Jouer sans jeux, Part II

26 Jan, 2026

A continuation from the previous section of the essay, this time looking at the second (of two) qualities of role-playing games that set them distinctly apart from other kinds of game.

Facts and Values

Chess, as a game, provides players with several facts. Play happens on a board, the board is an eight-by-eight square grid, each player has access to a certain set of pieces arranged in a certain way, individual pieces move and capture in specific ways, and so on. These things are true for chess. There are other versions of chess that amend these facts in various ways, but these we must construe as exceptions to and variations upon the baseline facts of the game, and ones that constitute their own set of facts that merely replace the expected ones. Additionally, chess provides players with a set of values: it is good to checkmate the opponent's king. Hume would perhaps construe these as two different kinds of statements, is-statements on the one hand, and ought-statements on the other. The board is a grid; one ought to checkmate the opponent's king. Notice that the ought-statement, the inherent value judgment of chess, is one and the same as its win condition. The mechanism by which chess passes judgment upon player actions is therefore through the lens of its win condition. An action is virtuous insofar as it contributes to its player's ability to win the game; a move is unvirtuous insofar as it inhibits its player's ability to win. This is therefore, at base, an ethical system unique to chess—though every game with a win condition can be understood to instantiate its own ethical system, which issues forth necessarily from its win condition to frame all possible actions in the game.

Hume would also likely point out that ought-statements do not follow necessarily from is-statements, that one cannot derive values from facts. This observation isn't particularly salient when talking about chess, as chess doesn't attempt to justify its win condition in the first place, it merely asserts its win condition as part of the game rules, alongside the facts of the rules. However, it becomes quite important when we view tabletop role-play through a similar lens. From its very beginning (or nearly the beginning, depending on how you categorize the immediate antecedents to Dungeons and Dragons in 1974), tabletop role-play has eschewed win conditions. Instead, in those early days play proceeded as a functionally endless "campaign," players and characters regularly falling in and out of the action, held together by the fact of a persistent, living world. Even now, though campaigns have taken on myriad new shapes over time, tabletop role-play is nonetheless marked by a lack of win conditions. Play must therefore be assumed, even in those earliest days, to happen outside of the tyrannical judgment of the game system.

Combat

This is, of course, not entirely true. Taking smaller instances of play inside of the grand structure of Dungeons and Dragons, for instance, we can occasionally glimpse a kind of ad hoc win condition. Combat is perhaps the most obvious example. One can be said to win combat if one makes it out alive: "One ought not to die," then. A loss condition is as good as a win condition for instantiating a value system in a game. We could go further, and observe that these parts of tabletop role-play where an implicit win condition pops up (such as "one ought not to die") are also the places where the rules take on a shape more akin to board games or video games. It's simple enough to make the connection: a win condition, being itself constitutive of an ethics, brings a necessary rigidity of form, such that player actions can be legible to the game. Which is to say that, in order for the win condition to come to bear on player actions in combat, play must understand itself in terms of that win condition—at least for the period in which the win condition is relevant. Therefore combat rules in tabletop role-playing books like Dungeons and Dragons tend to look more like a board game (more like chess, for instance) than do the rules elsewhere. And similarly for play, which is generally more structured and formal in combat situations than in non-combat play.

But we must understand this local win condition as an exception that illustrates the larger reality that tabletop role-play is a kind of play with a unique relationship to value judgments. In chess, one must take on the value system of the game as a precondition of playing it, insofar as the win condition is a—perhaps the—central rule of chess. (One could, of course, play in "unvirtuous" ways on purpose—one thinks of grandmasters who play bad or even nonsense openings against weaker opponents in an effort to stave off boredom—but those actions always already exist inside the explicit value system of chess as you have agreed to play it. A player maintains the prerogative to do as they please, but not to do so unjudged by the game.) In tabletop role-play, there is no such value system to take on because there is no win condition in the final sense, and so players must provide their own.

Put differently: in answer to the question "How is the game going?", one could feed the in-progress board state of a chess game into a chess engine and receive a full mathematical analysis of the game, and this analysis would, as a matter of course, be entirely in terms of the win condition of chess. One could do no such corollary thing with an in-progress game of Dungeons and Dragons, not only because the game state is infinitely more complex and largely incommunicable, but more critically because there is no final win condition by which to make such an analysis. A human is required—a player, or a group of players—to make such a judgment, who can bring their own value system to bear in assessing their own play.

Advancement

Let's take a moment to address the elephant in the room: an advancement system common to most versions and clones of Dungeons and Dragons generally known as "Gold-as-XP." Because very few people talking about this issue have read it, I provide here the relevant text on experience points from the original game as released in 1974:

Experience Points: Experience points are awarded to players by the referee with appropriate bonuses or penalties for prime requisite scores. As characters meet monsters in mortal combat and defeat them, and when they obtain various forms of treasure (money, gems, jewelry, magical items, etc.), they gain “experience.” This adds to their experience point total, gradually moving them upwards through the levels. Gains in experience points will be relative; thus an 8th-level Magic-User operating on the 5th dungeon level would be awarded 5/8 experience. Let us assume he gains 7,000 Gold Pieces by defeating a troll (which is a 7th-level monster, as it has over 6 hit dice). Had the monster been only a 5th-level one, experience would be awarded on a 5/8 basis as already stated, but as the monster guarding the treasure was a 7th-level one, experience would be awarded on a 7/8 basis thus; 7,000 GP + 700 for killing the troll = 7,700 divided by 8 = 962.5 × 7 = 6,037.5. Experience points are never awarded above a 1 for 1 basis, so even if a character defeats a higher-level monster he will not receive experience points above the total of treasure combined with the monster’s kill value. It is also recommended that no more experience points be awarded for any single adventure than will suffice to move the character upwards one level. Thus a “veteran” (1st level) gains what would ordinarily be 5,000 experience points; however, as this would move him upwards two levels, the referee should award only sufficient points to bring him to “warrior” (2nd level), say 3,999 if the character began with 0 experience points.¹

Much of it is just rules-fussing, but the takeaway is that characters gain experience points for performing two basic activities: killing "monsters" and gaining "treasure." Additionally, experience points gained through treasure seems to greatly outstrip that from killing monsters, though presumably you'll be killing the monster in the process of taking its attendant treasure, so perhaps it's better to construe the two activities as one gesture instead. We'll call that gesture "violence." In exchange for performing this specific kind of violence, characters advance in either two or three specific ways (depending on their class): "Dice for Accumulative Hits," "Fighting Capability," and (for Magic-Users and Clerics) "Spells."²

We might first understand advancement in Dungeons and Dragons as a systemic incentive for players to have their characters trek into the wilderness, such as it may be, and do violence upon the denizens thereof. Certainly we can assert with some confidence that the authors of Dungeons and Dragons expect characters to venture into the wilderness and exact this particular kind of violence. Further, if we construe this incentive structure as a kind of local win condition, similar to how we framed combat above, then the whole becomes constitutive of an ethics in which characters ought to do this violence. However, in the case of Dungeons and Dragons as it was published in 1974, the justification for such a reading is somewhat tenuous. We might just as easily understand the incentive structure in terms of simulation and verisimilitude: if one chooses to behave violently, one becomes therefore better at exacting that same kind of violence. This is borne out in the advantages gained from advancement, which are relevant chiefly, or even only, to violence. A character who kills monsters and takes their treasure does not—in 1974 anyway—become better at spinning yarn or composing symphonies; they become better at taking bodily harm, and exacting bodily harm upon others. In this simulation-focused framing, the incentive structure is not purposed to pass judgment one way or another upon the ethics of violent acts (the way a true win condition would), but instead perhaps to instantiate the self-reinforcing nature of violence, the very real way in which violence infects its actor, making them increasingly violent.

As an example of how this very same type of incentive structure could be used more substantially by a text to establish violence as an inherently virtuous act, we need only look to later editions of the same game. The second edition of Advanced Dungeons and Dragons, for instance, includes rules (admittedly listed, as are many of the rules of that edition, as "Optional") for "Proficiencies." The book defines a proficiency as "a learned skill that isn't essential to the character's class." It further divides proficiencies into "weapon proficiencies" and "non-weapon proficiencies." Non-weapon proficiencies include skills such as brewing, cobbling, pottery, mining, blacksmithing, weaving, and singing. As the character collects experience points, they increase in level, and thus opportunities to gain non-weapon proficiencies.³ The scenarios in which a character gains experience points are not precisely the same as in the 1974 text, however:

All characters earn experience for victory over their foes. There are two important things to bear in mind here: First, this award applies only to foes or enemies of the player characters—the monster or NPC must present a real threat. Characters never receive experience for the defeat of non-hostile creatures (rabbits, cattle, deer, friendly unicorns) or NPCs (innkeepers, beggars, peasants). Second, no experience is earned for situations in which the PCs have an overwhelming advantage over their foes. [...] The characters must be victorious over the creature, which is not necessarily synonymous with killing it. Victory can take many forms: Slaying the enemy is obviously victory; accepting surrender is victory; routing the enemy is victory; pressuring the enemy to leave a particular neck of the woods because things are getting too hot is a kind of victory. The creature needn't even leave for the characters to score a victory. If the player characters ingeniously persuade the dragon to leave the village alone, this is as much (if not more) a victory as going in and chopping the beast into dragonburgers!⁴

This is, of course, accompanied by a rather arcane procedure for determining precisely how many experience points should be awarded for overcoming a foe, based on its Hit Dice and and additional (combat) abilities. The book does also encourage gamemasters to award experience points to players for clever and engaged role-play, as well as for certain activities related to the character's class, and for "completing" an adventure. Outside of an ambivalent note in a callout box stating that one might optionally allow for characters to gain experience points for the value (in gold) of their non-magical treasures,⁵ there is no mention of gold-as-XP. On balance, though the range of possible activities granting advancement has been somewhat expanded by the time of the 1989 game, the overall effect remains similar to that of 1974: a character advances chiefly through violence. The substantial difference between the advancement systems in these two books, then, is in the benefits of the advancement. In 1974, characters do violence to become better at violence; in 1989, characters do violence in order to become better at all sorts of things, in the form of non-weapon proficiencies. That is, by the time of Advanced Dungeons and Dragons 2nd Edition, doing violence makes a character not merely a more violent person, but a better person in any number of ways unrelated to violence.

Going back to the game as it was published in 1974, we are left to ask a simple question: if the advancement system exists as a way to simulate, in the abstract, characters becoming more skilled at the same violence they exact upon monsters, why does the book not also include corresponding advancement systems for non-violent activities? We might assume that violent advancement exists in the book specifically because violence is de rigueur in play, and that is likely correct. However, this conclusion implies that non-violent play is somehow less central to Dungeons and Dragons, or even missing from it completely, which is a suspect claim at best—after all, as we established previously, player agency is not (and cannot be) limited by the structure of the game book or its systems.

So, a more pointed formulation of the question: why does Dungeons and Dragons feel it necessary to systematize advancement for violent activities, and not for non-violent ones? From here, one might simply point to the way advancement in Dungeons and Dragons proceeds naturally from the structural necessities of its combat system, given the attendant local win condition discussed earlier. In an effort to make player actions legible to the implied win condition of combat, the advancement system is quite useful, insofar as it rewards those actions that prove effective in that context—and, importantly, that those rewards are relevant only to that context. Advancement, then, becomes only one additional piece of the ethics of the combat system, instantiated ultimately by its win condition, and reinforcing a necessary abstraction familiar from board games and video games. This abstraction therefore moves play away from the form-specific qualities of tabletop role-play. One might contrast this with Traveller, which includes no non-diegetic advancement (all character advancement, violent or not, is accomplished by spending some number of years in training or at school).⁶ While the combat of Traveller has a similar local win condition to that of Dungeons and Dragons, and therefore employs similarly abstract systems for combat, the books nonetheless stop short of providing similar abstract advancement as a special case for violence. Consequently, Traveller is a game that feels, to this author at least, more concrete and grounded in the fiction and in role-play.