainly not the only possible journey after which character development arcs can be structured. Even for journey structures that are equally deeply rooted in Western culture, like the quest or the return home, most of these archetypes decay rapidly in terms of usefulness, regardless whether they’re personified characters or switchable masks.

Thus, we need to add one more node to our telephone game, and that is the functional character set as devised by Phillips and Huntley’s Dramatica theory, as introduced above. It’s a character set that is easy to learn, universally adaptable, and a breeze to apply. The Dramatica theory calls the characters in this set archetypal characters, carrying on that tradition. But we will stick to our terminology of functional characters to stress their structural aspects and dramatic functions.

Here’s how it works. First, Dramatica switches from the terminology of aspects—aspects of the human mind or the unconscious—to the terminology of arguments. Next, it creates a collection of arguments that best represent the dynamics of wants and needs on the one hand, and help and hindrance on the other. Then, it personifies these arguments into a set of functional (archetypal) characters which, together, represent the problem-solving processes of the human mind. Not unlike Vogler, who defines his set of archetypes as a metaphor for the human situation, Dramatica defines its set of personified arguments as the “story mind.” Finally, this story mind embodies the conflicting perspectives and proposals on how to overcome obstacles and satisfy wants and needs within the character’s mind, including opposed wants and needs as antagonistic forces. (As an aside, this is not too far away from Gustav Freytag’s view, which he expressed in 1863 in Technik des Dramas, that each dramatic character has a particular function with respect to the whole by representing a single common human trait.)

This might sound complicated at first, but it is really, really simple. It’s a set with two groups. In the first group, there are four functional characters: the protagonist who personifies the pursuit of wants and needs; the antagonist who personifies the preventive forces of diametrically opposed wants and needs; a character who personifies help and conscience called the Guardian; and a character who personifies hindrance and temptation called the contagonist, to which we will return in a moment.

The second group also contains four functional characters: personified Reason, personified Emotion, a character who personifies trust called the Sidekick, and a character who personifies doubt called the Skeptic.

That’s it—one set with two groups of four personified arguments, for a total of eight functional characters that you can learn by heart! They’re easily adaptable to different journey structures and can switch between characters, if need be, with the least possible amount of friction.

Now, the contagonist. As indicated, the contagonist personifies hindrance and temptation. The concept has its predecessors—notably, it identifiably relates to the Shadow figure or the Trickster. But to consolidate these lineages into the functional character of the contagonist is perhaps Dramatica’s most remarkable contribution to the dimension and use of archetypes. The contagonist, as the “dark” counterpart of the Guardian, might or might not relate to the antagonist in ways that are similar to how the Guardian relates to the protagonist. Unlike the antagonist, the contagonist doesn’t personify wants and needs directly opposed to those of the protagonist. Thus, the goals of the antagonist and the goals of the contagonist might or might not align, or might just loosely align, but they’re rarely if ever identical. The argument the contagonist makes in the story, and that is its core function, is not promoted by assault but by attractiveness, and its objective is not annihilation but seduction.

In the original Star Wars trilogy—of which particularly the first movie is used as a central example by Dramatica—the protagonist is Luke and the antagonist is the empire, later personified by the emperor. Each side wants to destroy the other. Vader, in contrast, is the contagonist: while his goals mostly align with the goals of the empire, he has his own agenda, and he doesn’t want to destroy Luke but seduce him with the promise of unlimited power.

J. R. R. Tolkien’s The Lord of the Rings, as another example, has many story subsets with their own local protagonists and antagonists and so on. But on the top-level, where Frodo and Sauron serve as protagonist and antagonist, similar to Luke and the empire, the One Ring embodies the contagonist. Again, its goals are aligned with those of the antagonist, Sauron, but it wants to seduce, not destroy—and Gollum, as a classical mirror-character, is what Frodo will become if the One Ring has its way.

As a third example, let’s take a look at Ash in Ridley Scott’s Alien. Ash doesn’t “seduce” anyone, and he actively tries to kill Ripley when she finds out about Special Order 937 after gaining access to Mother, the ship’s computer. But whether the crew lives or dies is none of his concern. All he does is subtly manipulating situations and influence decisions to protect the xenomorph. With that, his goals align with those of the titular organism, who wants to kill the whole crew, or use them as hosts. But their goals are by no means identical.

As you can infer from these examples, neither the antagonist nor the contagonist have to be sentient beings. The antagonist can be a collective force like the empire or a force of nature like the Alien, and the contagonist can be a dangerous artifact like the One Ring or Ash, the synthetic person.

But we’re not done. To repeat, the contagonist wants the same thing as the antagonist, roughly, and accomplishes that by seducing instead of annihilating the protagonist, or trying to. The contagonist is not defined by being good or bad. Neither is the Guardian! In Ryan Coogler’s Black Panther, for example, the character who wants the exact same thing as the antagonist Erik “Killmonger” Stevens is none other than Nakia, T’Challa’s former girlfriend. Through persuasion and implicit seduction, she tries to bring T’Challa to change his mind and reveal Wakanda to the world and use its powers to help the oppressed. That’s exactly what Erik wants! (Albeit with some striking differences in method.) Then, there’s T’Chaka as Guardian who, it turns out, brought about the world’s imbalance and the ensuing conflict through his moral failure in the past. (That’s a motif the later Star Wars movies also toy with, but neither as consistently nor as forcefully as Black Panther.)

With the unfamiliar concept of the contagonist out of the way, let’s look at a few examples for the other functional characters in our set.

Fig.4.45 Game Character Model II: Character Set
Fig.4.45 Game Character Model II: Character Set

The first group is almost always fully loaded and carefully defined. In Star Wars, that’s Luke as protagonist, the empire as antagonist, Vader as contagonist, and Obi-Wan as Guardian. In The Lord of the Rings, the first group from the top-level story line has Frodo as protagonist, Sauron as antagonist, the One Ring as contagonist, and Gandalf as Guardian. But there are many subordinated story lines in The Lord of the Rings that populate their first group differently; in one of these groups, for example, it’s Saruman who embodies the role of the contagonist.

The second group from our functional character set, in contrast, rarely appears as fully defined and as clear-cut as the second group from our Star Wars example. In Star Wars, it’s Leia as Reason (making plans and all the rational decisions), Chewbacca as Emotion (“he likes to pull opponents’ arms out when he loses”), Han as Skeptic (“that’s suicide”), and R2-D2 and C-3PO as Luke’s trusty Sidekick team who also provide comic relief. In more expansive works like The Lord of the Rings, these roles shift all the time between different characters and are often more reminiscent of Vogler’s masks than Dramatica’s arguments. The Sidekick role, for example, is sometimes embodied by Legolas and Gimli, sometimes by Pippin and Merry, depending on context and story line, with each team providing both trust and comic relief. And while Aragorn almost always embodies the role of Reason, the remaining two functions Emotion and Skeptic are forever on the move within and between different character constellations.

However, all this doesn’t mean that you can only become creative with your functional character set’s second group. Alien, for example, with its very intricate set for such a small cast, plays with its first group in clever and creative ways. Ripley is a covert protagonist for about eighty pages into the final script, and when she becomes the protagonist, interesting things happen. To see which, let’s wind back for a moment. Earlier, when the movie starts, it looks as if Kane were the protagonist. After he dies, Dallas seems a good pick, and everything looks reasonably neat. There’s Dallas as protagonist, the Alien as antagonist, the ship’s computer, Mother, as Guardian, and there are enough clues to suspect Ash as contagonist early on. But there are also clues that something is seriously wrong with this constellation; one example is that the Guardian doesn’t seem to do anything, least of all providing guidance to Dallas. When Dallas dies and Ripley is revealed as the protagonist, things falls into place. Ripley switches over into the first group. This gives her access to Mother, the Guardian, which she didn’t have before, and in contrast to Dallas, she receives vital information from her (because she is the protagonist and Dallas wasn’t). Later, Mother still continues to guide Ripley by continuously providing vital information about the ship’s status. (It’s Ripley who messes up the self-destruct sequence, after all.) There’s a lot more going on, especially with respect to the second group. But it should suffice to give you ideas.

Just like reverse-engineering a theme, reverse-engineering a functional character set can be fun. But that’s not what it’s for. The functional character set is not a gotcha game, and there is no score. It’s a structural design tool that helps you develop a full set of arguments for the “mind” of your story. Whether you want to use the whole set, use each functional character exactly once, or let one or more functional characters switch roles in different contexts, that’s entirely up to you.

As our examples demonstrate, the functional character set can hold itself up against vastly different proportions—from a tightly-plotted blockbuster screenplay to a novel that spans six books in three volumes and took twelve years to write to a chamber play with a total of ten characters, three of which are a ship’s computer, a xenomorph, and a cat. That’s because the functional character set is a form, not a formula. Well, okay, it’s also a formula. But it isn’t formulaic! It’s an incredibly flexible formula that you can bend any which way you want through the force of your own creative will, always providing you with a versatile and resilient structure that supports what you’re doing and helps you keep track of all the roles or arguments in your game.

With all that under your belt, from the difference between interesting and functional characters to a solid understanding of archetypes to the creative possibilities of what the Dramatica theory calls the Story Mind, you can now go and sketch fresh, complex, dynamic functional characters that your game writer or team of game writers will make interesting later on, all based on a simple set of eight easily memorizable dramatic functions.

Let’s continue with a related topic. Up to now, as you might have noticed, the player character was never called the “protagonist.” The reason is that there’s a functional difference between the protagonist of a story and the main character of a story, and that the player character can be one or the other, depending on your game’s dramatic structure.

Differentiating between the protagonist and the main character of a story is another substantial contribution of the Dramatica theory, and it is again incredibly simple and immensely useful. Here’s what it is about.

The protagonist is the character who drives change and pushes the plot forward. The main character is the character through whose eyes we experience the story. That’s it! In any story, the protagonist and the main character can be one and the same, as Luke in Star Wars or Decimus Meridius Maximus in Gladiator or James Bond or Lara Croft or Nathan Drake, and so on. Indeed, this seems to be the default option. But then, there are Arthur Conan Doyle’s Sherlock Holmes stories, Harper Lee’s To Kill a Mocking Bird, Terminator 2: Judgment Day, or Mad Max: Fury Road. In each of these stories, we have a protagonist—Sherlock Holmes, Atticus Finch, Sarah Connor, and Imperator Furiosa, who drive the story forward. And in each case, we don’t experience the story through the eyes of these protagonists, but through the eyes of main characters—Dr. Watson, Scout, John Connor, and Max Rockatansky.

Differentiating between a protagonist and a main character can be particularly useful for game design, but in games it doesn’t work in the exact same way as in our examples from other media above. It can’t! If the player experienced the game through the eyes of the main character avatar while a non-player character protagonist did all the heavy lifting, that wouldn’t be a very satisfying playing experience.

Instead, think of the great number of different levels and subplots and side quest and, perhaps, the occasional story line contingent on player decisions. There’s absolutely no reason to cast the player character as the protagonist everywhere all the time. To temporarily relieve the player from having to drive the story forward gives you a huge range of design opportunities. For example, your game can start with the player character as main character to defuse some of the classic problems attached to beginnings, as discussed in Beat 2. Prime Positions, after which the player character slowly develops into the game’s protagonist. You can design different story lines or subplots or side quests where the player character switches roles, depending on context. You can realize difficult plot points in more natural ways by casting the player character temporarily as main character and let a temporary protagonist make a number of decisions, maybe with the help of soft cutscenes or even no cutscenes at all. Also, temporarily casting the player character as main character gives you more opportunities for the design of participatory play, as discussed in Beat 5. Participation. It’s a very versatile tool.

Fig.4.46 Game Character Model III: Protagonist vs. Main Character
Fig.4.46 Game Character Model III: Protagonist vs. Main Character

But, you might ask, what about the term “hero?” Up front, it’s too diffuse. On the one hand, everybody is the hero of their own story, and everybody is involved in their own hero’s journey or quest or return home in one way or another. On the other hand, someone who overcomes seemingly insurmountable obstacles, including oneself, to drive profound change can certainly be called heroic. Thus, the label “hero” could apply to the main character as well as the protagonist, to the character development arc as well as the story development arc. That makes it fairly useless as a technical term because technical terms should discriminate between different meanings, that’s the one job they have. Terms like “hero’s journey” or “hero NPC,” in contrast, are highly specific and fit for use. (We’ll cover hero NPCs further below.)

For a practical example, let’s put this split between protagonist and main character to use in ways that allow choices and decisions for physical, cognitive, and also empathic tasks, as discussed in Beat 4. Problem Space. Imagine you were given the chance to make the original Star Wars movie into a game! How would you go about it? How would you realize player agency in a story that’s fixed, finite, and famous?

Luke is the protagonist, no doubt about that, and if you want the player to play the protagonist, that would leave you with the first major headache right after the crawl. Because the protagonist isn’t present at the inciting incident! As it should go without saying by now, if you think “cutscene,” you lose. But if you simply dropped the whole sequence and let Leia “tell” in an extended holographic message what’s happened on her cruiser and “explain” why this incident is so special, reminiscent of all the bad habits we touched upon in Beat 2. Prime Positions, then all you got were a weak and drawn-out second-hand ‘splaination moment instead of a distinctive and forceful beginning. That’s not what you want. Next, what kind of meaningful actions and decisions can you give the player as tasks? Until the Jawas arrive and mount their garage sale, Luke can attend to the farm or waste some time with his friends at Tosche Station, that’s about it. Then, together with his uncle, he has to buy exactly two droids from the Jawas. And the correct droids, for that matter. Then he has to clean and repair them. Then to come across Leia’s message. Then to give chase to R2-D2, only to be knocked out by a group of nomadic raiders and rescued by Obi-Wan. And so on. There’s no room for player agency to speak of.

But what if, for your interactive video game version, you made a split between protagonist and main character and let the player character be R2-D2?

Granted, that would necessitate a lot of market research, as explored in the Preparation phase. But let’s assume you identified a primary target audience for your concept that gives you a promising market estimate. R2-D2 is an immensely talented droid, and a vitally important character. R2-D2 is always at the center of the action, from the initial fight against Vader’s boarding party up to and including Luke’s X-Wing run along the trench of the Death Star. For the player, R2-D2 could provide a wide range of gameplay mechanics—cognitive skills, starship mechanic and astromech skills, hacking skills (think cyberpunk netrunner), and perhaps even electronic combat skills by supporting on-board computers for navigation, maneuvering, and targeting during dogfights.

Most notably, though, R2-D2’s not as bound by the story as Luke is. On Tatooine, right after the drop, your player can face any number of situations that demand actions and decisions and puts R2-D2 in the role of the protagonist. The same applies to the belly of the Jawas’ sandcrawler. Later, there’s room for a major adventure outside Mos Eisley’s cantina; a lot can happen during the hour or so that Luke and Obi-Wan spend inside having drinks, talking to Wookies, separating guests from their limbs, and haggling with smugglers. And so on, always exposing your player to the full gamut of physical, cognitive, and especially empathic tasks.

Over the course of these adventures, the character R2-D2 can undergo, e.g., a hero’s journey (an ambitious but rewarding job for your writer or team of writers later). It starts off with the Call to Adventure when R2-D2 is entrusted with the Death Star plans—so that, this time, the inciting incident and the Call to Adventure run side-by-side. Sure, R2-D2 wouldn’t have a Refusal of the Call like Luke, but no hero’s journey has ever been perfect. (C-3PO is a walking and talking Refusal of the Call in any case.) Yet, you could let the game start a little earlier on Leia’s cruiser, maybe shortly before at the battle of Scarif, with a first Call to Adventure that the player/R2-D2 must refuse for practical reasons. Either way, the game would start off in the thick of the action; all three development arcs would be in full effect; and nowhere will you, the game designer, be tempted to fall into the trap of the narrative/gameplay balancing paradigm that we discussed and abandoned in Beat 3. Paradigms Lost.

For any medium, a story is then, and only then, the right story when it’s both great and a great fit for its intended mode of expression. If you have the right story for your game, the dreadful chimera of narrative (story) vs. gameplay (choices) simply ceases to exist. In a nutshell: if the story you have in mind for your dramatically complete game demands trade-offs in either the autonomy/agency department or the plot points/stages department, you’re either trying to tell your story in the wrong medium, or you’re trying to tell your story from the wrong perspective. Correct for that, and you’ll be fine.

Let’s proceed to Dramatica’s third and final contribution to our game character model, the steadfast character.

Since time immemorial, general wisdom has it that the central character of a story has to “change.” This seems to go back to Egri Lajos’s hugely influential publication of Dramatic Writing: Its Basis in the Creative Interpretation of Human Motives in 1946. Change is inevitable, certainly, because experience is inevitable—both in the sense of learning and in the sense of awareness, as discussed in Level One: Integral Perspectives I. But that’s not the kind of change Egri meant and that everybody understood. Egri explicitly states that a fixed “nature” of a character constitutes bad writing. From there, the question of “has the central character sufficiently changed or not?” became everybody’s favorite yardstick to separate “good” writing (highbrow) from “bad” writing (lowbrow).

This we should leave behind. There are terrific characters in great stories who “change” in the sense that they grow and do not cease to learn, but who do not transform their nature. Maximus in Gladiator is an excellent example, and so is Furiosa in Mad Max: Fury Road. (Fury Road’s main character Max Rockatansky, in contrast, does transform.) What Maximus and Furiosa are, in Dramatica parlance, are steadfast characters.

Like the protagonist vs. main character distinction, the transforming vs. steadfast distinction offers a greater range of design options for games that invite the player to identify with their player character. If the player character is a detached character like Manny in Grim Fandango or Kate in Syberia, as discussed in Level One: Integral Perspectives I, then that character can transform as much as stories or journeys require without causing trouble. But in games like those from the Mass Effect trilogy, it becomes very difficult to transform the player character against the will of the player without impacting player motivation and player emotion in undesirable ways. For such games, it can be more opportune to design the player character as steadfast, as a character who overcomes obstacles and achieves their goals by staying true to themselves, not by changing their nature.

The most effective application of this design strategy is to give the player a choice—to transform or remain steadfast along a series of well-designed empathic tasks throughout the game. This hooks right into the “what” vs. “how” distinction from Beat 4. Problem Space. “What” the player has to achieve for the game’s dramatic solution is fixed, but “how” the player achieves it, through transforming or remaining steadfast, modifies this ending’s circumstances, emotions, outlook, and so forth, in ways that are compelling, plausible, and necessary.

Fig.4.47 Game Character Model IV: Transforming vs. Steadfast
Fig.4.47 Game Character Model IV: Transforming vs. Steadfast

To wrap up this beat about functional characters, we need to examine one more thing: major non-player characters as “hero NPCs.” A hero NPC is a special type of functional character. It is related to Egri Lajos’s concept of “pivotal” characters, but better suited for our purpose of designing games.

The term “hero NPC” is adopted from Chris L’Etoile, particularly from his sidebar contribution to Lee Sheldon’s Character Development and Storytelling for Games. Against the backdrop of conflict, bad things happen to player characters all the time, from being ridiculed to being tortured to being murdered as part of the plot, all of which happens surprisingly often. Designing such events certainly has its challenges, especially the death of the player character. But these challenges are unrelated to the problem at hand. The problem at hand is about terrible things that happen to the player character’s friends and family members that look like plot points but actually are less about plot and more about motivation and the player’s level of emotional involvement, the most egregious of which is known as the “Women in Refrigerators” plot device.

In games, beyond the problem of good or bad storytelling, such plot devices pose a structural problem. There is a line that should not be overstepped: when unreasonable demands or expectations of actual emotional suffering are imposed upon the player. Letting loved ones of an undetached player character suffer or die after the player has begun to identify with this player character will almost always backfire. First off, it doesn’t work. The player would need to empathize with their grieving player character, and it should be clear by now, after Level One: Integral Perspectives I, that you can’t empathize with yourself, or pity yourself, except in the form of self-pity. If it does work, miraculously, then it runs the risk of imposing an unacceptable amount of negative emotions on the player. Which is something you shouldn’t do, at least not without the player’s explicit approval. It’s about media ethics, and that’s a thing. There’s a lot to dislike, to say the least, about the rating guidelines from ESRA, Iran’s Entertainment Software Rating Association. But ESRA has a terrific content descriptor named “Hopelessness.” This descriptor specifies, among other noteworthy player emotions, the “deep feeling of sorrow due to the death of a lovable character.”

This is where Chris L’Etoile’s concept of hero NPCs comes to the rescue. Hero NPCs elicit empathy by “absorbing the burden of tragedy,” a burden the designer can’t and shouldn’t impose on the player directly for all the reasons just elaborated. When a beloved character dies as a dramatic necessity, it shouldn’t be a character beloved by the player character. It should be a character beloved by a hero NPC. That way, the player can empathize with the hero NPC’s grief. They don’t have to empathize with themselves, which is impossible, or weather the full impact of the loss, which is possible but ill-advised. Hero NPCs absorb the burden of tragedy, drive the plot forward, and elicit empathy all at once, without imposing undue emotional demands upon the player directly.

Fig.4.48 Game Character Model V: Hero-NPCs
Fig.4.48 Game Character Model V: Hero-NPCs

To repeat a point that needs to be stressed, all this doesn’t apply to games with detached player characters that players can have empathy with. It should also be stressed that none of this applies to games where the player can prevent the suffering or death of a beloved character through gameplay by becoming a better player and, particularly, by making better decisions. It also doesn’t apply when a non-player character has sufficient agency and decides to do something that can and will lead to their own suffering or death. All these cases do not pose structural problems, and they’re certainly not bad storytelling per se at all.

But then, what about the introduction sequence of The Last of Us? Clearly, Joel is not a detached character like Manny or Kate, the player can’t save Sarah through better gameplay, and there’s no agency involved on Sarah’s side concerning her death. Doesn’t this sequence, then, place the tragic burden, the death of the player character’s daughter, directly on the player?

No. If you analyze this introduction sequence closely, you will see that Neil Druckmann, who co-directed and wrote The Last of Us, uses a whole palette of distancing techniques to keep exactly that from happening.

To begin with, the introduction sequence is twenty years removed from the actual events in the game world and serves expositional and motivational purposes before the bond between player and player character is firmly established. Then, the introduction sequence’s primary player character is Sarah, peppered with cutscenes and soft cutscenes. (The death of a player character, as mentioned, does not pose design problems in our context.) It’s only during a brief chase scene toward the end of the introduction sequence that the player controls Joel, who will later become the primary player character. This practical distancing is reinforced by behavioral distancing when the player-controlled Sarah, her uncle Tommy, and presumably also the player strongly disagree with Joel when he aggressively prevents them from picking up a desperate family at the roadside. (It wouldn’t have helped them a lot, but still.)

The whole introduction scene, then, works as well as it does because the player is able to empathize with Joel and his situational emotions before Joel becomes the primary player character when the game proper begins.

Thus, if you want the player to experience a particular range and intensity of emotional suffering without harming the player or running afoul of self-empathy, you need to create distance. That distance can be created by letting not the player character suffer these emotions, but a hero NPC.

That should suffice. The game character model provides you with all the tools you need to sketch a set of functional characters for your dramatically complete game that your writer or team of writers—including yourself, if you happen to be a writer—can build on during development.

up | down

Outlook (The Story State Machine)

Twice in this phase, we touched upon fruitless attempts at designing procedural storytelling and plot point–generating algorithms. But there might be a possibility to deliver powerful story nodes that escalate all three arcs—the story development, the character development, and the player development arc—and ramp up emotional involvement without micro-controlling the player’s movement and individual progression.

It’s a thought experiment, nothing more, nothing less. It might create more problems than it solves, like an elaborate chindōgu, or devour more resources than its effects would justify. But it is well within our technological capabilities, as our discussion of data-tracking in The Witcher 3: Wild Hunt or Cyberpunk 2077 in this level showed. It is conceptually related to, but not based upon, Penny Sweetser’s approach in Emergence in Games, which guides dramatic tension through accumulated player data and threshold values that trigger plot point–related events, and also to Left4Dead’s already mentioned AI Director which creates different intensity curves for each cooperative play-through with context-dependent dynamic difficulty, enemy and item placing, and aesthetic effects (which includes dynamic audio, a topic discussed in Level Four: Narrativity).

Basically, it’s about tapping into the other three territories’ riches, Interactivity, Narrativity, and Plurimediality, and tie these riches to player data in what we will call the Story State Machine.

Here’s the premise. Emotional impact and a sense of urgency are not the exclusive domain of what happens. It’s a compound experience, composed of what happens and how it happens and where it happens.

Imagine you are designing a series of three important encounters. At each encounter, the player receives a piece of information (Story Development Arc) that will also reveal an insight about their role in the game world (Character Development arc) by winning a fight against an enemy (Player Development Arc). Ideally, urgency and intensity should increase from encounter to encounter. But you cannot know the sequence in which your player arrives at these encounters!

What you can do is assign style elements (Plurimediality), difficulty level (Interactivity), and some well-chosen narrative qualities or properties (Narrativity) dynamically, so that each encounter plays out depending on its actual position in the sequence.

Let’s look at an arbitrary encounter in the sequence. If the player arrives there first, it plays out in the pleasing environment of a clearing on a sunny day and demands a reasonable amount of skill from the player. If the player arrives at that same encounter later in the sequence, it plays out during a sunset that casts the clearing into long and uncanny shadows, demands a challenging combination of skills, and maybe the fight must be won before nightfall to add some time pressure. If this is the player’s final encounter in the sequence, it plays out in the nighttime during a thunderstorm and requires every bit of player skill and character abilities and equipment. Not only are the enemy’s skills cranked up in sync with the actual sequence, but also their looks, voice, and general demeanor. Camera angles become increasingly sharper, or tilted. Music and sound go from upbeat to frightening to infernal. Other, and more intense, mythical backstory and game lore elements become involved. And so on. As the majority of movable assets trade places along the sequence, the player will have enjoyed most of the design work by the end of the last encounter.

Technically, everything would be a single package that switches to different states contingent on actual player movement. The player can make any number of decisions and encounter plot points or journey stages in any order, but still experience all three arcs—story, character, and player development arc—with consistently escalated significance and emotional intensity.

In Ludotronics parlance, as noted, this package is called the Story State Machine. (It could as well be called the Journey or Proficiency State Machine, but alliterations go a long way in terms of memorability.) It’s a thought experiment that might perhaps lead to other, more elaborate thought experiments, which in turn might lead to practical solutions with enough potential to spread. In Level Two: Interactivity, it is briefly argued that procedural storytelling must fail because stories are emergence simulators. Thus, we should try out different directions, and that’s what the Ludotronics Story State Machine attempts to do.

Fig.4.49 The Story State Machine
Fig.4.49 The Story State Machine

Congratulations! You have beaten this level. Your sketches from this territory will go through many iterations, and you will later need a writer or team of writers to flesh out everything, devise plot points and journey stages and functional characters that hold water and are also interesting, offer choices and decisions that fully utilize your game’s available problem space, and deliver your playing experience with the right combination of agency and participation. At this point in time, though, your sketches will be more than adequate to support your pitch and possibly your prototype.

The primary motivator for this territory is autonomy/agency, and that’s what we focused on throughout this level. But don’t forget, or neglect, Architectonic’s secondary motivator, purpose/goal! Meaningful choices alone cannot provide the enjoyable and satisfying playing experience you have in mind. For this playing experience to take shape in the mind of your player, these choices must be interlaced with purpose and pushed toward increasingly relevant goals.