One of the things for which Numenera (and the Cypher system as a whole) is rightfully lauded is how easy it is for the GM to prep material for the game. The cornerstone for this is creating stats for NPCs: You can literally just say, “He’s level 3.” And that’s it. You’re done. By assigning that single number, you know everything you need to know in order to run the NPC.

(In this, Numenera is actually quite similar to older editions of D&D, where assigning a Hit Die to a monster was basically 95% of what you needed to do. But Numenera streamlines the process even further.)

This is nice for pre-game prep, but where it really empowers the GM is during play: If the PCs go somewhere unexpected, it’s trivial to keep up with them. If you have a cool idea for a creature, you don’t need to table it until you have time to stat it up. The creative act of adding a character to the game world is closely wedded to the act of realizing that idea mechanically.

Experienced GMs will, as they get comfortable with a system, figure out how to improvise stat blocks no matter how complicated they are: They’ll know the key stats they need to decide up front, and they’ll learn the various options well enough that they can sort of lay them in on-the-fly. For example, when I’m running D&D 3rd Edition I’ll jot down the skill point total for an improvised NPC and basically “spend” them when it feels appropriate for the NPC to have a particular skill. You can do something similar with spellcasters and their spell lists. What you can see here is that Numenera basically gives even brand new GMs the same ability that experienced GMs need to spend time mastering in more complicated systems.

So why would anyone choose a more complicated system? Why wouldn’t every game go with a simple “pick one number and you’re done” approach?

Because complexity can be leveraged to make the game cooler. It’s basically the same reason that every board game isn’t “roll 2d6 and race each other around a board with blank space.” Simulationists appreciate being able to make more detailed or consistent mechanical models. Gamists appreciate the variety of tactical challenges varied stat blocks can create. And so forth.

FRACTAL NPCs

Okay, so Numenera trades the advantages of complexity for the advantages of simplicity.

Well… not quite. Because this is where Numenera does something very clever, in a way that is often overlooked as people focus on its really fantastic “easy prep” features: When you’re statting up an NPC, you don’t have to stop after saying, “Level 3.” You can keep adding layers of mechanical detail. You could give a Numenera NPC as much detail as a D&D 4th Edition stat block — replete with differentiated skills and a grab bag of special abilities — and the system lets you seamlessly do that. (And, importantly, provides just enough mechanical structure so that these additional details are mechanically relevant.)

The way this works is that “level 3” remains the baseline for the NPC, and everything else is an exception to that baseline.

You can get a good sense of the different ways you can push (or choose not to push) the system by looking at the creature stat blocks in the core rulebook or the Ninth World Bestiary. But you can actually push it even further than that: You see value in having an NPC with a different rating in thirty different skills? You can do that. And just because you do it with one NPC, it doesn’t mean you need to do it with the next NPC.

I’ve come to think of this as the Numenera stat block being “fractal” in nature: The closer you look at it, the more detail you can see. And what’s interesting is that it can also be selectively fractal; you can look for more detail in one aspect of an NPC, while allowing other aspects of the NPC to simply default to a broader and simpler structure.

You may notice that this also neatly formalizes what the experienced GM described above is doing an informal fashion: Starting with a broad definition of what the character is capable of and then selectively adding detail (specific skills, specific powers, etc.) as desired and/or needed.

Agnarr beamed: “A dog!” He caught it deftly in his hand.

The dog continued struggling, trying ineffectually to claw and bite at Agnarr. It also continued its shrill barking: Yip! Yip! Yip! Yip! Yip! Yip!

Tee grimaced. “Agnarr…”

Yip! Yip! Yip! Yip! Yip! Yip!

“It’s a dog! Not a real dog… but a dog!”

Yip! Yip! Yip! Yip! Yip! Yip!

Tee grabbed the dog from Agnarr’s hand and smashed it to smithereens on the floor. Agnarr’s face fell…. but at least the incessant yipping stopped.

Arguably the most powerful and memorable moments you will experience at a gaming table are those in which a PC faces a crucible of character: Where the outcome of a particular moment will not only change the direction of future events, but will in a very meaningful way rewrite who the character is on a fundamental level.

Often these moments will blindside you: You won’t even realize the significance of what’s happening until you’re abruptly in the moment and the dial has suddenly cranked up to 11.

And in some cases you may be sitting at the same table — or even running the game — and not realize that a character crucible is happening right next to you until long after the fact. This is because, during a roleplaying session, everyone has a unique and privileged view of the game: As a GM, I am gifted with near-omniscience and can see all that is happening from a vast perspective. But each player is given the exclusive ability to experience what’s happening inside the head of their character.

Take, for example, this yipping dog.

At this point in the campaign, it had become a running gag that Agnarr (a) wanting a dog and (b) could not find a goddamn dog. Speaking of the unique path of actual play, this running gag was not planned: It emerged organically because every time Agnarr would go looking for a stray dog in Ptolus he could completely flub his Animal Handling check. (Hence the quote at the end of Session 20A: “It’s like there are no damn dogs in this entire city!”)

That brings us to Pythoness House, where the PCs find a porcelain dog that was put there by Monte Cook:

Six of these statuettes remain intact. Due to the power of the spirits inhabiting this place, they leap off and attack 1 round after anyone enters the room. Each figurine is a Tiny animated object. Most are statuettes of people, although one is a dog and one is a winged angel. The angel figurine flies rather than walks.

The yipping was my idea, and it was probably crucial to this moment happening because it was really annoying and almost certainly prompted Tee to take the unilateral action of smashing the dog.

AGNARR’S TWO PATHS

At this point in his life, Agnarr was deeply discontented: He’d lost his village. He’d lost a chunk of his life (and, with it, seemingly any chance of recovering his village). He found the crowded, confusing streets of Ptolus disconcerting and frustrating.

When Seth took over the character of Agnarr, he saw the conflict the character was in and he responded by briefly sketching out two potential paths the character might follow as they leveled up: On one path, Agnarr would multiclass into the Tactical Soldier prestige class and specialize in feats and abilities that would make him a natural leader for the team; coordinating the actions of others and enhancing their achievements through his presence.  And on the other path, Agnarr would multiclass into Frenzied Berserker, consumed by his rage to the point where he would even become a hazard to his companions.

In the present moment, however, Agnarr wanted a dog. It was a fairly definitional desire that had been established by Dave, his original player, and Seth had continued that pursuit. What was a running joke to me and the rest of the table (including Seth as a player) was actually really serious for Agnarr as a character: All he wanted was an animal companion; a faithful hound that would remind him of the simple and natural life that had been taken from him. But the confusing, frustrating city denied him even this simple thing.

In this moment, with the yipping porcelain dog, Agnarr felt a legitimate emoment of simple joy: The city had finally given him one thing. It wasn’t exactly what he wanted, but it was still a kind of victory.

And then Tee smashed the dog.

Right in front of his face.

And none of his “friends” cared.

They laughed about it.

And that was it. The choice had been set. Agnarr was turned towards the path of rage and alienation. It wasn’t like a switch had been flipped — he didn’t suddenly Hulk out or anything — but he had been set into motion.

Later there would be another crucible. A moment that would turn Agnarr away from that path and towards another. (You might spot it in the campaign journal when it happens.)

Now, here’s the thing: I didn’t know about any of this. Nobody at the table did except for Seth. In fact, I think it was literally years later that Seth revealed this.

But that doesn’t make the moment any less important. Or special.

In fact, quite the opposite: In the best campaigns, the fact that everyone is experiencing a personal narrative from their unique perspective of their character is what makes the totality of the shared narrative woven together from those threads so incredibly powerful and unlike the experience of any other medium.

If you’d like another example of this sort of “private narrative” being experienced by a single player — one so powerful that it literally moves people to tears — check out Matt Colville’s beautiful summary of the climax of the first season of Critical Role:

SESSION 20E: INTO PYTHONESS HOUSE

April 27th, 2008
The 9th Day of Kadal in the 790th Year of the Seyrunian Dynasty

Tee and Dominic returned to the Ghostly Minstrel only a few minutes before Ranthir emerged from his room. Ranthir introduced Erin to his friends and grabbed a bite to eat for himself.

Then they headed towards Oldtown and, within a quarter of an hour, they were standing on the street before Pythoness House.

The keep-like house seemed dreary beneath the noon sun – grimed and crumbling from years of neglect. Tee and Elestra were able to slip through the iron gate facing Emmitt Street, but Agnarr was forced to shove the rusty metal to one side causing it to emit a horrible shriek.

Looking through the stone arch on the front of the house they could see through a short passage into an interior courtyard. As the others came up the hill, Tee took the lead and headed through.

She was halfway through the stone passage when some instinct caused her to look up: A small, black metallic sphere was being dropped through a small murder hole!

Tee leaped forward as the powder bomb landed behind her and exploded. She managed to avoid the worst of the blast, rolling into the inner courtyard as several small mice scattered ahead of her. Agnarr and Tor, seeing the explosion, came running up – only to dash headlong into a second powder bomb.

Tee rolled to her feet and tried to find a target with her dragon pistol – but the opening was too small and the angle poor. She couldn’t see anything.

“COME TO ME…” The disembodied voice seemed to spring up from all around Tee – echoing through the courtyard and dancing through the empty windows and doors of the house. Tee whirled around, trying to find the source of it… but there was nothing there.

Elestra, Ranthir, and Dominic dashed through the passage into the courtyard. Agnarr and Tor pulled a rear guard, and barely managed to dive out of the way as a third bomb filled with dung was dropped.

Agnarr hauled himself to his feet, wiping a few flecks of disgusting excrement off of his armor. “That was digusting. Wait—listen!” His sharp ears had caught the sounds of skittering claws racing across stone – whoever or whatever had been using the murder hole was running off to the west through the upper passages. Then something large was thrown to the floor, and there was a booming noise – a large door being slammed.

Then there was silence. (more…)

Last week, in discussing how System Matters, I posited the hypothetical situation of an RPG which didn’t include a combat system. I did so because the entire concept is inherently radical: Including a D&D-derived combat system (even if it’s several generations distant from its progenitor) is such an essentially universal element of RPG design that the idea of a game which doesn’t do that is perceived as bizarre. I’ve actually seen people get angry at the suggestion, so ingrained has the expectation become.

So let’s talk about an RPG that actually does this: Technoir.

HOW TECHNOIR WORKS

We need to start by explaining how Technoir works:

First, characters are given numerical ratings in nine Verbs: Coax, Detect, Fight, Hack, Move, Operate, Prowl, Shoot, and Treat. These are basically the equivalent of ability scores seen in most roleplaying games.

The core mechanic of Technoir is to use a Verb to push an Adjective onto a target. The Adjective is literally any adjective: A word describing a quality possessed by a noun (i.e., the person or thing that you targeted). By pushing an Adjective onto a target, you are changing the description of that thing. In doing so, of course, you are fundamentally affecting the fictional reality of that thing: A computer system which has been hacked or bot-slaved is fundamentally different from that has been secured.

The mechanics of the game elaborate on this central conceit by (a) varying the severity of the Adjective you can apply (from fleeting to sticky to locked), (b) giving you mechanical structures for using these Adjective (so that they’re not just nebulous descriptions of fictional state, but also mechanically meaningful constructs), and (c) providing mechanical methods for removing negative Adjectives.

The cool thing about these mechanics is that they’re not dependent on the Adjectives you’re pushing: The mechanical structures work whether you’re pushing bullet-riddled or sympathetic or terrified.

STAKES IN TECHNOIR

In most RPGs, the combat system defaults to “I want him dead” as the stakes for the scene. In fact, this is often the only time traditional RPGs mechanically declare whether or not the stakes of a scene are achieved. (Most RPGs provide lots of mechanics for determining whether a specific method chosen by a character has been successful, but ultimately leave the question of whether or not that allows the character to achieve their actual goal to the GM fiat. See The Art of Rulings for a lengthier discussion of this.)

Technoir goes in almost exactly the opposite direction: You can’t push “unconscious” or “dead” or anything else that could remove the target’s agency in one hit unless your target is a Henchman (i.e., a mook). You can trivially set virtually anything else as the stakes for a roll, but not death.

This is a mechanical implementation of one of the pillars that Technoir identifies in the hard-boiled stories it’s seeking to emulate:

COME AT THEM SIDEWAYS

The conspiracy is bigger than her. The forces she’s dealing with could crush her like an insect. She can’t just kill all her enemies, she’ll have to outwit them. As such, this game doesn’t have a combat system designed for defeating opponents. It has a social manipulation system that gives the player the means to influence characters and nudge the narrative direction in which violent methods may be employed.

One of the semi-invisible consequences of combat systems and death being the only place where traditional RPGs mechanically declare a scene’s outcome is that a lot of GMs and players have become heavily conditioned over the years to think of death as the default ending for any conflict. And this is really weird when you think about it, because it’s not the way it generally works in other storytelling mediums (and it’s definitely not the way it works in real life).

Technoir breaks that conditioning. This means that the GM and players alike need to think about what they really want and how they can achieve that. The concept of vectors is really important in Technoir: If you want to achieve X by doing Y, then there needs to be a clear path by which Y can achieve X. You need to line up your shot. The fact that vectors aren’t always a straight line is, of course, the whole point. Sometimes the PCs have to push one adjective onto an NPC before they can push the one that they want (you need to get them trusting you before you can seduce them into being vulnerable). Or you’ll need to make it stick before you can lock it down.

The same is true for NPCs: If you’re the GM, make sure you know what they actually want out of the scene. Once they’ve got it, they aren’t going to stick around. Have them push the escaped adjective onto themselves, spend the Push dice they need to lock it against the players, and skedaddle.

DEATH IN TECHNOIR

All of this is not to say that character death is taken completely off the table in Technoir. What good noir story doesn’t feature death, after all? What if what an PC wants is to kill a PC? For example, let’s say the GM wants to frame a scene where a sniper takes a shot at a PC because they’ve been asking the wrong sort of questions.

The assassin’s bullet is going to push mortally wounded as an adjective onto the PC and the GM is going to lock that fucker in place by spending a couple of Push dice. And the important part is that once they’ve taken that shot — once they’ve pushed that adjective — the scene probably ends.

At this point the rules for “Lethal Consequences” are going to kick in: “At the end of any scene in which one or more adjectives were asserted that describe physical harm against a character, there is a chance that adjective might lead to the character’s death.” The specific mechanics for working that out aren’t particularly important for this discussion (but you can find them on page 138 of the rulebook). The point is that the PC has to check to see if he’s dying from this adjective or not. Even if he’s not dying from the wound, of course, he’s still going to be permanently debilitated by his injury until somebody pays to chrome him up.

And when the assassin finds out his target somehow survived the shot, he’s going to be coming back to try again. What are the PCs going to do about that?

That final bit is the really key element to grokking Technoir: The question of, “Do you live or die?” is not intrinsically interesting. You might die, but the game is far more interested in making you own your adjectives and then asking, “And what do you do about that? How do you fix it? How do you live with it?” Mortal injuries from sniper rifles aren’t inherently more significant in that equation that having your heart metaphorically ripped out.

As far as NPC death is concerned, I’ve already mentioned that Henchmen can be killed in a single shot (by spending Push dice to make dead lock onto them). The rulebook is a little less clear about what it takes to knock out Heavies and Connections (the other categories of NPCs), but what it boils down to is vector: If the PCs can put themselves in a position where they can justify getting the bullet to stick, the GM should let it stick. (That might mean beating them bloody. It might mean compromising their security. It might mean breaking their leg and making them helpless before you cap a bullet into their skull. It’ll depend on circumstances.) Note, too, that the “Lethal Consequences” rules don’t specify player characters; in fact, it says the opposite. The GM can (and should) use those rules for NPCs, too: You shot the Bride in the head and left her dead, but she woke up.

What are you going to do about that?

George R.R. Martin has a reputation for indiscriminately killing his main characters.

The popular conception is that he has this reputation because deaths are random and capricious. But this is not he case. His reputation for unexpected lethality is actually because he uses character death as a way of introducing radical reversals into his plot.

(For an in-depth discussion of reversals, check out 3-Point Plotting by John Rogers, the creator of Leverage. I also discuss it over here, but the short version: It’s the moment when the story you thought you were experiencing becomes a completely different story. It’s when Die Hard stops being a film about a guy trying to fix his marriage problems and becomes a hostage situation; it’s also the point where it’s revealed that Die Hard is a heist and not a hostage-taking.)

Martin has proven to be a master of making you think that he’s telling Story A only to dramatically reveal that Story A is never, ever going to happen. “Ah, this is the story of how Ned Stark saves the kingdom from the forces of corruption and– Oh. I guess not. So it must be the story of how Ned Stark’s son avenges his– Oh. I guess not. Well, then it must be– Ah, never mind.”

What makes this work particularly well is that these moments of reversal — rather than just being random — are completely logical and fully justified. It’s not just the author screaming, “GOTCHA!” It’s brilliant writing.

But the motherfucka is still gonna kill you off.

As another example, consider Tyrion’s trial. Once you realize that the trial is rigged, Tyrion seems doomed. Lots of stuff piles up to make that a virtual certainty (even after Tyrion demands trial by combat)… and then there’s the game changer when Oberyn agrees to be his champion. The narrative impetus appears to be that Oberyn will change the outcome, defeat the Mountain, and Tyrion will walk free. There’s a whole potential plot flowing out of that which makes a lot of sense (House Martell allying with Tyrion, Tyrion vs. Cersei in cutthroat politics, etc.).

And then just as that locomotive is starting to build up some momentum… Martin yanks it away again.

There’s a couple of reasons to do this:

First, by this point in the narrative we’ve all started recognizing the reversals. If Martin doesn’t throw some fake reversals / double reversals into the mix, the reversals themselves will become purely predictable. If he wants to keep you off-balance (which he does, because an audience getting too far ahead of the plot is the absolute death of narrative ninety-nine times out of a hundred) he has to mix it up.

Second, this process has a really significant impact on a lot of different characters and a lot of relationships between characters. And it does have a lasting impact on how events play out.

So there’s both a procedural and a dramatic reason for things to play out like this.