Classifying Characters in TTRPGs

Concepts for discussing rich campaign activity

TTRPG discussions have a simple and long-used terminology to describe the various player/character divisions. The referee (also called the Dungeon Master, Judge, Storyteller, and so on in different contexts) oversees and adjudicates the gameplay. The players control Player Characters (PCs). Any other characters in the gameworld are controlled by the referee and are considered Non-Player Characters (NPCs).

At first glance, it seems like this problem is thoroughly solved. We appear to be able to make all important distinctions between characters, players, and the referee.

But even in games with the most conventional designs, we encounter situations calling for more nuanced distinctions. Let’s investigate this problem and find a more complete understanding of characters and their role in a TTRPG.

The 4-Point Semantics

Since there are four distinct central terms, we will refer to the conventional system as the 4-Point Semantics.

• referee

• player

• player character (PC)

• non-player character (NPC)

Strictly on the basis of language used, the 4-Point Semantics already present us with difficulties. If the referee is participating in the game, is he not a player? What if the referee goes through the character creation process to make a PC—does this change his status as a player? Is he both a player and referee? Essentially, is the referee a player by virtue of playing the game or must he control a PC to become one?

If one of our players has a Beastmaster PC which has a pet companion, is that pet a PC or NPC? Does this depend on whether the referee or the player controls the actions of the pet?

Imagine we set up a game designed to cleanly follow the 4-Point Semantics. Players control single PCs without any complicating factors like henchmen, lackeys, or pets. The referee controls all other characters.

One of our players has to move away and cannot participate in regular sessions because of timing conflicts, but he still wants to participate in the game. He decides that he can contribute by determining the week-to-week actions of an important NPC—a vampire lord plotting a necromantic ritual to extend the night-time.

The arrangement would amuse that player while decreasing the workload for the referee, who would no longer need to decide what that NPC will do. In this situation, 4-Point Semantics would tell us we have a player-controlled non-player character—an exquisite semantic trainwreck.

The categories in the 4-Point Semantics do not stand up to scrutiny even in common situations produced by conventional play. Partially this is due to simple linguistic confusion, but it’s also the case that we are missing out on important logical categories. In more nuanced gameplay where interactions with the campaign are more dense and less constrained, 4-Point Semantics entirely flops.

Controller Semantics

A more descriptive set of categories can be derived from focusing on the actors in the gameworld rather than the players at the table. For every character in the gameworld, we can always ask, “who controls this character?”

If a human participant is deciding what that character would do, we will call that a Controlled Character (CC). If, instead, that character’s actions and behavior are an emergent product of game mechanics, we will call that a Non-Controlled Character (NCC). These are not mutually exclusive categories in the strict sense, but we will clarify smaller distinctions later on.

In Controller Semantics, a Controller is the human participant controlling the character. Since the referee is often controlling characters as part of his arbitration role, he too is a Controller.

Controlling Whom?

What about someone that isn’t controlling a character but is otherwise playing the game?

Let’s imagine that someone wanted to participate in the game but not as a character. Perhaps they wanted to be “the weather” or some other abstraction.

We go through the process of working with them. Each day we ask The Weather what he is doing that day. “Today, I am windy. But tomorrow? Thunderstorms.” Is this “player” really participating in the game properly?

This is kind of a goofy line of questioning, but it reinforces our instinct that the fundamental nature of the TTRPG—focused on the “role” aspect—is that of guiding characters through the gameworld. What constitutes a “character” is up for debate on the fringes, but it must be something imbued, in principle, with decision-making capacity—the precise capacity which the player is providing.

Non-Controlled Characters

Consider a sentry guarding the entrance to a high-value warehouse.

If no other character encounters this sentry, his existence is arguably meaningless. This is like the tree that falls in a forest with no one nearby to hear it. However, if a Controlled Character (CC) encounters this sentry, he instantly becomes meaningful! The player controlling the CC is discovering the world through the CC’s eyes, and the sentry is part of that world.

If a Controlled Character (CC) is distinguished by requiring a Controller lending decision-making capacity to that character, then the Non-Controlled Character (NCC) must be making decisions on some other basis. The sentry’s behavior is determined by the game rules—perhaps a reaction roll determines his ultimate disposition towards the CC; alternatively, the game might have an extensive Sentry routine that the NCC would follow.

The NCC is an essential part in fulfilling the game’s promise of a rich, discoverable gameworld.

Emergent Characters

Imagine the referee decides to step in; he takes into account the sentry’s likely motivations and tendencies before deciding on the sentry’s response to the CC encounter.

In such cases, the sentry ceases to be an NCC and becomes a CC. The same reasoning—and same conclusion—applies to e.g. an encounter with zombies in a crypt or a merchant on the road. This highlights two special points about NCCs.

Procedure as NCC Essence

Entities which are summoned into place by the game are intrinsically NCC. Since their existence is procedural, their motivations and decision-making tendencies are implied to be procedural as well. Games whose rules disrupt this natural pattern are almost certainly making an error.

We state that, as a matter of principle, emergent characters are not controlled by the referee but by the game’s procedures. If circumstances demand more attention to a character than the game’s procedures can cover, then that character must pragmatically become a CC. A more perfect TTRPG will decrease the incidence and necessity of this switchover.

NCC as the Asymptote Character

Similarly, consider a CC whose Controller lapses in their duty or otherwise is unavailable to provide decision-making. There is an opportunity here for that CC to become an NCC. In the absence of the Controller, a CC’s behavior can be determined by procedure.

This may be perceived as more “fair” than alternatives for two reasons. First, the ownership of the character is not overwritten at any point. Second, the game is implicitly understood to be a just arbiter of elements it oversees.

A more perfect TTRPG will make the process of CC → NCC more concrete and more readily available.

The Divergent Purposes of CC and NCC

Thus, an important asymmetry exists between the CC and the NCC. A more perfect TTRPG will see the rate of NCC → CC decline towards zero while the rate, acceptability, and mechanical support for the CC → NCC conversion will increase, allowing CCs to seamlessly become NCCs.

In 4-Point Semantics parlance, allowing PCs to be treated as NPCs when necessary should be easy and acceptable enough that it is a simple and reputable practice—but requiring numerous NPCs to be played explicitly by a player should be minimized and shunned. By these means, the game will more emphatically express its essential strengths—and the players are more properly participants in the game rather than part-time showrunners or stand-in game designers.

Canonicals and Universals

Imagine a TTRPG where every character in the gameworld conforms to some system of character creation. Beasts and similar entities use the Creature Creator. Members of an exalted pantheon of gods use the Deity Creator. Humans and human-like sentients use the process outlined in the Character Creation chapter of the book—essentially a Mortal Creator. In this game system—and in all the gameworlds we can spawn with it—every character can be classified as belonging to one of the Universal Categories (Creature, Deity, Mortal, and whatever others).

In many TTRPGs, the canonical mode of play involves creating characters who are exceptional and distinguished from the great mass of beings otherwise in their Universal Category. Usage of Controller Semantics alone does not allow us to hold these characters properly apart from the mass. Thus, we propose the concept of the Canonical Controlled Character. Since this descriptive style of naming will inevitably cause us trouble, we’ll call them Canonicals.

In AD&D, consider the Universal Category of mankind. The great mass of this Universal Category, the Universals, are undeniably distinct from those with class levels, the Canonicals. AD&D’s human Universals are farmers, merchants, tradesmen etc. The Canonicals are Fighters, Paladins, Thieves etc. Not only is this distinction absolutely clear, there is even an explicit pathway from Universal → Canonical: gaining a class level!

Imbued With Purpose

The Canonical is a standout example of a Universal Category. But if we take a step back, there is an even deeper reason to identify and mark them.

When the game designer put the rulebook together, it was the Canonicals that filled his considerations. It was the Canonical for whom exceptions and clarifications were made. Controlling a Canonical—in his thinking—is the most “correct” way to play the game. A certain kind of player—he imagines—will create this type of character and run them through sessions in the canonical fashion.

Thus, even when a Canonical is not elevated explicitly (by being of a special caste or role in the gameworld), he is nevertheless still elevated implicitly by being the prime, obvious way for players to interact with the gameworld. For this reason, we suggest shifting to this deeper meaning of the Canonical label: the Canonical is a character intended to be the focus of the gameplay, and they are often (though not necessarily) of elevated status—via the game mechanics—compared to similar characters.

Even when Canonicals and Universals have no mechanical differences (such as in examples like Classic Traveller), the Canonical label is still signaling the importance of their controllers’ role in the gameplay.

Imposed vs. Organic

Another useful distinction for characters is a measure of their authenticity or belonging with respect to the gameworld. Because the gameworld is the product of the game’s rules, this idea is suggestive of an alignment between characters and the game’s ruleset.

If a character resides in a position or circumstance because a player thought it would be interesting to place them there, that character’s presence has been forced on the game by outside influence. We will call this an Imposed character. If, in contrast, a character’s presence and station were a result of the game’s internal procedures, that character more naturally belongs in the gameworld. We will call this an Organic character.

This is a narrow concept that is not meant as a means to judge the aesthetic quality or fitness of characters but simply to classify them strictly with respect to their function and role in the gameworld. The difference between Imposed and Organic is often clear, but it is easy to conjure examples where a character is one or the other by degrees.

Collective Impact

Consider a referee building an exploration map. He has marked out a number of hexes and is rolling through procedures to determine terrain and “special” features in the hex. In one hex, he rolls both a “water” result and a “witch’s hut” result. He interprets this as a swamp inhabited by a witch. He looks up the “witch” result in the bestiary and rolls an example, taking note of possible behaviors the witch might have.

Consider a different referee also building a map for an upcoming campaign. In one of his marked out hexes, he rolls a “water” result but decides that it should be a swamp inhabited by a witch. He is pleased with this idea and starts making notes on the character, filling in interesting motivations the witch might have and pondering what magic she wields.

Even though both examples have similar outcomes, the pattern in the first is that of an Organic character whereas the second is clearly an Imposed character.

The single outcome (the witch) doesn’t present much difference, but the campaign’s overall feel will be very different with one mindset vs. the other. A campaign full of Imposed characters will not necessarily reflect the underlying game’s design, aesthetic, or ethos. A campaign filled with Organic characters whose behavior and characteristics are ripped from the game’s own tables is necessarily the more authentic in this narrow definition.

Controllers and Belonging

Based on our measures of authenticity, Controlled Characters (CC) will tend more towards the Imposed side of authenticity whereas Non-Controlled Characters (NCC) will generally be Organic examples.

If we envision a campaign’s gameworld as inhabited and appropriately stationed by a host of NCCs (both explicit and implicit), the campaign state in this moment is the most Organic it could possibly achieve. If players seeking a means to make an impact on the gameworld select a handful of these NCCs to control, that is certainly going to move things at least a few notches towards Imposed on the global scale.

If the usual method of play is for players to create Canonical characters and “add” them to the campaign’s gameworld, we can mark this is a point of imposition.

Preferred Directionality

Yet, if a Canonical character has made a big impact on the gameworld and become an important facet of the campaign over time, their absence would be the imposition.

Thus we can see another utility of this narrow metric of belonging. Long-lasting characters that make an impact on the campaign can move the needle of belonging towards Organic even if they were strongly Imposed in origin. The opposite result is possible as well: a character that diminishes the cohesiveness of the campaign’s gameworld can become increasingly Imposed over time.

We posit that a more perfect TTRPG will encourage and aid Canonicals in becoming more Organic through their campaign contributions.

Scope and Utility

A loose hierarchical representation, stolen from a DOD database

We have extensively covered the scope of these ideas. Let’s see how they can be applied.

In a continuous Braunstein¹, a gameworld is evolved in time from a well-specified starting point. The actors involved may be ordinary denizens, organization leaders, or even faction rulers—and each of them have individual win conditions. As they each press towards their win conditions, the gameworld’s pot is stirred; a great bubbling of activities, reactions, rumors, and scheming rises to the surface. The campaign comes alive as the gameworld becomes too complex for any individual participant to understand in whole².

Because the actors can belong to any strata of society and have a wide array of different roles and mechanical implications, speaking about the characters in Braunsteins is confusing with 4-Point Semantics; using our augmented Controller Semantics can alleviate this.

Some of the Braunstein actors might be player-driven Controlled Characters (CC) and others procedure-driven Non-Controlled Characters (NCC). Many of them might be mechanically indistinguishable from ordinary denizens—Universals without any special qualities—while others may have class levels or other distinctions which separate them from the masses.

Some of these Braunstein actors may be played entirely “in the background” by eschewing session play; their plans may be of a wide scope with a long time horizon, requiring little up-close attention. However, some of the Braunstein actors may be Canonical characters, in every sense of meaning. They might be played by Controllers who frequently put them through session play, utilizing the “downtime” in between sessions to carry out their more abstract plans. The 4-Point Semantics would call these PCs, but they have explosive potential beyond contributions to conventionally played sessions.

The referees generally participate in TTRPGs in a fundamentally different way than other Controllers, but there is no a priori principle barring them from having their own Canonicals at hand—or other CCs undertaking non-canonical campaign play—as long as the campaign managers detect no conflicts of interest which could diminish the campaign itself.

Setting and Belonging

A Greek-mythology-themed campaign might have Hector as an NCC; this is a highly Organic-leaning character. But if a Controller were to direct his actions, we would consider this as moving Hector towards Imposed.

If Hector’s Controller behaved in an outlandish manner judged to be of poor fit to the campaign, the degree to which Hector is Imposed would grow. But consider other possibilities. If Hector’s Controller relinquishes control (i.e Hector becomes NCC), this is undeniably a step towards Organic—the game itself now controls Hector’s behavior. If the Controller’s outlandish behavior is skillfully used to summon support from the campaign’s other actors to back Hector’s schemes, this is also a move towards a more Organic Hector.

By the same token, Vlad the Impaler would be a severely Imposed character in this setting! Nevertheless, a skillful Controller could navigate and channel the campaign’s inner forces to become an Organic member of the action—not from any literary standpoint, but by using the narrow definitions we’ve provided.

Nuance Upon Nuance

Consider an example of a brand new player being handed a level 11 character to play. This is a subtle variety of Imposed that also leads us to question to what degree that character can be considered Canonical. The degree is different, for example, if we gave that player a level 4 character instead!

In some games, there is an explicit expectation that every Canonical will start from level 1 (or level 0 or whatever the baseline is) and earn their way into higher levels. Even if there is no explicit call to play this way, the attitude persists for many reasons.

A character which was brought up to a higher level as a result of campaign activity is much more Organic than one that simply sprang into existence at a higher level—a real character history is much more Organic than an implied one.

More Treasure, More Problems

In revitalizing intensive TTRPG supremacy, we see a great many things in a new light. Some of those things puzzle and challenge us. Others inspire. In the long run, we will inevitably find further problems inspiring further questions and solutions.

The concept of the continuous Braunstein brings about a richness of possible character roles and character interactions that makes conventional party-centric play³ seem simplistic in the extreme. Though parties of characters have a crucial role to play in these games, that role is only one piece of a large tapestry of actors and interaction. This writeup, and others like it, are an attempt to more concretely understand and convey what is possible in the greatest gameplay medium we have yet encountered.

---

Thank you for your readership! Primeval Patterns thrives on the basis of the sincere interest and support of hobbyists like you.

Some other interesting things:

• Follow the author on twitter.

• Check out BMD, a far-future wargame-infused TTRPG about slaughtering aliens.

• Think carefully when designing or using skill systems!

• Understand starting an ACKS campaign from scratch.

If you want to support this effort further, becoming a paid subscriber here on Substack is the easiest way. This directly supports our design research and the development of BMD.

As a bonus, paid subscribers can read extensive BMD development blogs and are guaranteed a PDF copy on BMD’s release.

1. See reflections on Chantsonian Patrons and Braunsteins.

2. See remarks about BrOSR campaigns.

3. See Deprogramming Design Inversions discussion about party-centric vs. gameworld-centric gameplay.