Bathtub Reviews are an excuse for me to read modules a little more closely. I’m doing them to critique a wide range of modules from the perspective of my own table and to learn for my own module design. They’re stream of consciousness and unedited critiques. I’m writing them on my phone in the bath.

The end of January marks my 100th Bathtub Review (not counting my I Read Reviews, so the number may be closer to 150), so I’m celebrating with an Introductory Extravaganza, where I review a different Dungeons and Dragons starter module every week for a few weeks. Last week I covered the Village of Hommlet (AD&D), this week I’ll cover the Sunless Citadel (D&D 3e) and coming up are the Lost Mine of Phandelver (D&D 5e), The Iron Coral (Into the Odd) and Blancmange and Thistle (Troika!).

The Sunless Citadel is a 32 page module for D&D 3rd edition, written by Bruce Cordell. In it, the player characters investigate a dungeon whose upper levels are occupied by feuding kobold and goblin tribes, before driving deeper to find a dark evil: An evil tree that grows a single apple each summer and winter, the seeds of which sprout evil blights. It’s a 55 room dungeon that spends no time on the surrounding area at all; of the total page count only 6 are not dungeon key — 3 are introductory, and 3 are a stat block appendix. Compared to both the Village of Hommlet and the Lost Mine of Phandelver, it’s a shockingly self-contained module, more similar in scope to the Iron Coral.

I skipped the third and fourth editions of Dungeons and Dressing entirely, opting instead to play board games for a decade, and so I’m a little surprised to find that The Sunless Citadel feels very strongly transitional, and not so much of a break from the 1st and 2nd edition modules I’m more familiar with. I assumed that this would suffer from the bloat that I’ve heard was endemic in 3.5e, but the truth is, while it inherits bad habits from the Dragonlance era of 2nd edition, it’s actually got a strong authorial voice and reminds me a lot of the better dungeons of AD&D.

In terms of layout, maps occupy the inside back and front covers, and interior text is a dense justified two column layout that makes you rely heavily on headings for navigation. That said, main headings are replicated in headers, and are differentiated clearly by font, size, indentation and decoration. I intensely dislike the choice to lay sidebars over the external margin art — the external margin art crowds the rest as is, but this choice makes those columns affected scarcely legible. Aside from this, small things like leading, paragraph spacing and padding make this generally more legible than any 1st or 2nd edition module. It’s a cut above, but still beholden to its past. Interestingly, compared to the Village of Hommlet, the keys are more difficult to read: Beyond differentiating the boxed read aloud text and stat blocks, there is no text flagging or highlighting for points of interest or treasure. Given these keys tend to bloat as the stat blocks are heftier than previous editions and are included in-line (excellent choice for an introductory module), the difficulty identifying key information is a problem. You’ll have to go through this with a few different coloured highlighters on your read through to make it smoother to run. There are only 7 interior art pieces, not too different from what came before and not the incredible over the top art density that has come to characterise modern editions. These are, too, black and white line art of a style more typical of earlier editions, although the realism that came to characterise Dungeons and Dragons is beginning to creep in in place of the less polished, janky stylings of AD&D.

The introductory pages introduce jack all, to be honest. Even encounter levels — intended to show at a glance whether your party is likely to be destroyed by an encounter — refer to the DMG. Boxed text and sidebars are explained but little else. The faction summaries aren’t terrible, but don’t give you much juice, and make assumptions regarding what will happen: “The adventurers have less luck dealing with the goblins…” The 3 hooks are not interesting enough to warrant existing and don’t impact the modules’ play in any way; you could be generous and say that is because it’s for beginners, but it largely seems lazy. The rumours aren’t given in list form, and while you can gain them “by role playing”, they really don’t give useful information that will impact play or information that will really lure them to the Sunless Citadel itself; while there’s a social contract to go to the dungeon of course, why are these even here if they don’t provide either in-game justification or contribute to the gameplay. There’s at least one genuine mistake: While the tree bears only one fruit, somehow this is supposed to provide the goblins with some kind of industry selling magical fruit and spreading the seeds. This feels like a massive editing mistake, to me? Oakhurst as a village also feels like it exists in greater detail somewhere on the cutting room floor — a conjecture also supported by the existence of a genuinely lovely panoramic map for a town that isn’t described otherwise. In general, it assumes you remember the text far too much — I wondered why saplings were missing in Oakhurst as a mystery hook, when all it meant was that they had turned into blights, obvious in retrospect, but I missed it in the walls of text.

Overall the introductory section assumes a deep familiarity with both the Player’s Handbook and the Dungeon Masters Guide, and doesn’t do anything to tutorialise the referee at all. Does it tutorialise the players, though? Well, it does gradually expose the player characters to the kind of challenges that they should expect, and does so at first in low-risk situations. In the first few areas, there are sign of low risk creatures, there are language puzzles and search puzzles that foreshadow future encounters, and environmental challenges that require conscious decision making. All of these are mirrored deeper into the module, with traps and secret rooms. But, it gets pretty gnarly pretty quickly; I’m not sure if this is a positive or if the very regular hazards will be a negative experience for new players (there are 5 traps in the first 12 rooms). There is, at least, theme behind the traps: There’s an ongoing war between kobold and goblin that the player characters are stumbling into, but it feels like this dungeon is more punishing than the Ruined Moathouse in Village of Hommlet or especially compared to the Goblin Cave in Lost Mine of Phandelver. You’re less likely to have a rollicking good time here; especially as this adventure assumes you’re mapping as you go and that marching order is important. It’s a ten-foot-pole-poking adventure, moreso even than Hommlet was.

At least until you encounter the kobolds, who you’re supposed to ally with to defeat the goblins that block your way to the deeper druidic grove level. This sounds great in theory, but there’s no simple solution to building reputation with the kobolds, let alone with the goblins, and as factions there is very little to them aside from “squabbling over territory” which isn’t even “squabbling over resources”. There are named kobolds, but they don’t have desires or needs of their own to speak of; the only goblin named is their chief. There’s potential for faction play here, but it’s not set up adequately at all in my opinion, and would be more interesting again if we could incorporate our corrupt druid into the politics, but he’s confined to the grove below. This means that in theory this isn’t a hack and slash dungeon, but in practice it’s likely to be. It’ll take a lot of work to squeeze social play out of it, but it has good bones.

Overall, the Sunless Citadel isn’t something I’d choose to run as an experienced referee for experienced players. It, more so than the Village of Hommlet, assumes a lot of background knowledge of the game from everyone at the table. What’s more, the dungeon itself isn’t deeply compelling as it stands. However, it has a lot of components that could, if arranged differently, make for some compelling and interesting play. It’s inspiring, that’s for sure. I could see a version of this module that would make some hella interesting memories, but where are the compelling NPCs to meet? Where are the exploitable areas? How can we claim parts of the dungeons and make allies and wage war and play them off against each other?

It doesn’t tutorialise at all for the referee, but it arguably does for the players, at least in the sense that the level design provides a graduated difficulty curve and lethality so that they can learn to poke things with a long stick; but, given the issues with the design of the module, I think it will also discourage their best impulses.

That all said, I think, like the Village of Hommlet, it does reflect the expectations for play in 3rd edition: This is skill-roll focused, hazard and encounter-based play. It’s straying from hirelings; now you’re solo heroes. You’re levelling up quickly and facing high level foes by the depths of the dungeon. In this way, the Sunless Citadel is a fascinating insight into the modules that would drive play for over a decade, and potentially, still drive it in in 5th edition.

Idle Cartulary

Playful Void is a production of Idle Cartulary. If you liked this article, please consider liking, sharing, and subscribing to the Idle Digest Newsletter. If you want to support Idle Cartulary continuing to provide Bathtub Reviews, I Read Reviews, and Dungeon Regular, please consider a one-off donation or becoming a regular supporter of Idle Cartulary on Ko-fi.