I tend to prefer the term “metagame games” rather than “thematic games”, but thematic sounds cooler. In these games, the rules describe the “world” in which play takes place. While world games often had a touch of metagaming, such as in Vampire’s humanity attribute or Call of Cthulhu’s Sanity characteristic, thematic games are full of such “tricks”.

Many of these games, because the authors paid attention to their rules, don’t have game aspects that were once universal. All of the previous games, D&D, post-D&D, and world-games, shared the same basic structure. That all of them used dice was only the most obvious and most quickly overthrown. Others included:

• There is always one player per character.
• There is no “winner”.
• The Game goes on forever.
• There is a game master and one or more other players.

Even James Bond 007, a game emulating a film and book series with a single protagonist, assumed multiple players with one game master. But all of those “rules” are broken by at least one of the games in this article.

Most importantly, all of the previous games had some form of what’s called Rule Zero: if the rules don’t make sense, ignore the rules. In these games, there’s not as much “ignore the rules”, and rule zero is either nonexistent or has less force. The rules are there because they describe how the game works: system matters. Of the games covered here, only the early games, Amber and Everway, contain a strong form of rule zero.

Often these metagame games will come with a specific world just as their predecessors did, but they don’t have to. Burning Wheel doesn’t have a world any more than AD&D did—probably less. But it does recognize that its rules create a specific kind of world. These games try to be aware of how their rules guide play. That makes their choices on how to reward and advance characters especially interesting.

Many of these new games have no character advancement: instead, they have a defined end that will occur within a few game sessions. Such games explicitly acknowledge, as earlier games like Creeks & Crawdads hinted at, that the games are not meant for “campaign play”. Some games aren’t expected to last more than one session. These games have no need for character advancement, putting them out of the range of this series.

In earlier parts of this series I tried to choose memorable games that either had some effect on the industry or showed an interesting trend. Here, we don’t always know which games will be remembered ten years later as the influential games, or twenty years later as the emblematic games. So I apologize in advance if you think I’m wrong. And some games that I think should be in here aren’t. These games are often published by individuals who run out of stock for long periods of time.

Also, because these game rules are about something, I will often make a statement regarding what I think the game is about. I make this statement based on my reading of the game rules, and I can be wrong about that, too.

Amber

Amber is, in a very strong sense, the father of Forge-like metagame games. When Eric Wujcik published Amber as “Amber Diceless Roleplaying”, White Wolf’s five reviewers (May/June 1992, issue 31) focused on the lack of rules in general. It was the “diceless” that generated controversy, but the rules themselves sustained it. That controversy extended throughout the gaming magazines of the time as well as gaming clubs, and continued through the emerging Usenet.

In Amber, the player characters were the extraordinarily powerful third generation of Amberites. While not as powerful as their parents, they were on average more powerful at any task than the most skilled human.

Amber characters were so powerful that the character improvement system had no rule for acquiring skills. If a player wanted their character to have a skill, they pretty much had it.

What Amber players paid points for were attributes, powers, allies, and artifacts. The most important of these were the attributes. Amber characters had four attributes: Psyche, Strength, Endurance, and Warfare. By default an Amber character had “Amber level” in each of these attributes: better than any human could have.

Players started the character creation process with 100 points. The first step in character creation was getting all of the players together to bid on attributes in the Attribute Auction. The character whose player bid the most was the best of their generation in that attribute.

At Amber level each attribute was ranked according to the bidding. If there were three bidders, there would be three Amber ranks for that attribute. The bids set the point cost for ranks in an attribute.

If the final bidding on Psyche was 43, 41, and 29, then Psyche had three Amber ranks: First Rank (43 points), Second Rank (41 points), and Third Rank (29 points).

Later in the character creation process, or later in the game using Advancement Points, other players could buy to those levels. If a player wanted to acquire Second Rank in Psyche for their character, they needed to spend 41 points. (If they had already reached Third Rank, they needed to spend 12 points to get to Second Rank.)

However, the Auction bidder was always best at that rank. The character whose player won Second Rank Psyche in the Attribute Auction had a better Psyche than the character whose player bought their character’s Psyche up to Second Rank later.

The winner of an Attribute Auction is unbeatable. Whoever wins for an Attribute is that Attribute’s ultimate winner (among the younger generation). First place is the only safe place. If you get first place in any Attribute Auction you will know you are the best of your generation. Anyone, after the Auction, can buy up, even spending as many points as the first place winner, but they can never beat, or equal, whoever gets first place.

Ranks were important, because there was no random action resolution in Amber. Each rank was so much better than the previous rank that no randomness was necessary: the better rank won.

The Amber auction system was designed specifically to create bitter sibling rivalries among the player characters.

The auction system also basically assumed that all players would be in the game from the start. Players who entered the game after the auction ended could never be the best at any attribute: they were treated as if they were present for the auction but didn’t bid on any attribute.

If players spent more than a hundred points during character creation, their character would have “bad stuff” equal to the amount they went over. Bad stuff was basically bad luck at every encounter. If players spent less than a hundred points, their character would have “good stuff”, or basically good luck at every encounter.

Notably lacking in the Amber ruleset was any means of handling skills. They were listed under “Character Freebies” along with Physical Description and Age.

…in Amber, if you want skills, you can get them. Why?… Time is always available. Any character with the power to travel through Shadow can always take a side trip to a “fast time” Shadow. For example, someone can spend eight years in a Shadow medical school and surgery residency, while only a lazy weekend passes in Castle Amber.… Rather than trying to list all the various skills you’ve picked up, it’s better to list the experiences that the character has lived through.

Similarly, “material possessions come ridiculously easy to Amberites… Objects are just objects, so help yourself.” Entire worlds counted as material possessions, because Amberites with Pattern could walk the pattern to bring worlds into existence.

Character improvement in Amber was pretty much the same, without the auction. At the end of every adventure players were awarded Advancement Points, possibly “a couple of points for every couple of sessions of play”. Wujcik estimated that the Patternfall War in the Chronicles of Amber was “worth somewhere between twenty-five and seventy-five advancement points. Call it fifty.” All players received an “identical number of points”.

Players could use these points to improve their Attributes, or gain Powers and other advantages, at the same cost as during character creation. There were, however, two major twists to Amber’s advancement system:

1. Players weren’t told how many points they received.
2. Players weren’t told what they bought with those points.

Players knew basically what they were buying, because they gave the Game Master an ordered list of what they wanted, along with a note saying “whether or not they are willing to take Bad Stuff” and how much Bad Stuff they were willing to take.

The GM looked at the point total for that adventure and went down the list. As long as there were enough points left, the character received the improvement. If there were not enough points left, the character might still receive that one improvement that went over the total: was the player willing to take enough Bad Stuff to afford that improvement? Then the character got that one, too, as well as the Bad Stuff. Leftover points meant Good Stuff.

Players didn’t know what their attributes or other capabilities were, or even if they had Bad Stuff or Good Stuff, though obviously some of this would come out during play.

This contributes to the mystery of the game, and makes engaging in player versus player conflicts a tad more interesting, since the players can no longer be certain (at all!) of their own scores.

Player rivalry in Amber was important. The world of Amber was one of intrigue and mystery.

Everway

In Jonathan Tweet’s 1995 Everway, players created heroes out of the four elements: Fire, Earth, Air, and Water. They also started the game with Powers, Specialties, and perhaps Magic. Heroes improved as their players learned “more about the setting, your hero’s abilities, your gamemaster’s style, and so on. This increased understanding helps your hero be more effective when faced with challenges.”

Heroes also gained “boons” during their quests. A boon could be “anything that benefits a hero, such as a magic amulet, esoteric training, a bonus to Element scores, healing herbs, and so on. These boons also make your hero more effective.”

The Gamemastering Guide gave more detail about boons. Boons were treasures found during the quest, rewards granted, and direct or indirect effects from the quest. Typically, boons had “a special effect that only works once” but they could also be a new power or other special ability.

Everway did not otherwise have an improvement system. The characters started the game with “abilities well above the norm”. There was no specific means to increase a character’s elements, gain new powers and specialties, or gain or improve their magic. When Jonathan Tweet suggested a simple experience system in response to queries on a discussion group, one of the posters asked why it hadn’t been included in the original. Tweet replied:

Because in over a year of playtesting, I never saw the need for it. In terms of game play, I still don’t see a need for it. But if people want it, who am I to deny it to them?

In response to a different system and its justification by a fan, Tweet wrote:

I would agree, and it’s exactly this sort of ad hoc, character-oriented development that I didn’t want to preclude with a standard, abstract system. Providing a lowest-common-denominator system (such as the system I’m describing) would discourage people from using more ad hoc, personal systems.

Everway was very freeform. While it included a rule zero there weren’t many rules to throw out. As in Amber, heroes generally succeeded when their abilities were better than their opponent (The Law of Karma). Everway also provided the laws of Drama (“the needs of the plot determine the outcome”) and Fortune (“a draw from the Fortune Deck determines the outcome”). Everway’s character improvement system, such that it was, required taking part in the drama. Heroes gained tangible improvements only through the events in the story.

Sorcerer

Ron Edwards’s 1998 Sorceror began what would be called the Forge style of role-playing games. The edition I have (I’m using the 2001 edition) includes the System Does Matter essay. It also includes a short description of the publishing process that Sorcerer went through, ending with “you can do it too”. His essays were influential to those writing this style of games. Every game that follows in this article either thanks Ron Edwards or notes Sorcerer as a major influence.

The game itself was not as upbeat as the essays. The characters were sorcerers who summoned demons. They had four Scores that they could improve: Stamina, Will, Lore, and Cover. “Cover” was the character’s personal life, and basically covered any normal skills that the sorcerer had. The character’s other scores had descriptions that could also be considered sources of skills.

The player was allowed to increase one of those scores for each full adventure. They had to make a Humanity roll against that score to increase the score; if their first choice failed, they could go to their second choice, and so on, until a roll succeeded or there wasn’t anything left to improve.

GMs should monitor how fast characters progress by experience. If they improve every adventure, you’re going too easy on them. Force Humanity down by making them desperate and thereby Summoning up things and Binding them in order to deal with their problems. Have no mercy; no Sorcerer protagonist is worth squat until he’s felt the breath of chaos on his ass.

Sorcerer was about summoning external powers to solve problems. It was about dealing with the dehumanization that entails. It was more about sliding into a soulless abyss than personal improvement.

In this role-playing game, becoming more effective through earning experience isn’t really the point.… The keystone of the sorcerous mind, above all else, is arrogance.… Nothing—literally nothing—is more important than what you want out of a situation, and whoever has to be Punished, Bound, or otherwise made to obey is by the Pit going to get it.

Sorcerer had one resolution mechanic: roll a number of dice against another number of dice. The side that rolled the highest number was the successful side. Improvements were rolled against Humanity. If a player’s character had a Will of 5 and a Humanity of 3, the player would roll 3 dice (Humanity) against 5 dice (Will) if they wanted to improve their Will by one point. If the Humanity side (the 3 dice) had the highest number, their Will improved. If the Will side (the 5 dice) had the highest number, their Will did not improve.

As in Call of Cthulhu’s Sanity, a low Humanity made further Humanity losses more likely. A character whose Humanity dropped to zero was out of the game. Every time a character contacted a demon, summoned a demon, or bound a demon, the player needed to make a Humanity roll against the demon’s Power. Failure meant a one-point Humanity loss. Characters also lost Humanity for “heinous acts, such as sacrificing the newspaper boy to Summon something or going on a killing spree for some reason.”

Characters could get Humanity back by Banishing demons whose Power exceeded the sorcerer’s Humanity. The character might also do “something that, in the GM’s opinion, confirms the character as a decent human being”. Humanity and its fluctuations was important enough to warrant a full supplement, The Sorcerer’s Soul.

Finally, characters in Sorcerer had Kickers, the reason that they were part of the game. If a character’s Kicker was resolved, the player could either retire the character or rewrite the character. Rewriting meant that the character’s Scores and their descriptions could change. “The only thing that must remain unchanged is Humanity.” While not necessarily improvement, Kicker resolution meant significant character development.

Donjon

Whenever I buy a copy of my own game from Lulu, I also try to pick up something of someone else’s, either a novel or another game. Clinton R. Nixon’s 2002 Donjon, with its fascinating sales blurb, became the first of the Forge-era games I read. It’s well worth picking up.

Donjon was, relative to AD&D, a very free-form game. Players directly affected the world through successful dice rolls for their characters. In other games, a successful Perception roll looking for a secret door meant that the character either found or didn’t find a secret door. It depended on whether such a door was there. In Donjon, if the roll was successful the door was there. The player then narrated the finding of the secret door according to the number of successes they rolled on their dice.

Characters in Donjon started the game with six Attributes (Virility, Cerebrality, Discernment, Adroitness, Wherewithal, and Sociality). These were rolled randomly and ranged from 1 to 6. Characters also had Flesh Wounds, two Saving Throws, one Main Ability, and four Supporting Abilities. The player had 20 points to distribute amongst these scores. No score could be more than 4. Finally, characters had Provisions (what the character carried) and Wealth (the character’s finances). The player set one of them to five and the other to three.

Each character had a class, as in D&D, but the player made up the class. A Donjon class was basically a name that the player made up, the Main Ability, and the Supporting Abilities.

Characters in Donjon gained experience that, when it reached specific totals, increased their level. Characters gained experience from winning combat:

This does not necessarily mean that they killed or knocked out the opponents, although it normally does. If they used strategy to get rid of the opponents, they can be considered to have overcome the opponents. If the opponents run away to fight another day, though—no experience for the characters.

Basically, the GM totalled the levels of all opponents, divided by the number of player characters, and awarded this as experience.

GMs also set Goal Awards for achieving “a goal that does not involve combat”. The character received an experience award equal to the Donjon level plus the difficulty of the task that achieved the goal.

And finally, “the GM may allocate one to three experience points to a player for entertaining everyone”. Normally, this meant one point, though if the GM “shed a tear” that could mean two points and if the GM “beat his chest” it could mean three.

Each level increase brought five dice that the player could apply as desired to Flesh Wounds, Saving Throws, and Abilities. None of those scores could be above level+3.

Characters also gained one point in one Attribute, or one new Supporting Ability, every third level. And they gained a new permanent possession or Magic Word every even level.

Level did not otherwise affect a character, though it made a good estimate of their capability relative to other characters.

The resolution system in Donjon meant each “side”—the player and the GM—rolling a number of dice. Whoever rolled the highest die “won”. The GM always received extra dice equal to the Donjon Level. Thus, climbing a wall on Donjon Level 1 was easier than climbing a wall on Donjon Level 6, unless the player had been improving the character’s climbing ability.

This is entirely on purpose. As you play Donjon, your character is not the only one to gain experience: you learn how to play as well. At lower levels, players should feel free to try out all sorts of ideas in play, and their chances of succeeding will be higher. As they rise in levels, their choices will have to become more focused.

Donjon characters could improve their capabilities by finding cool items just as in other games, by looting their defeated enemies. In Donjon the player chose what the character was going to find.

When a combat is over, all opponents that are knocked out or dead can be looted. The players must decide which character gets to loot each body. The player who rolls decides whether he is looking for money or items. If he is looking for money, he rolls the opponent’s Level in a Test against his own current Wealth. Any successes are added to the character’s Wealth score.

If the player is looking for items, he must decide exactly what he wants his character to find. The higher the Worth of an item, the less of a chance that the item will be found.… The player rolls the opponent’s Level in a Test against the item’s Worth. If successful, the character gets the item; if unsuccessful, nothing was found on the body.

Players could also add curse dice to the item they were looking for to make it easier to “find”. Curse dice made the item less useful “when doing some activity”. That activity was up to the GM, who did not have to tell the player what it was.

Donjon also provided a formula for experience points and level advancement that allowed groups to fine-tune how quickly new levels were gained.

Universalis

There were neither player characters nor game masters in Ralph Mazza and Mike Holmes’ 2002 Universalis. What it had were Components, Traits, Events, and Complications. Players controlled Components of the story, which could be “people, places, and things”. Players could establish Facts only of Components that they controlled or through a Complication. Players could seize control of any Component, including “characters”. All control over the story had to be paid for with Coins.

At the beginning of each scene, the players bid for “the privilege of framing the next scene” using their Coins. Each player had certain things that they wanted to happen. Players who wanted specific Components, Traits, or Complications in the story needed enough Coins to pay for introducing them, and needed to convince the other players not to challenge those Components when the player introduced them; or they needed enough Coins—either their own or from supporting players—to fight off any challenges to the introduction. Even interrupting another player required Coins.

They then also needed Coins or support to maintain that their vision of that Component’s place in the story.

Coins measure story power. If you spend Coins, you gain power over the story.

Players began the game with 25 Coins. At the end of each scene, each player received 5 Coins.

Players also received Coins for winning a Complication over some Component. Each Trait that the Component had that applied to the Complication meant an extra die for controlling the Complication. Players could also buy extra dice with Coins. The side that won the Complication received more bonus Coins added to their Wealth than the losing side. It was very cost effective to spend Coins—wisely—to create Traits that moved the story to the player’s desired end. This was the main (perhaps only) way for players to gain Coins besides the end-of-scene refresh.

Players could also lose coins for “egregious behaviors or abuses”. A player had to call for a Fine; all players voted. If more players voted for the Fine, the accused lost as many Coins as voted for it. If more players voted against the Fine, the accuser lost as many Coins as voted against it.

The revised edition contained optional rules devised by various groups who played the previous edition. Some of these adjusted the costs of different kinds of actions in order to encourage them. For example, where an Event normally cost one coin, there was a “free dialog” optional rule that made dialog between characters free “to encourage first person conversations between players”.

The Burning Wheel

Luke Crane’s 2002 The Burning Wheel was inspired in its first edition by games such as AD&D, Cyberpunk 2020, and Shadowrun. Its revisions were inspired by games such as Dogs in the Vineyard, My Life with Master, and Sorcerer. I’m using the revised 2003 edition, which expands Artha into a more complicated system than the original rules had.

In Burning Wheel, characters improved their stats (Will, Perception, Agility, Speed, Power, Forte) and their skills individually, by marking when they were used during play.

This advancement system gives the feeling of a living, sweating, breathing, bleeding being in the game. It is vital that the characters in Burning Wheel grow and change as they are played. As players and GMs call for tests for characters, marks are made on the ability’s experience log. Mark enough tests of varying difficulty, and the ability advances.

Where other games, such as Runequest, allowed for advancement rolls as long as a skill was used during the adventure, Burning Wheel advancement was not random but required several uses of the ability. Those uses did not have to be successful uses, but they did have to be of varying levels of difficulty. For skills, the character needed to meet a number of Routine, Difficult, and Challenging (“life and death”) tests of that skill. For stats, the character needed to meet a number of Difficult and Challenging tests of that stat.

Tests are only awarded to players when their characters act in appropriate and applicable in-game situations. Everything else just counts as practice.

Characters could practice to improve abilities, but improvements through practice generally took years to accumulate all of the necessary test types.

In Burning Wheel, a characters wealth and influence were also measured as a skill. Characters had Resources and Circles to measure “how economically and financially viable a character is” and “the character’s social influence”. Perception, Resources, and Faith required successful tests, not merely having tests, to improve.

Abilities improved immediately when the requisite number and kind of tests were accumulated.

Doesn’t matter if it’s in the middle of a sword fight or while researching a new spell. Once the test is earned, the ability increases one rank. There’s nothing more pleasing to my ears than a player crowing, “I just learned something!”

Burning Wheel also awarded players three different kinds of Artha: Fate, Persona, and Deeds. The player could use artha points to affect die rolls in the game.

If the mechanics are the fuel, players are the fire, and artha is the spin on the wheel and the coruscation of the flames. It’s an interface and interaction between GM, player and system. GMs reward players for driving the story forward, and players reward themselves by using their characters to further complicate and dramatize the story at hand.

Players were awarded artha for their adherence to and use of their character’s Beliefs, Instincts, and traits.

The flow of artha creates a connection between player, character and GM based on themes and issues important to their game.… Setting out BITs [Beliefs, Instrincts, Traits] for his character, a player states to the GM and the group what his goals in play are for this character. He lets everyone know how and when he wants to be rewarded for playing his character.

It was the GMs responsibility to watch players for artha-earning actions by their characters. Fate, for example, was earned for acting out a Belief in “a convincing and entertaining manner… [that] serves a purpose and drives the game forward”. Fate was earned for “playing Instincts when such play gets the character in trouble or creates a difficult or awkward situation.” And Fate was earned for “using traits, but only if invoking that trait alters the direction of the story in an unforeseen way or makes life difficult for the character.”

Personal artha was more difficult to acquire. Players gained Persona artha for their character by capturing the mood of the table and driving the story forward, for playing out an inner turmoil “where his Beliefs, Instincts and traits conflict with a decision he must make”, for accomplishing personal and group goals, for being the “most relied upon” character, and for being voted the “most valuable player”.

Deeds artha were the most difficult to acquire and were “reserved solely for accomplishing goals larger than a character’s personal agenda.… Such an accomplishment must come with some cost or sacrifice”.

Players spent Fate to reroll sixes (Burning Wheel used six-sided dice), to reroll a traitor (unsuccessful die), or to shrug off minor pain. They spent Persona to gain a bonus die for a test, to shrug off wound penalties, or to attempt to survive a mortal wound. They spent Deeds to double the dice from “a single stat, skill or attribute” in a test, or to reroll all traitors in a test.

An optional rule allowed spending a Persona point to turn a dangerous failure into a complicated, expensive success. A character about to die from a failed roll to Avoid a troll’s hammer, for example, might instead:

spend a Persona point to duck the blow, but to have stumbled back and nearly fallen from the ledge? My character would have dropped his sword, and he’d be hanging on for his life as the troll advances.

Earning artha was not completely its own reward. Spending it helped the character advance, because “bonus dice from artha never count toward the total dice rolled for purposes of advancement.” Thus, artha made it easier to live through meeting that Challenging test necessary to advance a skill or stat.

Further, using artha for a skill or stat roll moved the character “toward an epiphany”. After a specific number of artha-enhanced tests, the ability “automatically and immediately shade shifts one step lighter.” Shifting a shade on an ability made the character much better at that ability. Normally, players needed to roll a 4 or better on d6 to have a “success”. If they rolled five dice, and three came up four or better, they had three successes and two traitors. Having to roll 4 or better was “black” shade. On a shade shift to “gray” shade for an ability, the player needed to roll only 3 or better. On a shade shift to “white” shade for an ability, the player needed to roll only 2 or better.

Players were thus encouraged to focus their artha spending on specific abilities in order to shade-shift those abilities.

Fastlane

Alexander Cherry’s 2004 Fastlane reduced characters to their most basic function: fast-burning vehicles for narrating an exciting story.

Nothing in Fastlane is solid—the numbers on your character sheet are as fluid as alcohol, ever-changing in response to circumstance. Nothing stays the same, not even the things you care about most.

Players in Fastlane had a set number of chips, left over from character creation. They created their characters by putting chips into their characters’ facets (People, Assets, Nerve, Guile, and Sobriety), styles, and life connections.

Fastlane didn’t completely jettison the game master, but did give its croupier only as many chips as the players had. Each participant in a conflict placed chips on a roulette wheel, bidding a number of chips up to the value of the facet that applied to the conflict plus any lives laid on the line, with an extra free chip if a style applied, and another extra free chip placed on the character’s lucky number.

The croupier spun the wheel, and players collected their winnings (or bemoaned their losses). A player’s “take” was whatever they did not bid from the maximum bid allowed (their facet plus modifications) plus whatever they won. Players then allocated their take to “the various contests in the conflict”. The player who allocated the most chips to a contest won the right to narrate that contest’s resolution. Players could also “burn” facets, styles, factions, favors, and lives to gain chips to use in a contest. Burned facets, factions, favors, and lives were reduced by 1 permanently; burned styles were lost for good.

If, at the end of a conflict, players still had chips in their take, they could either put those chips in their bank or they could use them to improve their facets and factions, and to gain new styles.

Whether burning or improving numbers, players needed to be able to reasonably narrate the loss or improvement with “an acceptable in-game reason for the change related to the conflict or its resolution”. Players could also spend a chip to move points from one score to another score.

The croupier also stood to improve and lose capability as the game progressed. The croupier couldn’t bid at the roulette wheel: “the croupier’s only source of income is the losses of the players.” The croupier could spend their chips freely to increase the difficulty of contests, but once the croupier’s chips were gone the croupier was no longer able to add such obstacles until the croupier gained chips from a player loss.

Fastlane had a defined end for individual characters: a character whose life connections all burned or were reduced to zero was burnt-out, and had to be narrated out of the game “in their very next conflict”. Players were encouraged to “burn as much of the character as possible” in this conflict, “as the blaze of glory is one of the things Fastlane is all about”.

Shadow of Yesterday

In Clinton R. Nixon’s 2004 The Shadow of Yesterday, characters gained experience points through their Keys and by reaching story goals set by the Story Guide. (I’m using the 2005 revised edition of The Shadow of Yesterday.)

Keys are the primary method of increasing a character’s abilities. These are goals, emotional ties, or vows a character has. By bringing these into the story, the player gains experience points (XP) she can use to advance the character, increasing pools and abilities, or learning new Secrets and Keys.

A character with the Key of Conscience, for example, gained 1 XP for helping someone who couldn’t help themselves, 2 XP for defending someone “with might who is in danger and cannot save themselves”, and 5 XP for taking “someone in an unfortunate situation and [changing] their life to where they can help themselves.”

There were two kinds of Keys: motivations, and everything else. Motivations had 1 XP and 3 XP levels. Everything else had 1, 3, and 5 XP levels. At their lower (1 XP) level the character could gain “everything else” XP only three times per game session, but the other levels, and motivations, were unlimited. Experience points were awarded immediately, as soon as the Key’s mechanisms came into play.

Every 5 experience points gave the character an advance. Advances could be used to improve a character’s abilities or pools (Vigor, Instinct, and Reason), and to buy new Secrets or Keys. Improvements cost from 1 to 4 advances. Characters began the game with 5 advances.

Players could gain immediate experience for their character by “buying off” a Key. Each Key stated the means by which it could be bought off. For example, the Key of Conscience could be bought off by ignoring “a request for help”. Players were not required to buy off a Key when the buyoff action occurred. If the character performed the action and the player chose to invoke the buyoff, the character lost the Key and gained 10 XP.

This encouraged life-changing actions on the part of the characters. Since a new Key cost 1 advance, which cost 5 XP, buying off a Key was basically free: a player could buy off a Key and then buy a new Key while coming out one advance ahead. This did, of course, change the character conception. It also might bring the character closer to accidental transcendence and removal from the game.

Players were forbidden from improving the same thing twice in a row, or from buying two Secrets or two Keys in a row. (The latter prevented players from buying off a Key and then buying two Keys with the 10 XP). They were also forbidden from buying a previously-bought-off Key again.

Besides Key experience, characters also gained experience for meeting Key Scenes. A Key Scene might be “confronting the wild beast” or “meeting the overlord”.

Whenever a character is present in a Key Scene, she earns one to three experience points, as determined beforehand by the Story Guide. These experience points are given as soon as the scene is over.

Groups could easily fine-tune improvement speed, by changing the cost of advancements. Unlike Nixon’s Donjon, this mechanism was simple and easy to understand. Change advancements to cost 10 experience points instead of 5, and improvements happen half as often.

The Shadow of Yesterday also had a means by which powerful characters were removed from play. There were eight levels of success in the game, from 0 (Failure) thru 5 (Legendary), 6 (Ultimate), and finally 7 (Transcendent). If a player rolled a Transcendent success on an ability check, the character reached Transcendence.

It signals the end of a character’s story, and is a special occasion for that character’s player. With this result, the player should feel free to narrate the outcome of her roll herself, with any help she likes from the other players and Story Guide.…

The story should immediately focus upon the transcendent character. She has just accomplished a feat that will be spoke of by her companions forever, and the day is her. Within 24 game hours of the moemnt she became transcendent, her story will be over. The character may die; she may retire for a quiet life; she may disappear over the hills; or she may become something else entirely. Her story will end and she will be retired from play.

Polaris

Ben Lehman’s 2005 Polaris was a beautiful game, and like the others in this section it differed considerably from “standard” role-playing games.

The goals of playing Polaris are no more and no less than the goals of a child’s play-pretend—to create a good story, to explore an imaginary world, and to have fun doing it.

That’s not too different from what older games wrote in their introductions. If Polaris had been like those other games, it would have then gone on to describe how each player creates a character. Polaris had one protagonist per player. However, in Polaris, each player took a different guiding role towards each protagonist. There were four players and four protagonists. Each player was the Heart of one protagonist, the Mistaken for another protagonist, the New Moon for a third protagonist, and the Full Moon for a fourth protagonist.

The Heart was the closest to the traditional player character and was often treated as such in the rules. The Heart of a protagonist guided the protagonist. “When there is conflict, the Heart negotiates in favor of the protagonist.” It would be more strictly true, however, to say that the Heart negotiated in favor of a strong place for the protagonist in the narrative. The Heart’s negotiations were often not to the protagonist’s advantage, except insofar as they ensured that the protagonist held a powerful place within the story.

The Heart also managed the protagonist’s Cosmos, which was basically the character sheet.

The Mistaken guided the antagonists, such as the demons and environment. The Full Moon guided “the secondary characters with whom the protagonist has a primarily societal or hierarchical relationship” such as knights, senators, and distant family members. The New Moon guided “the secondary characters with whom the protagonist has a primarily emotional and personal relationship” such as lovers, close friends, and close family members.

Each of the four guides took on some of the roles that would have been handled by the game master in other games. The Mistaken, Full Moon, and New Moon were closer to GM than the Heart in the sense that they handled non-protagonists and environmental conflict.

As the story’s focus moved from one protagonist to another, the player roles shifted. The Heart of one protagonist would be the Mistaken of another protagonist, and so on.

Each protagonist had three Values: Ice, Light, and Zeal. Ice and Light started at one, Zeal at four. Ice and Light were closest to a traditional statistic. Ice measured the “strength of the knight’s relationship to the world and society”, and Light measured “internal strength and prowess”. Each of these improved as the game moved forward.

Zeal was closer to statistics such as Call of Cthulhu’s Sanity. As the game progressed, Zeal faded. When Zeal reached zero, it was replaced with Weariness. Weariness grew as the game moved forward, and if it reached five the protagonist was corrupted to the Mistaken.

Protagonists also had characteristics listed within four Themes:

• Offices: formal roles in society, such as being a knight, being a senator’s aide, etc.;
• Fates: events, ideas, and relationships important to the protagonist’s story;
• Blessings: things and people that the protagonist can count on in the battle against the Mistaken, such as swords, boon companions, and yearnings;
• Abilities: personal characteristics, skills, and talents of the protagonist.

As in Universalis, Polaris put control of the narrative into play as an explicit player goal. Besides the Key Phrases used to handle story paths, Hearts and Antagonists could exhaust one of their protagonist’s Themes to make a statement about a conflict. They could do so once each until that particular Theme was refreshed.

In Polaris, advancement was “about how protagonists change and grow, how storylines move towards their final tragic endings and what shape that ending might take.” Experience, for example,

marks the progression of the knight from a untested novice, full of zeal and will to fight, to a seasoned veteran, skilled but weighed down by the horrors he has seen.

Experience was checked whenever the Heart lost a conflict roll or had their protagonist act “in some way indicative of sympathy for a demon or the demons as a whole, hatred of a person or the people as a whole, apathy, callousness, cynicism, doubt, or despair.”

The Mistaken called for an Experience check for the protagonist if one of those criteria were met. On an Experience check, the Heart rolled a d6. If it was less than or equal to the protagonist’s Zeal or Weariness, the Heart added one to the protagonist’s Ice or Light, up to a maximum of five. If the protagonist had Zeal, it dropped by one. If the protagonist had Weariness, it increased by one.

On an advance, the Heart could also remove aspects from or add aspects to Themes.

If, on the other hand, the die roll came up greater than the protagonist’s Zeal or Weariness, their exhausted themes were refreshed.

When a protagonist switched from Zeal to Weariness, they also gained the Fate Theme “Betrayal of the People”. Protagonists were only allowed to die once they had Weariness. If the protagonist did not die before their Weariness reached five, they became a demon. They were no longer considered a protagonist, but unlike games such as Call of Cthulhu were not out of play:

Even though the character has turned permanently to a foul demonic creature, it does not mean that the character has left play. This character, a former protagonist, must now return to play as a demonic antagonist. Further, whenever the character is present in a scene, the character’s old Heart still guides him.

Either the Heart or the Mistaken could “frame one last scene focusing on this protagonist which shows the exact moment of his descent.”

Summary

Some of these games explicitly recognized that the form of the award also guides play. Donjon specifically stated where play should progress: from a general “try anything” approach to a focused application of the abilities the character has learned.

Burning Wheel turned the spending of rewards into further rewards, but rewards that encouraged focused character concepts.

Fastlane vastly reduced the power of the game master, making the world basically another character in the game subject to some of the same character improvement rules as the other players. If the croupier ran out of resources, “everything becomes easy for the protagonists, and very little can stand in their way.”

Role-playing games often have advice for GMs on how to hold back on experience points or give more experience points, but Clinton R. Nixon began experimenting with putting experience accelerators and brakes into the rules. Donjon’s involved a mathy formula, and then The Shadow of Yesterday made altering the improvement rate easy.

Games also started looking more closely at what happened when characters were removed from play. Fastlane and The Shadow of Yesterday—games in which character removal occurred for thematically opposite reasons—both turned significant narration over to the player whose character was about to disappear.

In games with a Game Master, a character with a Sanity-style score that reached its limit might be taken out of play and controlled by the GM. In Polaris all players shared the role of GM, so a character whose Weariness reached the maximum basically remained in play.

One of the interesting aspects of Polaris, from a character “improvement” perspective, is that Novices with high Zeal quickly gained advances at first, because it was easy to roll four or less on a d6. As their Zeal dropped, advances came more slowly. And as they lost their Zeal to Weariness, advances also came slowly. But it came more and more quickly as their Weariness increased. The mechanic meant that protagonists were held for longer periods of time in the middle areas. They didn’t remain raw Novices for long, and once they became very weary Veterans, the end came quickly.

This contrasted with most other advancement mechanics, which tended towards slower and slower advancement as the character grew more powerful.