Go to Part 1

I believe that the Kraken Society first appears in The Savage Frontier (1988) by Jennell Jaquays, where they are described as the Thieves’ Guild of the North and connected to the Kraken of the Purple Rocks. They came to a particular prominence with their appearances in two SSI AD&D computer games — Gateway to the Savage Frontier and Treasures of the Savage Frontier (which I haven’t played).

Much of the material from The Savage Frontier was incorporated and expanded into The North, a 1996 boxed set. This more or less forms the foundation of the Kraken Society, but it received its most expansive treatment in Cloak & Dagger (2000), one of the very last AD&D books ever written. (This book appears to draw on continuity developed by Elaine Cunningham in several tie-in novels, but I haven’t read those, either.)

The 3^rd Edition supplement Lords of Darkness (2001) also had a full write-up of the Kraken Society, but it was basically just an abbreviation of the material from Cloak & Dagger.

The Kraken Society didn’t appear in 4^th Edition, as far as I can tell.

Purple Rocks, but not the Kraken Society, next appear in the Sword Coast Adventurer’s Guide (2015).

Finally, I’ve drawn material from both Princes of the Apocalypse (2015) and Storm King’s Thunder (2016) to flesh out into the Society’s current activities in Yartar. Background details unrelated to the Kraken Society were drawn from Shawn Merwin’s “Backdrop: Moonshae Isles” article in Dungeon #196 (which, as far as I know, remains the only definitive treatment of Moonshae in the 15^th century).

My goal with this project was to advance the Kraken Society into the 15^th century and provide a definitive reference for it. This meant:

• Advancing the Society’s timeline. I didn’t want the organization to have simply been stuck in a form of stasis between the 1380’s and the 1490’s.
• Resolving the various continuity discrepancies between the official sources. Some of this could simply be swept under the rug (it happened a hundred years ago, so there’s no reason to worry about the details). The rest was primarily a matter of creating Slarkethrel’s myriad mystery cults and then diegetically seeding the contradictory continuity into the game world. There are many “truths” of Slarkethrel… and that’s just the way it likes it.
• Expanding and detailing the Society.

The final point took many forms. Too many to catalogue, frankly. (If you’re curious, however, you can compare the material here to that found in Cloak & Dagger and the 5E adventure books to easily identify it.) But, for example:

• Ascarlian satrapies did not previously exist.
• The Assassins of the Purple Veil are new.
• Society operations in Neverwinter, Thornhold, Caer Westphal, and Luskan had not been previously detailed.
• Concepts like low tide, high tide, and bucklings — along with other cultural aspects and slang of the tentacles of the deep — are original creations.

And so forth.

I’ve done textual histories of the Forgotten Realms in the past (Zariel, Elturel, Trade Way, Arveiturace) which only seek to provide a gestalt summary of existing Realms lore. But the primary goal of this piece was to creatively expand the Kraken Society (and Slarkethrel’s domain in general).

Go to Storm King’s Remix

Roleplaying games can vary a lot in how lethal they are for player characters. And, perhaps even more importantly, they can vary a lot in how permanent death is when it occurs.

This can even be true within a single system. In some classic versions of D&D, for example, PCs can begin in an ultra-fragile state (in which any hit in combat could automatically kill them) and then level up to a point where death is just a minor inconvenience.

But let’s assume that we’re playing a game where death really is the final frontier: If your character dies, then they’re dead. No take-backs. If you’re playing in a campaign with a game like that, how can you handle death?

One way is to duck the question entirely with script immunity. In campaigns with script immunity, PCs simply can’t die. This may be a feature of the rules (as in Magical Kitties Save the Day), but is often a metagame conceit openly or silently respected by the table. There are a number of techniques that are useful for making script immunity work, but the two most common approaches are to either disallow the mechanical outcome of death (NPCs will always miss the final blow) or to interpret the mechanical outcome of death to mean something other than death (e.g., the PC is instead knocked unconscious). These are often combined with a caveat which allows the PC to die if their player wants it to happen. (This is because script immunity is usually a technique favored by dramatists, and being able to have death occur only when it is dramatically appropriate and satisfying is desirable. But I digress.)

What I actually want to focus on is not how to avoid PC death, but rather what comes next.

The first option is to actually remove the player from the campaign. The PC was their agent in the game world. Now that their agent has been destroyed, they have no ability to participate in the game.

This is, in some ways, the opposite extreme from script immunity — where immunity completely removes lethal consequences from the game, the all-or-nothing approach makes those consequences about as meaningful as they can be. On the other hand, script immunity and one-PC-only both recognize how momentous and important death can be to a narrative and simply emphasize it in radically different ways.

Personally, I’ve never seen one-PC-only used in a multi-session campaign. (And I could only really imagine doing so if it was very deliberately the focus of experimental play.) But it’s far from unheard of to see it used in one-shots, and it can be built into games like Ten Candles and Dread.

The far more common approach, of course, is to replace the player’s PC with a new PC. Your character is dead, so create a new one that can join the group.

BACKUP PCs

There are two impediments to consider when using replacement PCs.

First, the time required to create the new character. This can range from trivial in some systems to laborsome to baffling (in games which feature interconnected character creation mechanics but neglect to account for how new PCs could be added to the group).

Second, how to organically integrate the new PC into the existing group. Even when the group leans into the metagame conceit of the replacement (“we trust this newcomer implicitly because we know Mark is playing her”), there can still be the question — when the group is in the middle of a vast dungeon or lost in the untracked wastes of an uncharted jungle — of how and when this new character can actually show up.

Sometimes these two problems can nicely cancel each other out (the time taken to create the new character neatly covers some or all of the time it takes the rest of the group to reach a point where a new character can be naturally introduced), but there can still be logistical and logical hurdles to clear.

The core tip here is that you can solve the first problem by preparing a backup PC. In other words, before your character dies you can already have the replacement character prepared and ready to go. This obviously simplifies things, as you can simply pull out the new sheet and get back to playing lickety-split.

Tangential Tip – Inheritance: If you’re playing in a game where gear is important (e.g., D&D), make sure backup PCs don’t come fully equipped for their current level. You need to work from the assumption that they will either directly or indirectly inherit the wealth/gear of the PC they’re creating. Otherwise every dead PC becomes an incredibly rich looting opportunity and death, rather than being a failure to be avoided, paradoxically becomes a payday which dramatically rewards the group.

You can extend this technique to begin addressing the second problem by giving the backup PC a clear connection to the group. This will often be through the PC who died. For example, “I heard my brother was killed! I have come to avenge them!” (Early versions of D&D actually included rules and guidelines for handling PC-to-PC inheritance and probate.)

BEFORE YOU DIE

Now that you have a slate of backup PCs waiting to step up if a current PC should die AND those backup PCs have existing relationships with the PCs (sisters, apprentices, old college roommates), you can incorporate the backup PCs into the game while the current PCs are still alive.

In many ways, this just makes sense. If you’ve prepped an apprentice who can replace your character Obi-Wan if they die, it makes sense that the apprentice would be part of the story before Obi-Wan’s death.

Of course, once a backup character comes onstage like this, it’s certainly possible that the evolving narrative will make them unavailable or inappropriate for being a PC. That’s fine. (Obi-Wan survived long enough and things went crazy enough with their apprentice that they ended up giving the apprentice a son and just copy-pasting their stat block onto it.)

Tangential Tip – Promoting NPCs: You can flip cause-and-effect here by letting a player of a dead PC take on the role of an established NPC. Even though the NPC wasn’t intended to become a PC, they have an existing relationship with the other PCs and are already integrated into the narrative.

Onstage backup PCs can be played by the GM, but it’s often more effective if the player takes on the role “prematurely” when necessary. To that end, it can be most effective for your backup character to have a connection to a different player’s PC. If you’re playing your own apprentice, there’ll be lots of moments when you’d have to roleplay with yourself (which can lead to skipping or abbreviating those scenes). But if you’re playing Alejandra’s apprentice, then you’ll both be able to frame up interesting scenes and small interactions that will enrich the game.

A variation of this is to create a common pool of backup characters, rather than having each player create their own. You might have one or two or three such characters, and whichever player’s PC dies first (if any) simply grabs whatever character is most convenient from the pool.

These backup characters might also just be temporary roles, which can be played until it’s convenient to create and bring in a fully fledged new PC. (In old school D&D, taking on the role of another player’s hireling is an informal version of this.)

These backup PCs can even account for Total Party Kills (TPKs). “So-and-so has mysteriously vanished/been killed and I’m looking through their notes” is a well-established trope in Lovecraftian fiction, for example, and can easily be transferred to other genres. Laying the groundwork for this sort of insurance policy can be used to frame epistolary play and bluebooking, encouraging note-keeping and enabling different forms of roleplaying that greatly augment a campaign.

You also don’t have to wait for a PC to die in order to swap to your alternate PC. Any number of circumstances might suggest it: Your primary PC might want to retire, be called away on a family emergency unrelated to the main thrust of the campaign, disappear into the Fairylands, or go into a witness protection plan. Or maybe you just want to switch things up.

In fact, you can swap back and forth between your PCs. Or across multiple characters. If your group has established a common pool of PC options, you might even find yourself playing the same character that was previously played by a different player.

TROUPE PLAY

… and we’ve just reinvented troupe play.

First created by Jonathan Tweet and Mark Rein*Hagen in the Ars Magica roleplaying game, troupe play can be broadly defined as a campaign in which the cast of player characters is larger than the number of players and the expectation is that the players will take on different roles for each session or scenario.

This style of play is, of course, quite impervious to the question of, “What does Naito do when his PC dies?” because it has already eschewed the one-to-one relationship between player and PC. More importantly, however, troupe play techniques unlock a lot of unique opportunities at the game table.

Most notably, the constant shuffling of the current group creates a huge variety in personal dynamics and relationships. (You can get a similar dynamic with an open table. The distinction here is that you can get the same effect with a small, dedicated group of players who are sharing all of the experiences in common.)

Ars Magica notably uses the technique to create the dynamic found, for example, in The Hobbit: Gandalf is clearly a much more powerful character than everyone else in the adventuring party. Ars Magica solves the problem of, “Who gets to play Gandalf?” by letting everyone create their own Gandalf and then rotating who’s playing Gandalf and who’s playing the motley assembly of mortals each week.

Similarly, a Star Trek-style space opera can run into the question of, “Who gets to be the captain?” In troupe play, the captain could be a communal character shared by all the players (each of whom also has another bridge crew member as a PC). Each session, a different player gets to play the shared role of the captain.

Ars Magica also associated the concept of a rotating GM into troupe play. I think of this as actually being a distinct technique, but it does combine very well with troupe play. (Since the campaign dynamic already has characters constantly swapping and realigning, it’s easy enough for each GM to have their characters skip the adventures that they’re running.)

Go to Part 1

The agents of the so-called Thieves’ Guild of the North, known as krakenar, identify themselves through the secret mark of a purple squid with an incredible number of tentacles. (The number and styling of the tentacles, perhaps encouraged by the decentralized structure of the Society, has actually shifted over time and in different places. This is not an intentional variation, but those who are deeply familiar with the krakenar may be able to identify the origin of a particular instance of their sign.) It would be a mistake, however, to think of the Kraken Society as purely a bunch of thieves. They are skilled spies and information brokers, with their services often being sought even by legitimate powers. The krakenar thus include highwaymen, merchants, adventurers, and more. Anyone with the right skills or in the right place may be welcomed in the tentacles’ embrace.

In fact, the Kraken Society can often abide in these legitimate, mutually beneficial relationships for years. The krakenar refer to this as the “low tide,” getting their foot in the door by selling valuable information, using their newly privileged position to gain access to more information, rinse, wash, repeat.

What the krakenar are looking for is the “crack in the seawall” (which they may refer to in slang simply as a “crack” or “buckling”); a weakness in the existing systems of power which can be exploited. When they find it, the tide comes in.

Operating at “high tide,” a krakenar cell will use all of its illicit knowledge, covert skills, and criminal resources to disrupt the targeted government or organization. They may attempt to destroy the entire institution, but it’s more common for them to target specific individuals. When the leadership has been crippled or wiped out, the krakenar agents who silently infiltrated the organization during low tide simply slip into the power vacuum. It’s not unusual for krakenar agents to enter power promising to end a threat that was created by the Society in the first place, ensuring acclaim as they rapidly “solve” the problem.

ORGANIZATION

The Kraken Society is cell-based. Each city or region is controlled by a Lieutenant, and although the various regional cells cooperate with each other when it’s beneficial, each is operated as an independent organization and has deliberately limited contact. The local structures of the cells are not proscribed, and can vary greatly (although they tend to value secrecy and usually reflect the Society’s general cell-based methodology).

The inner circle of the Society, which consists of the Lieutenants and a handful of other senior members (called Followers), refers to itself as the coterie. The coterie is privy to the inner mysteries of the cult, which feature a highly mythologized “truth” about the Kraken of the Purple Rocks: They believe themselves to be the servitors of the King of the Trackless Depths (Slarkethrel), who is destined to regain the lost hegemony of the kraken empire which once ruled the seas of Toril. Those who advance further into the mystery perform rites which forge a communion (or supposedly do) with the kraken psychopomps who escaped their empire’s destruction within the sanctuaries of the abyssal rifts.

Those becoming Lieutenants travel to the Purple Rocks, where the Heralds of the Deep indoctrinate them through a series of strange rites. Lieutenants who earn great honor may even be invited to Ascarle.

Most krakenar agents, however, are not part of the coterie and just think of themselves as being part of a criminal organization. Even further out on the periphery is the vast web of front organizations which serve the Society’s will without most of their members or employees ever knowing they are part of it.

CURRENT CELLS

Waterdeep/Skullport: The Skum Lord is an aboleth which lairs beneath Skullport. It has quietly spent centuries inexorably expanding its influence within the Port of Shadows, creating a network of telepathically linked agents strategically placed in the nexus of the underworld. Information flows through Skullport, and the Skum Lord skims his take.

Those long-enthralled to the Skum Lord can be noted by their pale, translucent skin. Known by some as the “pale servitors,” they can be seen moving throughout Skullport and, increasingly, the city above. (The Skum Lord used to have a surface counterpart who served as the lieutenant of Waterdeep, but this hasn’t been true in decades. Slarkethrel’s resurgence in recent years has prompted the Skum Lord to become more active in expanding the Society’s presence in Waterdeep. His agents had some success doing business with Lord Neverember’s administration, but have been disappointed by Laeral Silverhand. It’s uncertain whether the Skum Lord wants to expand his base of power, or if he would be happier finding a new Waterdhavian lieutenant.)

In Skullport, the Skum Lord is also rumored to own forty percent or more of the city’s buildings. Each month the pale servitors come to collect their rent: Sometimes they ask for money. Sometimes they ask questions.

Yartar: Yartar was once the primary center of power for the Kraken Society, but in the late 14^th century the Harpers and the Waterbaron systematically broke their power and drove them out of the city. The Society has only returned to Yartar within the last few years.

• Ghald (sahuagin) and Unferth (male Tethyrian human priest) are partners jointly in charge of Society business in the Dessarin Valley. Their approach has been to recruit powerful, ambitious individuals who already have existing power bases and simultaneously turn them towards myriad schemes. Their theory is that they can easily jettison the failures and consolidate the victors into a larger organization. The actual organization that Ghald and Unferth have reporting directly to them now can actually be described only generously as skeletal, but they’ve managed to project to most of the people they’re working with the illusion that the Society is a monolithic force. (Princes of the Apocalypse, p. 210)
• Lord Khaspere Drylund, one of Yartar’s nobles, is a member of the society and spearheading an effort to replace Ruthiol as the city’s Waterbaron. Drylund notably owns and operates the Grand Dame, a riverboat gambling casino. (Storm King’s Thunder, p. 216)

The Hand of Yartar is an all-female thieves’ guild in Yartar which was actually founded as a front organization for the Kraken Society. In the 14^th century it was led by Semmonemily, a doppelganger who assumed the identity of the Hand’s guildmistress, Emily Iramalac. When the Society was purged from Yartar, however, Semmonemily was killed. The remaining members of the Hand had no idea they’d been operating as part of the Kraken Society and no living connection to the organization. They’ve continued operating independently over the last century, and ironically are now one of the strongest opponents to the Kraken Society as the Society attempts to reassert its control of the underworld in Yartar.

Neverwinter: The Reefkin are a group of merfolk who have set up a commune under Neverwinter Harbor. The upper ranks of the Reefkin deliberately infect themselves with lycanthropy, allowing them to transform into wolves when they wish to pursue shoreside agendas.

The Reefkin have a familiar modus operandi among Slarkethrel’s cults: They rescue drowning sailors and, sometimes after showing them some underwater wonder or vision, deliver them to shore. They claim a life-debt from these sailors, however, and subtly coerce them into the Society.

(Do the Reefkin create their own supply of sailors-to-be-saved by covertly sinking ships with damage disguised to look as if they ran aground on the reefs? Of course they do.)

What happens next depends on circumstances: Some can be immediately forced to do terrible things. Others will be manipulated into countless small acts of treachery, each binding them closer to the Reefkin. The end goal is either recruitment, blackmail, or both.

The Reefkin are also known to have formed alliances with fey powers within Evernight, the dark reflection of Neverwinter which lies within the Shadowfell.

Thornhold: North of Waterdeep along the Sword Coast, Thornhold is an ancient fortress which was unwittingly built by the Margaster noble family directly below the caverns of Clan Stoneshaft in the Underdark below. Control of Thornhold has passed through numerous hands and its connections to the Stoneshaft dwarves have grown stronger (primarily due to a period in which Zhentarim slavers were oppressing and exploiting the dwarves).

Most recently, the Margasters have once again taken possession of Thornhold and forged a tentative alliance with the Stoneshaft dwarves: The Margasters protect the Stoneshaft interests, provide defense against surface threats, and facilitate trade relationships with the merchants of Waterdeep. The Stoneshaft dwarves benefit from having economic access to the wealth of the North; the Margasters reap a healthy profit from their role as middlemen.

There are those among the Stoneshaft dwarves, however, who resent the Margasters: They remember the past betrayals of other surface “allies” and they question what the humans are doing to justify “stealing” their wealth. That’s the sort of buckling that the Kraken Society looks for. A krakenar dwarf named Urnom Telrokak has been organizing disaffected Stoneshaft dwarves. What would be ideal is if the Kraken Society could place an agent within the Margasters, who could then be used to exacerbate the tensions between the two factions.

Caer Westphal: Caer Westphal is the capital of Snowdon, a small isle in the southeastern corner of the Moonshaes which was recently annexed by Amnian nobles. The Society has implanted itself among the native Ffolk of the island, characterizing themselves as a popular uprising.

They have an uneasy alliance with the Brothers of the Beast, another resistance group run by a druid named Heinrich Mucklepratt who can summon strange allies from the island’s moonwells. Their mutual efforts have forced Lady Erliza to openly reveal more and more of her dark powers, increasing tensions on the island.

Luskan: The City of Sails is worth mentioning here because, like Yartar, it was once a major center of power for the Kraken Society. The regional lieutenant here was actually a member of the High Captains who ruled the city, often turning Luskan’s extensive resources to the Society’s purposes. As in Yartar, however, the Society was purged in the late 14^th century. Recent efforts have attempted to find a toehold here, but so far Jarlaxle Baenre and the Bregan D’aerthe have been successful in stamping them out.

Addendum: A Brief Word on My Sources

One of the GM’s most fundamental skills is description. The GM’s words are, after all, the players’ window into the game world. While some performance-enhancing work-arounds do exist (pictures, miniatures, maps, etc.), the bulk of the heavy-lifting boils down to what the GM says and how they say it.

I’ve given general tips for crafting effective descriptions in the past, but if you’re looking to boil things down to some basic procedures, here are some simple formulas. (Our goal here will be to create descriptions that are efficient, effective, evocative, and also easy.)

DUNGEON ROOMS

For a dungeon room:

1. List every notable thing in the room and its position. (“Notable” here can be broadly understood as “thing the PCs will interact with / check out / will inform their actions.” Check out The Art of the Key if you want to delve deeper here.)
2. Use the Three of Five rule by dropping descriptive tags on some or all of the notable things. (In short: “Think about your five senses. Try to include three of them in each description.”)
3. If appropriate, add a verb. (Add action to the scene; e.g., instead of “there is a waterfall,” there is “a waterfall tumbles down the far wall.”)

For example, a room has:

• wardrobe
• bookshelf with arcane tomes
• a goblin

Use the formula to generate:

A horrid stench [smell] emanates from a wardrobe off to your left. On the opposite side of the room, there’s a bookshelf stuffed full [sight] of thick tomes and tightly wrapped scrolls. There’s a goblin pawing through the books on the shelf, knocking them to the ground [verb]. Seeing you, the goblin gapes its maw and screeches [sound].

NONPLAYER CHARACTERS

Whether improvising an NPC on the fly or prepping a Universal NPC Roleplaying template, you can use this formula:

1. An action.
2. What are they wearing?
3. One physical trait.

For example:

Lady Silva is wearing a beautiful blue dress [clothing] that compliments her sapphire eyes [physical trait]. She taps her finger thoughtfully on her chin while looking you up and down [action].

When using the Universal NPC Roleplaying Template, the NPC’s action may be drawn from the Roleplaying section of the template (i.e., the action may be the NPC’s common mannerism). But it doesn’t have to be. The key thing is that you’re establishing the NPC as someone living in the game world; you’re not describing their headshot, you’re establishing them in the scene.

MONSTERS

When encountering a monster:

1. Look at their abilities and attacks; describe how one or more of those are physically manifest.
2. Describe one non-ability-based physical trait. (Use ability scores as inspiration if you’re coming up dry.)
3. Add a verb.

Let’s pick some monsters at random and see how this plays out.

Hill Giant. They have a greatclub and a rock attack. Include the club in the description. You might mention that they’re standing near a pile of rubble (from which you can later describe them snatching up rocks to throw at interlopers). They’re a giant, so… they’re quite tall. (You don’t have to make this complicated.) What are they doing? Gnawing a bone, scratching their head, chatting amongst themselves, swatting giant flies swarming around their head? (Again, you don’t have to be particularly clever. Establish the idea and move into the scene.)

Revenant. One of their abilities is Vengeful Glare, so describe them as having eyes that burn with an eerie blue light. It’s undead and has 18 Charisma, so we can add that it’s an incredibly handsome figure with chalky gray-white skin. Simply add an action appropriate to the current scene.

(As with our other formulas, don’t get hung up on the order here. The description of the revenant, for example, can be: “A dark figure perched atop a rocky promontory, gazing out across the valley. The man’s features are handsome, but its eyes burn with an eerie blue light and its skin is a chalky gray-white.”)

Xenomorph. Their acid blood won’t be immediately obvious, but perhaps we could riff on the idea of bodily fluids by having their jaws slavering with some alien fluid. Their serrated tail whips back and forth, while the flickering fluorescent lights gleam off their black, chitinous exoskeleton.

A WORD ABOUT FORMULAS

Formulas are… well, formulaic. They’ll only take you so far, they can easily become repetitive, and it’s trivial to find examples where the generic formula will be a poor fit for the material. The role of these formulas is not to be the be-all or end-all of RPG descriptions. But if you’re stuck, you can use them as simple recipes to get your brain churning. In fact, you’ll often find that starting with a formula will quickly inspire you to spin out of the mold and create something completely different.

If it does?

Mission accomplished.

PERSISTENT DESCRIPTION

As you’re looking to expand beyond the simple formula, one thing to keep in mind is that description should persist throughout a scene.

I’ve mentioned in the past that the legacy of boxed text can condition GMs to think of description as something you only do at the beginning of a scene or when the PCs enter a room. But that’s an artificial limitation of published modules that you should try to move beyond as you’re running the game.

Add sensory details. You mentioned that there was broken glass on the floor [sight], but as the PCs move into the room you can add the description of the grass crunching under their boots [sound and touch]. The waterfall was described as roaring [hearing], but you can build on that by mentioning the cool mist it throws up [touch] or the fractured reflections of light dancing across the walls of the cave [sight].

Investigate to find new details. When the rogue heads over to check out the wardrobe, take the opportunity to describe the intricate carvings on its doors. After seeing Lady Silva for the first time they grab their drinks and head over to talk to her, and you can take the opportunity to add the color of hair as she turns to look at them.

In the case of a keyed dungeon room, this can actually be structured and prepped. Remember how we listed each notable thing in the room? When the PCs go to investigate or interact with a notable thing, that action simply triggers additional details.

Discover new things. In addition to finding new details about things you already know, the PCs may also discover entirely new things in the environment as they explore it. (What’s inside the wardrobe and causing that terrible smell?) Matryoshka techniques can be a powerful way of running this.

Introduce new elements. A cold wind blows through the cracked window. An otyugh shambles through the door. A police siren wails past the apartment building. A flickering hologram manifests in the center of the chamber. You don’t have to wait for the PCs to investigate to add something new to the scene. The world is a dynamic and active place.

Combine. All of these methods can be used in conjunction. Someone throws a rock with a note wrapped around it through the window. As Bryan goes to pick up the rock, describe the crunch of broken glass under his feet. The rock itself is black obsidian. As he pulls off the note, describe the texture of the vellum.

An image I find particularly evocative is to think of the description of the game world as being layered. You don’t have to exhaustively describe every single detail of an environment in one big glob of exposition. Make sure that the players have the key information they need to orient themselves and understand what’s happening; but then either peel back or add on (whichever visual analogy works better for you) additional layers of description as the scene plays out, slowly building up the mental image of the place like a painter laying down paint on their canvas.

EVOLVING MONSTER DESCRIPTIONS

Speaking of layers, here’s an extra tip when it comes to describing monsters.

The first time the players encounter a monster, you need to establish the monster’s basic visual image. (And you can use the formula described above to do that.) Once the players are familiar with a monster and are able to put a name to it – goblin, gelatinous cube, blood terror, little fuzzies, etc. — you don’t need to re-establish the monster’s basic description each time.

However, this can lead you into the trap of replacing evocative description with bland labels: “You see six goblins.”

What you usually want to do instead, once the players know what a monster is (when you can say “it’s a worg” instead of describing the worg), is to dig one layer deeper by customizing the monster.

These aren’t just six goblins; they’re six goblins dressed in opera dresses. This isn’t just a worg; it’s a worg with a bright red scar over its left eye. This yeti’s fur is matted and filthy. This ogre’s hair is tied up in a topknot. This ghoul’s left arm is broken; it’s hand flopping back and forth on a loose flap of sinew. This wraith wears an iron crown of Angorak.

When the PCs run into a mob, you don’t need to customize every single one of them. (And it’s usually counterproductive to do so.) As a rule of thumb, you’ll want to either customize one of them (it’s a group of yetis, and one of them is wearing a sapphire amulet) or focus on what’s notable about all of them (all of these yetis are missing a finger on their left hand).

(If you’re making one member of a group stand out, it will often be the leader. But this isn’t necessarily true. For example, one worg out of the pack might be limping and ostracized by the others.)

For humanoids, you might use the full formula for an NPC described above. But for monsters you often just need one salient detail to distinguish them.