Friday 4 September 2020

City of Spires part 1: Theory vs. Practise

16. John Soane's Rotunda of the Bank of England in ruins, Joseph... |  Download Scientific Diagram

There's a city over the mountains. The people call it the City of Spires.

In the days of your great-grandparents it was rich and proud and prosperous. Traders thronged its streets. Coloured lamps flared in every window. Gold poured through the silk-gloved hands of its laughing lords.

In the days of your grandparents the caravans from the west stopped coming. The roads were all closed, and could not be reopened. The city's markets fell silent. 

In the days of your parents the spires started to crumble. The wealthy fled. The poor whispered of strange sights in the twilight. People said that the city was under a curse.

In the days of your youth, war came to the city. Faction rose against faction, house against house. The palaces burned. At last a great iron serpent tore its way out of the earth and destroyed all before it until only its masters remained, preening self-crowned kings of a city of ruins.

Yesterday the local lord rode into your village with a retinue of mounted soldiers. He said that your hovels were equally offensive to the eyes and to the nostrils, and had furthermore produced no tax revenue worth collecting for the last nine years, so he was going to knock the whole place down and turn it into a game reserve instead.

Where shall we go?' the people asked him. And he shrugged and gestured down the road that leads to the City of Spires.


As I mentioned in my last post, for the last year I've been running a (modified) B/X campaign called 'City of Spires'. We're currently thirty-something sessions in, with something like a hundred hours of total actual play behind us. It is not yet a campaign on the same scale as my previous 'Team Tsathogga' game, which ran for seventy-odd sessions over the course of three years: but it still represents a pretty considerable amount of gaming, and I'm very happy with the way it's run so far.

'City of Spires' was my attempt to put my ATWC material to use in an actual game, and as such it's prompted me to think about the differences between writing setting material and actually using it. When I started this blog, back in 2015, I was between gaming groups, and I wasn't really writing for anyone except myself: in fact, if I'm honest with myself, the reason I wrote about a campaign setting was because I wasn't getting the chance to run one. As a result, I wrote my early ATWC material without ever having to confront the key question: 'nice idea, but how exactly can I use it in this week's game?'  This isn't any kind of repudiation - I don't think any of the stuff I've written for the setting over the years is unusable - but it is an explanation for why what I initially wrote and what I ended up running ultimately turned out to be two quite different things.

When the 'Team Tsathogga' campaign finally ended due to some of the players moving away, the remaining players and I agreed that we should start a new campaign at level 0, in some other region of the same campaign world. I saw this as a chance to finally use my ATWC material, but I immediately faced several problems:

  1. Some of my players had read the blog, meaning that they'd know all about the setting (including lots of things they really shouldn't know) right from the start.
  2. ATWC assumes an animist cosmology with little or no standard D&D magic. But the world of Team Tsathogga, which the new campaign was going to be set in, was already established to be a science fantasy setting with standard D&D magic all over the place. 
  3. ATWC is a fairly high-concept setting, and requires a high level of player buy-in. But I knew that some of the players would be new to gaming, and didn't want to shove them into the deep end any more than I had to. 
  4. It's just too fucking big.

That last one was the real reality check moment. I'd spent years sketching out the ATWC setting, and the assumed campaign arc that went with it - go to the city, bounce around interacting with its various factions, make occasional trips into the outside world in source of resources or allies, and ultimately stage a coup or revolution, storm the King's Tower, and confront whatever lurked in the throne room of the Wicked King. What I hadn't confronted was the sheer logistics of it all. I'd given the city something like forty different factions: just meeting everyone could easily be a year's play. Its government was a massive, deeply entrenched tyranny, and while I'd deliberately designed it to be riddled with weaknesses, assembling a revolution capable of overthrowing it would still require immense amounts of investigation, organisation and diplomacy. Vast, complicated social worlds are great fun to write about, but can easily be a nightmare to GM. How many sessions would it take to play out a 'proper' ATWC campaign, with the PCs starting as complete outsiders and ending up as revolutionary masterminds capable of bringing down the Wicked King? Fifty? A hundred? More?

Ralamb's Ottoman Costumes Book (1658) | The Janissary-Archives
'Sorry - which faction are you guys from, again?'

So when I came to write City of Spires, I essentially did so by running a 'Condensation in Action' job on my own setting. My key principle was: 'like ATWC, but smaller'. 

The family resemblance is still very strong. It's still set in a ruinous city on a fantasy Silk Road somewhere beyond the eastern borders of fantasy Persia. There's still a tyranny, and the PCs are still trying to overthrow it, and the focus of the game is still on gathering networks of allies from among the city's strange and desperate inhabitants. But the city is physically smaller, and the population is much lower, and the government is just gangsterism writ large rather than the totalitarian nightmare of the Wicked City. Everything's a lot more ruined, in line with the principles I laid out here that the more thoroughly wrecked a setting is, the easier it is for PCs to exert real agency within it. There are fewer factions, and they're smaller, and most of them only have 1-3 significant NPCs each. As a result, after 'just' thirty-odd sessions, the PCs have been able to explore most of the city, meet most of the factions, and assemble plans and forces for an upcoming coup that just might actually work. 

I've reused many parts of the original ATWC setting, albeit in modified forms. Versions of the Cobweb, the Blue Necropolis, the clockworkers, the snake men, the merchant houses, the street gangs, the Rubble, and the Streets have all appeared in City of Spires. But they all shrank. The Cobweb shrank to a single tower. The clockwork armies shrank to a single huge automaton. The merchant houses shrank to a single merchant and his household. The Hortlaks of the Blue Necropolis shrank to a single undead princess. And so on. Over and over again, I looked at something that could potentially take a whole campaign to do justice to, and decided instead to condense it into something that could be dealt with in a single session. 

I should be clear, here, that I'm not saying that every campaign needs to do this, or that sprawl and vastness is always a problem for games. In a pre-plotted 'adventure path' campaign, it doesn't matter how big or complicated the setting is, because the plot will always ensure that the PCs interact with exactly the right parts of it to bring their story to a satisfying and level-appropriate conclusion. At the other extreme, in a totally wide-open sandbox, it doesn't matter that the PCs may only have scratched the surface of the world around them by the campaign's end: the story is just whatever happened to happen, and that's OK. But if you actually want your PCs to gain a decent level of knowledge of and mastery over their sandbox environment over the course of the game, then it's important to tailor its scale to the likely length of the campaign, and to be realistic about how quickly complexity starts to snowball as the number of interconnected factions grows. If I was starting this one over again, I'd put in less rather than more, at least to begin with. (Did the city really need six noble houses?) It's much easier to add things into a sandbox than it is to take things out.

Next post: the city as hexcrawl!


  1. I like the idea of the newer, pared-down version of the Wicked City being built in the ruins of the original, expansive one - the Cobwebs are standing but abandoned, the Murder Harlots have no one left to appreciate their art, and the Wicked King's inscrutable nature is just because it's been so long since people even pretended he was still in power. That way any idea can easily come back as a holdout or resurrected concept from the city's "glory" days (if you could call them glorious), but things are still small and tenuous enough for players to easily affect them.

    1. 'Small and tenuous enough for players to easily affect them' is the key, yes. The bigger things are, the more a campaign will need either heavy scripting or excessive length before the PCs are able to effect credible change. Whereas when everything's small and lying in ruins it's much easier for a small group of dynamic individuals like the PCs to bring about rapid and lasting shifts in the status quo.

      Regarding the Wicked City itself, I kinda think of it as already in a deeply unstable equilibrium. The current regime's only been in power for seventy years, and the whole polity is already a total wreck, so I doubt it has much of a long-term future in its current form. I can easily imagine it declining and declining until it's little more than a haunted ruin in the desert, inhabited only by ghosts and madmen and a mask-wearing murder cult descended from the original secret police. Might be a good setting for an adventure, actually...

  2. It takes a lot of discipline to point a gimlet eye at your favorite stuff and cut out everything that doesn't meet the needs at hand. Sounds hard.

    But this also sounds like a cool exercise in reinterpretation. You had this toolkit of stuff that you could pick and choose from and remix as needed to generate a given effect. (I mean, you have to do when running even a vanilla game - including everything in the Monster Manual would produce a weird and pretty disjointed setting. But you had a toolkit you'd made yourself.) Did that make the process different? (Or will this be a future post...)

    The idea of a Wicked City sort of endlessly refracted through different worlds is striking. Probably not game-able, but enticing in that old "Dancers at the End of Time" feel.

    1. I'm not sure it was that different, to be honest. It's the same way I approach everything. I never run a setting or module exactly as written: I always just use them as raw materials, available to be reworked into something that fits the needs of the actual game at hand. Same with other people's blog posts: I've used a number of ideas from Chris Tamm and Arnold K's blogs over the years, but always in adapted forms. So doing the same with my own work felt pretty natural.

      In RPGs, I think there's always a gap between theory and practise, what we write and how we play. I don't think my work is any exception to that!

    2. Fair enough! Thanks for the answer.

      That gap is a little weird, isn't it? A whole lot can happen in there. For whatever reason the gap feels more opaque or less obvious with blog-borne content (maybe because they're less formal and stable than a book?). Maybe that's just my misconception.

    3. No, I don't think it's a misconception. Blog content is much closer to the kind of notes we might make before or after actual play than most published content. No-one's actual-play adventures look like a published module. No-one's actual-play setting notes look like a published campaign setting. But blog content is scrappy and fragmentary, and thus much closer to things that actually get used. 'Yeah', one might think, 'I can use this guy. I can use this room. I can use this monster. I'll put them into tomorrow's game, between the bit with the hydra and the bit with all the zombies.'

      But, honestly, you'll still probably change them a bit along the way.

  3. I am currently setting up a second campaign in my own setting and making the same kind of experience. The previous campaign was only nominally set in my setting, being really about the journey to the Isle of Dread and using the cultures from my setting for NPCs.
    Now that I'm setting up another campaign right in the heart of the world, the process of making things playable mostly consists of making them smaller and turning them into digestible chunks that can be dealt with in one or two sittings, instead of each of them having enough scope for a entire campaign.

  4. Interesting post. Lots of good ideas. Especially with what is actually practically gameable vs what you’ve written and/or collected. I’ve long been interested in running a good city game but I think I’ve collected more than is suitable for a realistic game length. Your original AtWC posts augmented a lot of ideas I’d had, and spawned others such that they’re part of the background of my ItO “Bastionland” game. Before I go too far I think a gentle cull may be in order. As you say it is easier to add things (back) in than to remove them.