On Knave and Old-School Rulebooks

Knave by Ben Milton was my first light-weight old-school D&D sort of game that I ran for my friends. I suggested using it because although some of them were familiar with Fifth Edition, many of them just weren’t. I could tell that they would be overwhelmed with the rules of character creation and of the game itself. I also wanted to run our little campaign more like a party game where they could just get onto the Zoom call, generate their characters, and start playing.

This casual approach lends itself more to an “old school” style of game, which at that time I had been interested in playing for a while. For me, Fifth Edition didn’t just have baggage with respect to its huge amount of rules, but also with how it was expected to be played. The play culture expects you to create a whole character with a background and a motivation, and then the dungeon master creates a story with which to challenge the characters enough to show their respective strengths and advance their character arcs. I just like tabletop games as get-togethers, so it helps for the game to be more of a game proper. Old-school play, with its clear motivators and lack of narrative-domineering on the parts of both referee and player(s), appealed to me this way.

So I suggested Knave, and we ran with it. I added some modifications to account for class-based and racial abilities because some of my friends wanted the usual D&D trappings, and there was no reason not to include them. Besides that, I thought it was basically Knave as written, and the modifications didn’t often come into play anyway. Our experience was that, provided you have an idea of what “a D&D'' is supposed to be like, Knave offers what you need to facilitate that and then simplifies it. Most of my friends ended up creating their characters ahead of time just so we could work out any kinks, but it was a painless process.

The Campaign

We played towards the beginning of quarantine when Tiger King and Animal Crossing were both in vogue. The conceit of the game, then, was that the player-characters had just moved to a new continent and had some debt to pay off to their housing contractor. They decided to steal some exotic animals from the landlocked boat of an apocalyptic street preacher and ex-adventurer [1], and then deliver them to a sketchy zoo in the next city over. The crawl over to the big boat was a little bit boring because I didn't realize I should have keyed all the hexes in the area, but it was no big deal.

On the way to a village near the boat, they came across a big airship selling luxury goods like spell books and armor, and that gave them another reason to get their hands on some money. They were also almost stampeded by a bunch of slides herded by cowboys, but one of them charmed a slime before it was about to run her over. She kept it as a pet! [2] This was my first time really messing with random encounters, so I learned quickly which ones worked (i.e. kept my friends' interest) and which ones didn't. Really, though, the best ones just gave my friends opportunities to explore and try things out.

Once at the village, they met the ex-wife of the preacher and managed to glean from her where their old farmhouse was. I had to improvise the next morning where they went out a couple hours away to investigate the grounds. They learned from old family photos about the preacher’s three sons who still believe in his mission (and could thus be expected on the boat), and they also discovered a model of the ship in the basement. They took notes based on what they could see, and started planning on which animals they should extract from the boat and in which order.

The next day, they came to the boat and carried out their heist plan. They enjoyed the silly animals I made up; the gnome's racial ability to talk to animals (not in Knave as written) came in handy when negotiating with ligers not to eat the winged buffalo in exchange for their mutual freedom from the boat. Complications arose when one part of the team tried entering the boat from the main entrance to distract whoever was there, and they ended up punching one of the preacher's sons hard enough for him to faint at the front door. We loved the mess, though! The more complications, the more fun.

What I Thought We Played

I feel like I got a little carried away describing what happened, so I'll stop myself here. Overall, I was impressed with how Knave felt like training wheels to play (and run) an old-school D&D sort of thing. There didn't seem to be any pressure on the players' part to have to refer to their character sheets, except when it came to inventory management. The hard limit based on the characters' constitution scores felt somewhat punishing, so partway through I told my friends that they could buy a backpack to increase the number of inventory slots they had. Besides that, though, they seemed to have had a seamless experience figuring out their characters' actions.

Whenever it did come time to roll dice, Knave has a certain elegance to character scores that I haven't seen in other rulebooks. First, players roll for their characters' ability bonuses (+1 to +6). Then, they determine scores called ability defenses equal to 10 plus the respective ability bonus (11 to 16), which serve as target numbers for rolls made against the character. This rule on paper makes it easy for both the player or the referee to make rolls, since Character A adding their bonus to a roll against Character B's defense is statistically equivalent to Character B adding their bonus to a roll against Character A's defense. It also makes the rolls really transparent for the players, since it's easy to glean what they are rolling against and why.

There is, however, a slight point of... not confusion, but a "huh" moment. The rules say that you should not roll above or equal to the defense of the roll (i.e. the target number), but only above the defense. This is what guarantees that characters with equal bonuses, i.e. equal abilities, each have a 50% chance of success when contending with one another. However, the effect is that when rolling d20 > 11 with a +1 bonus (d20 + 1 ≥ 12), you're ultimately still rolling d20 ≥ 11. The combination of not succeeding when you roll equal, but having that also being canceled out by the math, feels slightly unwieldly. It's impossible not to have funky math somewhere in the mix, though, so I'd rather appreciate that Knave handles things elegantly with respect to math than focus on what's almost a suspension of disbelief in the dice.

But, Wait a Second...

While revisiting Knave as a text for this post, I realized it didn’t have everything I thought it had. When playing with my friends, I was referring to the fancypants version published by Joel Priddy [3]. Besides having a cleaner layout, Priddy also supplies some additional rules about dungeon and outdoors exploration that weren't in the original Knave. I say that I had only "referred" to the fancypants version because, in fact, my friends hadn't read any version of Knave in-name at all. Instead, I made a couple Google documents that explained character creation and the rules of the game we were going to play step-by-step. We played Knave in terms of dice-rolling and character creation, but the meat of the game wasn't supplied by Knave at all. I came up with very generic rules for timekeeping and random encounters that, of course, were mostly read by me and not the players (I'd think, anyway).

This is a tendency I see in a couple later D&D-likes, to mostly offer rules for combat but not anything substantial about what's supposed to be the point of the game (eXpLoRe DuNgEoNs, aCqUiRe GoLd). One edition of Into the Odd that I have also does this to an extent, except it at least gives an example of play where it explicates that the referee should make a luck roll for random monsters. The first edition of The Black Hack says that time is kept in either seconds or minutes depending on what's called for, and it has the referee roll every 15 real-life minutes for random encounters; these things, however, do not really communicate the rhythm of the game. Maze Rats, an earlier rulebook by Ben Milton, says that the referee should roll for random monsters every 10 in-game minutes inside a dungeon; it explains the function of these encounters, but it does not emphasize this check and the associated timekeeping as a structural feature of the game.

So it's kind of whatever, but it poses what I think is an interesting problem: why do these old-school rulebooks, or rather the ones from the mid-to-late 2010s, not clearly communicate the structure of the game if they even hint at one? There's lots of talk about focusing on player skill rather than character skill, a still-popular OSR slogan, but they don't really situate the players in the context where they will apply those supposed skills. I would think that a rulebook for this playstyle would do better to focus on the structure of the game more than anything else, if the point isn't supposed to be the hard-coded rules. Knave is my favorite of this period of old-school-inspired rulebooks for how it simplifies the annoying parts of the game tradition it comes from. However, like the other rulebooks, it doesn't really tell you how to play.

The Future of Knave

Ben Milton put up a first draft for the second edition of Knave on his Patreon [4]. It would be unfair to actually review this draft as a complete and published rulebook, but I wanted to offer some thoughts on it based on what I liked and didn't like about the first edition of Knave. Like the first edition, the first draft of the second edition is a classless D&D sort of thing. However, it's not as rules-light. This is likely due to some people wishing that Knave had something like a class system to inform characters, or due to Knave not really having exploration rules, or due to trends in the old-school game-enjoyer scene to produce more baroque and detailed rulebooks.

Knave 2E Draft 1 no longer has the bonus/defense structure for ability scores as did Knave 1E. This sort of disappoints me because what Knave 1E did made me believe in universal resolution systems, even for just a moment, because of its simplicity. However, character creation has been massively simplified in a way that I myself dig. You have the six usual abilities, each with a bonus that starts out at +0. Then you roll a die to figure out which ability score indexed 1-6 you add a +1 to. Finally, you sort of do this again twice, except you choose which other two abilities get +1 instead of rolling [5]. This is elegant in its own way which I appreciate. Since bonuses start at +0 instead of +1, it allows Milton to also resolve the weird roll-above-but-not-equal issue that Knave 1E had. This isn’t without its own kinks, though. Now to make a roll against another character or monster or whatever, the target number is equal to 11 plus the character's ability bonus. So, the same dice-math is there; it's just it's a little more awkward on the math-side instead of the dice-side. I'm picky.

Knave 2E Draft 1 also has rules for exploration! It uses a variation of the overloaded die [6] for both overworld and underworld exploration, which encapsulates random encounter checks, resource consumption (e.g. torches or rations), and straight-up timekeeping into one function. The main benefit of the overloaded die, for people who use it, is that it wraps up the core of the dungeon-crawly game into one little d6 table. There's no more keeping track of torches or when the party should rest; just roll the die after each turn and read what happens. I suspect that the introduction of this hack into every rulebook as of late is because it encapsulates both the base flow of the game and the play-culture supposed by the author. So, I would probably be in the minority of players and referees in saying that I don't really care for it. This is mostly because I don't like referring to the rulebook while I run a game (which having a table basically requires you to do), but I would be lying if I said it weren't also because it makes things slightly too easy. I know that's, like, crazy to say, but in short I think it obscures the rhythm of the game in a way that I don't think saying "Your torch lasts an hour, but you check for monsters every 10 minutes" does.

Regardless, the inclusion of an overloaded exploration die in Knave 2E Draft 1 reflects this rulebook's transformation of Knave from a generic reduction of a D&D thing (which I think is the selling point of 1E) to a rulebook composed by Ben Milton as the author of your game experience. We see this also in the draft's new rules for hit points: hit points are now simply the amount of damage your character can endure before they begin to accumulate "injury" slots in their inventory space. A character then dies when all their inventory slots are filled with injuries. I can't speak for how this works in practice, though I think it'd be a funny idea for a character to undress as they're hurt in combat--this might be a fun basis for strip D&D. The more important thing, to me, is that Knave 2E Draft 1 has a lot more for you to learn before you start playing, by virtue of it presenting more unique and complex rules rather than rules that should only be seen and not heard (if one can help it). This is not a problem with the rulebook, which is very clear in its presentation and sounds fun, but it implies a different aim than the first edition of Knave had.

This is indicative of the larger trend I see in rulebooks as of late to represent more auteur takes on the same D&D game, instead of finding new ways to minimize the game itself. It's as though you were looking at a particular referee's head and learning how they specifically run games. In other words, what motivates this coming generation of rulebooks seems to be what motivated Gygax to write Advanced Dungeons & Dragons (besides wanting to screw over Arneson): to present a cohesive, standard vision of a dungeon game through the lens of one author. Having a singular, authorial vision is an easier sell than having a further-simplified version of a generic vision, and I suspect that's why we're seeing more of them now.

Wrapping Up

The first edition of Knave really helped me find my footing as a new referee just wanting to run a game for my friends. Its rules were mostly out of the way, and when they did get in the way by necessity, they were really easy to grasp and play with. Still, the important part of the game (exploration) was mostly left unstated. Though I didn't realize it at the time, it was something that I had to import based on my own ideas of how it should be done. This is an issue which I think permeates old-school rulebooks of the time, and which I think later rulebooks like the first draft of Knave 2E tries to solve. However, like other contemporary books, it accomplishes this by presenting a highly-specific version of the gameplay it anticipates, which grates away at its potential to be a generic simplification of the dungeon-crawly game at large.

Knave 1E, despite the essential things it lacks, will always be my favorite reduction of the boilerplate rules for D&D. Just be sure, as with any rulebook from that time, to bring your own game!

[1] Another suggested mission was to deliver new furniture to a luper colony (i.e. of werewolves) in the woods. I thought it was clever.

[2] I forgot to keep track of how long the Charm spell was supposed to last, but also I have no problem with how this turned out.

[3] https://abominablefancy.blogspot.com/2018/10/knaves-fancypants.html

[4] https://www.patreon.com/questingbeast

[5]  I would rather just roll three dice from the beginning and add +1 to each of the ability bonuses indicated by the dice. In fact, I bet this was Milton's original plan, but maybe he didn't want player characters to start with one ability bonus of +3.

[6] Originated by Necropraxis at https://www.necropraxis.com/2014/02/03/overloading-the-encounter-die/