The Mysterious Appeal of Emergent Narrative

26 Feb, 2025

A few weeks ago I played in a session of Falling Tide, my brother Nick Hendriks's OSR game.¹ My PC Makepeace Barebones and the rest of the party were in a pirate town called Mosquito Bay, a wretched hive of scum and villainy if there ever was one, and there we witnessed a duel in the streets. We decided to let things play out, though two of us made bets with strangers in the crowd. (Alas, we lost.) It was a fun scene, though my summary of it might seem unremarkable. I bring this episode up because it is an example of something I find very compelling but have a hard time describing. It occurs in nearly all media, but tabletop role-playing games can deliver it much more consistently. I've tried for years to put into writing what I'm looking for, but have not succeeded. What I'm starting to settle on is that emergent narrative can produce whatever nameless thing I'm looking for. It's not the only thing that can produce it, but it's a place to start.

Early sketch of Mosquito Bay, by Nick Hendriks, source, used with permission and cropped.

Nick discusses how his game works in a few posts on his blog, most notably for this discussion in "I pre-roll my random encounters," in which he writes the following:

That's right, I pre-roll them, and I won't apologize for it. It gives me more time to think about how to make the results cool.

[...]

It's important to me to stick to the results of the dice and not interfere if at all possible.¹ But that doesn't prevent me from interpreting the random encounter results in interesting ways.

I don't typically use monster reaction tables when rolling random results, but instead give myself permission to decide what it makes sense for a creature or NPC to be doing in this region. When I created the random encounter tables for each region I selected each creature and NPC for a reason. I ought to be able to come up with plausible circumstances for them that don't stretch my players' sense of disbelief, while also taking the opportunity to make encounters more interesting than "the creature is sitting in a clearing waiting for player characters to arrive."

It's worth your time to read the rest of the post; in particular that footnote explains why he does things this way and what effect he's trying to create.

As a player in many of these sessions, I have to say I find this method effective. Often the encounters that Nick's method generates feel narratively appropriate; of course, that is partly because he prepared his encounter tables in such a way that the results are somewhat likely to be appropriate. It still might not happen that the creatures or events we encounter pertain to our motives and interests, but the odds they will aren't bad. When we headed into ape territory seeking an ape wearing a crown we once saw, the encounter tables for that area were of course heavier than usual with apes. And when my character, Makepeace Barebones, warned another PC that we should beware terror birds, only for a terrifying creature that was probably a terror bird to crash into our camp the following night, that also wasn't such a coincidence: Makepeace had seen terror birds nearby, which is why I mentioned it, and the encounter tables reflected that.

Of course another reason these encounters feel appropriate is that the tables' contents are broad enough that Nick can adapt them. Consider the anecdote with which I began this post. That duel was between the leaders of two factions, both of whom we had met or heard about: the administrator of the local trading company and a corsair boss. At the end of it, one of them was dead in the street, and I have every reason to believe that event will change the Snarl's political landscape in sessions to come. Afterwards my brother told us that the encounter table for the town only determined that a duel would happen; he decided who the participants would be, and he chose ones that would shake up the setting however the duel ended. That is what I mean when I say that the encounter tables' members are somewhat broad and can be adapted to the story. (But of course it was the dice that decided who won the duel, and that matters.)

The thing I have a hard time explaining is why I enjoy this so much more than I think I would enjoy it if Nick had simply chosen appropriate encounters. It has something to do with how meaning is produced from these joint activities that no one actually intended; it's sense-making with many of the parts exposed. I don't know why I like this kind of emergent narrative so much, but know that I do. It doesn't just appear in tabletop games, either. Adam Millard of Architect of Games discusses it in his video about one of my favourite video games, "Why Wildermyth's Best Story Wasn't Written By Anyone." I also enjoy, in I think the same way, seeing how extrinsic or arbitrary constraints shape stories in poetry or prose: how, for instance, authors of epics in the early modern period were expected to include trips to the underworld, descriptions of the dawn, talking trees, and snake-ladies guarding thresholds, and they had to find some use for these conventions that would contribute to their overall vision.

Despite my interest in emergent narrative of this type, I use hardly any random tables in my own tabletop games, and until recently I would have told you I probably never would. Just about the only exceptions were the few times I ran variants of a game I made about tracking creatures through the woods. In The Retired Adventurer's taxonomy, if I understand it correctly, the culture of my main campaign's table is traditionalist with significant neo-traditional characteristics, and maybe some classical characteristics as well.² I carefully make each encounter to reveal information about the world and its inhabitants, to reinforce the atmosphere I want for each of my provinces, or to showcase something about the PCs, either by challenging them or by giving them a chance to show off new abilities. My shorter games are often even more tightly defined: I have a particular scenario in mind for my players to explore, and I have a clear sense about what each part looks like.

Of course, that's not to say that no surprises emerge from this set-up; rather, the unplanned narrative emerges from the places where my players do something that I failed to anticipate. If that happens in combat, novel situations develop as the parts of the systems interact with each other and with the players' unfolding choices; if that happens outside of combat, novel situations develop as I react to their choices according to what I already decided about the setting and NPCs. The emergent narrative in my main campaign comes not from random tables, but instead from collaboration and from the usual randomization of Dungeons & Dragons: ability checks and saving throws. Encounter tables are not the whole of emergent narrative.

And just as encounter tables are not the whole of emergent narrative, so too emergent narrative is not the whole of this mysterious thing that appeals to me. After all, if what I like about encounter tables and what I like about epic conventions are related, then emergent narrative of the kind I'm discussing isn't actually necessary to achieve the thing I'm looking for. It's not the whole thing: not necessary, and maybe not sufficient. After all, the sorts of stories ChatGPT and its ilk can produce are emergent narratives as well, but I find them of limited interest. I still don't understand quite what I'm seeking, although it maybe involves an author or reader making meaning by repurposing images or ideas that were not made for that meaning, or in other words something like narrative kitbashing.

Given how much I enjoy the design of Falling Tides, I'm starting to think I might want to change my style a little. I don't see more than minor adjustments being possible or helpful in my main campaign; we're over five years in and I worry it would be too disruptive to change much now about the table culture. Still, we've tried new formats for brief stints before; maybe I can find a reason to employ a small hexcrawl somewhere. They'll be headed to the Feywild soon, and eventually to the Astral Sea, so maybe I could fit a hexcrawl into one of those expansive new locations. Someday this campaign will finish, too, and I presume I'll GM another campaign after that; that next one could build in more encounter tables. I like the look of "The Hexcrawl is the Clock," Kilian's inaugural post for On Shadowed Stone, and I've started following with some interest Xaosseed's tinkering with the hexcrawl format at Seed of Worlds. See "Hex-stocking by transfer/bleed-over" for an example. I've got some time to think about how I want to weave in more opportunities for emergent narrative.

I still haven't been able to put my finger on exactly what I like about emergent narrative, what beyond the mere emergence of narrative makes it appealing. What I have figured out is that, while it's true that I'm interested in narrative first and foremost, there is something uniquely charming about the story that emerges from how the narrative elements are built on the mechanical elements, be they gamist, simulationist, or anything else.³

I apologize for an unusually introspective post. If you want something you can immediately walk away with, perhaps consider my first footnote below. I'm going to try to stick to those titling conventions from now on.

#for GMs #for players #narrative kitbashing #style and approach