Agents of Description, or Agency?
You slide down into the air duct and nod at the rest of your team as they seal the cover in place over your face. You have thirty meters to slither awkwardly in a noisy metal duct barely wider than your shoulders and tight enough that it is hard to draw a full breath.
According to the blueprints, you will come to a junction after 10 meters, and at that point you should be able to turn over on your stomach. First, however, you have the alarm panel on the top of this section of ductwork to worry about.
When you reach it, you see that like the blueprints indicated, it is a cheap older model you can deactivate in your sleep. For fun, you disable it with your eyes closed, and then move on to the junction. Turning over isn’t as easy as you’d hoped. It might be time for fewer calories and more exercise. Wiping your brow with the back of your hand while you have the chance you grit your teeth and move on to the next part. The dangerous part. The frightening part.
The blueprints show a grate at the end of this duct which should overlook the boardroom table. If you can get there at the right time, the meeting you need to eavesdrop on will be in progress and you can record the whole thing. If you get there late, you know with a cold chill in your guts that you will need to enter the room and hack the terminal in the table.
Focus. No time to waste on what if. Gotta keep moving.
You reach the end of the duct and for the first time the blueprints are wrong. Instead of a vertical grate at the end of the duct overlooking the table, what you find, with a curse under your breath, is a horizontal grate right over the table. Someone expanded the room to take in some small office space next to it.
This arrangement has its advantages you realize, after a moment of frustration and anger. This will be better for the audio, and even though the video won’t catch display screens or faces, it will be able to see finger movements and the private dealings the attendees might get up to on their personal devices…
The meeting is about to start. You have to remain perfectly still or someone might hear the duct move or make noise. This might be the most difficult part of this whole caper.
One by one the board members walk in, looking fat and smug in tailored clothes that cost more than you make in a year of good runs. Most are the nameless, faceless sorts you expect in a research center like this, but the last in the door… you have to admit he looks cool. There is something about him which catches your attention and makes you think: don’t mess with that guy or he’ll use your head for a cereal bowl.
Over the top, but hopefully on point
This is a lengthy and somewhat extreme example, but one designed that way to provoke a reaction and demonstrate a point of how a player might feel after a night of this sort of description. After all, if the GM is free to tell you what you do, how you do it, and how you feel about it, why play at all?
If this were a passage quoted from a PBM, we might have a great deal of empathy for the GM and the difficult chore of balancing player agency with pacing and descriptive utility. If we write enough to truly inform the players in deep and meaningful ways, the game might take months to get to the first scene. If we try to limit the scale and scope of action, then players will often – intentionally or not depending on their needs and natures – butt up against those limits.
Around the table, however, this sort of description falls into a category of egregious error I find I just cannot tolerate. I have experienced it only four times ‘in the wild,’ and each time it gave me the same strong reaction, which I put into action by leaving the game when there was no sign of change. Now, in my kinder and gentler 40s I depart with an explanation of why I am departing and reassurances of no hard feelings. In my twenties, I would just find a reason why I could not play and make sure that was true. “Sorry, my shift changed at work. I can’t make game night anymore.” There was no need to mention that I swapped shifts so that I would not have to attend that game. It may seem like this less-direct method was kinder, but I think not allowing a dialogue about a repairable problem with the game shows a lack of regard for the game and its gamers.
Intrusive Description
Injecting moments of description from what is generally the players’ domain is one of the elements of GMing which, in conjunction with stories on rails, creates the perception of the GM as frustrated author. While intentions might be the best, what it leads us to is the GM being more often in the driver’s seat of the why’s and how’s of the characters connecting and interacting with the world, and is at least as active in defining how the characters are perceived.
As those of us who GM more than play can relate, not all of us are frustrated authors. Many of us just have limited time to prepare and play and we want to keep things fun, in motion, and entertaining. In that situation, that sometimes means setting limits on what and how things get done – for the good of the game. This is all well and good, providing we do not cross the line.
Crossing and not crossing the line
When you are just starting out, I am afraid it is more likely that the line will not be seen until it has been well and truly crossed. It becomes visible in the baleful glares of the players. Later on, especially when you know your players well, it really should be discernible without too much trouble. There is something to be said for negative feedback producing quick, if not generally pleasant, results.
The first time I piped up about dissatisfaction regarding intrusive description, I was helping one of my players get ready for his first time in the GM’s chair. This was the 90s and he wanted a chance to run some of the Ravenloft material as a counterpoint to all the Vampire: The Masquerade games cropping up at that time. I had no interest in playing D&D so the two of us conspired together to help him get his sea legs before he had to run the module for real. About mid-way through the first scene, where the characters were lucky enough to get chased by wolves on the road into Barovia, I asked my friend why it was that he kept telling me how my characters felt. His response was that as this was a horror module he wanted to create a sense of fear by telling me how scared my characters were. We talked about that for a while and although I didn’t think of it in terms of ‘show, don’t tell’ at that time, what my advice was amounts to the same thing: “Give me a reason to be afraid, and an opportunity to express my fear.”
That experience was good for me, (and I hope it was helpful for Steve) because it took the edge off of speaking out against an aspect of game play which was not conducive to fun. Prior to that, I was among those who would endure the suffering and then find a polite way out of the game. Who does that ever help? I cannot say it has always been easy to speak up, but I think it has always been beneficial.
Later experiences were for similar reasons; one wanting to set a strong and palpable atmosphere, one trying to show how in tune they were with the characters, and one who had too much to cover and not enough time to cover it. To these I can add my own weakness: trying to guide players into the spirit of a genre with which they are unfamiliar. Sometimes it just seems more efficient and efficacious to present a reaction, or a thought process, or a knowledge-based response as a complete description rather than doing what is really called for: providing enough of the right type of information to facilitate play of a type that the group has agreed that they want to experience.
The urge to produce an outcome of a particular type, or a scene that plays in a certain way, or end a session at a particular point can be very powerful, can contribute to the sense among some players that the game is going well, and can definitely give things your distinct stamp. It can also stamp out the very thing which makes the game a game and you a GM game instead of a story with an author of whatever degree of frustration fits. As GMs the option lies with us whether we will be agents of description, or of agency. As the former requires no players, and can in fact drive players away, I will opt for the latter.
Related articles
- 2 GM or not 2 GM (ukrpdc.wordpress.com)
- Why GM’s Will Love Fate Core (gnomestew.com)
- Minimalist Encounter Descriptions (bigballofnofun.blogspot.com)
Casting Shadows 
I can’t help but think of recent events in our shared gaming when reading your article. Certainly, PBEM and play-at-the-table are different animals, and since PBEM lacks dynanism and spontaneous reactions from either players or GMs, the GM will wind up crossing certain lines – unwittingly, sometimes – in order to have any kind of meaningful progression at all. Otherwise, as you noted, it will take months just to get to the first scene.
I’d like to add to the discussion of player agency using the scenario you used as your introduction to the topic. It’s problematic not just because the GM is rambling and not letting the player get a word in edgewise, but that there are no real choices to be had. Even if the GM stopped and solicited player input from time to time, the only real choices are, “I keep going” or, “I go back.” And, honestly, between those “choices”, the only one that keeps the action going is, “I keep going” (unless the scenario presents a compelling reason to turn back). Every now and then the player might be able to fluff out their contribution with a, “Now I know what a TV dinner feels like. I still keep going.”
To paraphrase a noted game designer, if the game purports to offer three choices, Fight, Bribe, Flee, but Fight is the only option that actually leads to consistent character/story progression, then, really, you have but one choice: Fight. The other two menu options might as well not be there at all for all their usefulness.
For player agency to be more than just the illusion of choice or input, the GM needs to present the player with actual choices that matter, and gloss over the ones that don’t. “You follow the map and get to Castle DungeonCrawlies after about half a day. Now how do you try to get inside?” is better than:
“You go north. Keep going?”
“Yes.”
“Ok, you keep going another two leagues. Still keep going?”
“Uh-huh.”
“You arrive at the junction where the map says you should go West. Do you?”
“Tolkien’s beard, YES!”
Technically the above offers the player more choices, but… I think you get the point.
The opening example is problematic because the GM is assuming control of the character. In a lot of cases this decision gets made because – as you note – there are no meaningful choices at that point in the session. In some modes of play, the assumption of control (although not to this extent!) makes good sense for that very reason. It is essentially how a video game must run, but it impacts strongly against modes of play wherein story is created by player choice. Story-first or strongly narrativist groups will likely have no significant problem with this decision. They are just part of the picture, though. The GM in the example opts to skip over meaningless choice-making to get to a point where choice becomes meaningful on their terms. Why choice is not meaningful here is the question I think we should ask.
Your travel to the castle scenario, having the entire journey glossed over because it is assumed the characters will not see/meet/decide/reconsider something which takes them in a different direction, and having no option to follow a new direction because of a preconceived story arc, decisively limits player agency. It provides more agency in the form of getting inside the castle, but skipping the ability to “turn back or do something else” is the same as providing the meaningless choice to continue or cease to progress in a worthwhile way. Story only exists if the group chooses to enter the castle. The rest of the world has become meaningless.
“Let’s not go to Camelot… it’s such a silly place.”
This sort of analysis of agency and choice is helpful, I think, in figuring out how what the group likes and doesn’t like about play, and identifying if they can work together to provide it. Building a campaign out of the concepts of Quests and Side Quests might make a lot of CRPG players very comfortable, and they might thrill at all the agency tabletop play allows for them in comparison to the video game du jour. That same game would drive me mad at how confining and empty of agency it really was. If on top of that I am hearing about my character’s thoughts and emotions from the GM…? I know that game is not for me.
In a situation where choice must be limited (one shots, play tests, strong themes/genre elements, etc) then to my mind it is imperative that the flow of the game be monitored very carefully to ensure that what little choice is open to the players is always open at full bore, and the players be entrusted with the responsibility to use their agency to work within that framework for the good of the group’s experience.
It would be a poor example of a GM offering meaningless choices if the choices had meaning after all. 🙂 In such a closed scenario it is assumed that the player group has decided fairly decisively that Castle DungeonCrawlies is in fact where they want to go, and that nothing of note interrupts or interferes with the journey to get there. It may be a decision that was made before play, such as the GM buying a module and the group wants to play it, or during play the group found a map, thought about all the treasure and experience points they’d gain, and ventured forth.
In the module scenario, it would almost be better to start in media res with the party standing within sight of their target, unless the journey through the surrounding Forest of Random and Not-So-Random Encounters is part of the module. In the more freeform situation, enterprising GMs of old may use random encounter tables to determine whether the group encounters roving bandits or inclement weather or stumble across The Forgotten Temple of Yucky Worms That Vomit Gold When They Die and decide that Camelot is a silly place and wouldn’t it be more fun to sack the temple and irritate the dormant gods instead?
Even in the free-est of scenarios, the random encounter table only exists if the GM decides to use it. Interruptions to the journey or alternate paths are most likely to be spawned by the GM as part of their world-building, or provided by a pre-made map/module. It is not for the player to say, “We stop in the Village In Imminent Peril to rest and resupply,” unless the GM (or module) has already established that there is a village that may or may not be in imminent peril within reasonable reachable distance.
In a dynamic world, though, a player may feed the GM ideas by inquiring, “Is there anywhere we can rest?” which may incite the GM to “spawn” a village or a trading post where they hadn’t thought to put one before.
It is important for a GM to stay out of a player-character’s head, though it is the GM’s responsibility to describe what the characters perceive with their five (or more) senses, and the GM does have to supply the player with the character’s knowledge of certain things in certain cases (“Roll Intelligence + Occult. Success? OK, here’s what your character knows about vampires…”).
Telling a player how their character feels is a very easy trap or shortcut to fall into, especially in game systems where there are mechanics for influencing a character’s emotions, or magical powers that can sway emotions. “The wizard casts Enchant Person on you: You are now wholly loyal to that wizard and would give your life for her.”
I don’t think a truly emotionless narrative is either possible or desirable. By using evocative language in descriptions, however (“You see a filthy, shifty-eyed urchin sneaking around”), is the GM not indirectly telling the players how their characters (are supposed to) feel?
Ok, a lot of the opus you have included in the reply has already been covered so I will take it as confirmation of us being on the same page and move on to the last paragraph wherein you ask about the connection between evocative description and intended reactions. 😉 I think this is a damn fine question although I disagree with your lead in statement that removing emotion is impossible and undesirable.
I think we already agree that there is a huge difference between, “You see a filthy, shifty-eyed urchin sneaking around,” and ““You see a filthy, shifty-eyed urchin sneaking around who reminds you of [someone good/bad/disgusting/worthy of pity/etc].” I am sure we already agree that there is a huge difference between your example and ““You see a filthy, shifty-eyed urchin sneaking around who makes you feel [emotion x]. Where we do not yet agree is if the descriptive statement, “You see a filthy, shifty-eyed urchin sneaking around,” and the implied value judgments it contains, require the player to respond in the way we intend. I think it is possible that some GMs are skilled enough at verbal manipulation that they can influence the responses of the people at their table. I think it is possible that others could fail to influence the people at their table with two hands and a sledgehammer. What interests me is how loaded do we make a description, why do we make that decision, and at what stage does it begin to intrude on the player’s domain of decision-making about the character’s response.
If I can, I like to stay completely out of the character’s head, except in terms of memory. If the aforementioned urchin actually does resemble another person in the game then if the player is not told outright, or if the description is not properly detailed to alert the player of the similarities, then I have denied the player access to information their character might have had were they real. If I am describing a PC or NPC I try to stick as closely as I can to what they say, how they move, and what can be observed about them. I try to follow the idea that I am not there to lead or mislead the players, I am just serving as their senses. They have to make sense of their senses.
In the case of the urchin, are the kid’s eyes shifty? If so, describe that and let the player decide if it is out of fear, greed, or shaky-eye syndrome. Is the kid sneaking around or not? Is there a less loaded way to describe what form of locomotion is being undertaken? Use it. Again, either way, let the player decide what they make of it. Sure, I might know the kid is a thief and I might be trying to imply the kid is a thief and the character might be appropriately skilled and able to read that at a glance, but it is not my place to say, “the kid is a thief and cannot be trusted” or worse, “the kid is a thief and you don’t trust them.” The character might opt to trust the thief like Indy trusted Short Round, or might be interested in shooting first and asking questions never. Who can say?
The player.
I agree that we can agree to agree. 🙂
I get your point about it being fair to make the narrative language as objective as possible in order not to interfere with a player’s sense of control over what their character thinks and feels. I agree with it in principle. There are some little areas of application in which I might disagree.
As well as memory, I would also supply a player with the knowledge that the character has, and provided they’ve made a successful Intelligence/Wisdom or equivalent check, it is fair to give the player some direction on how their knowledge applies to a current situation. After all, that’s what those checks are for, especially when there is a knowledge gap between the character and the player of that character.
Certain aspects of intuition or feelings that are not easily describable in language are a bit trickier. A GM needs to supply the player with the character’s “sixth sense” (especially if it’s a game where there are stats explicitly called “sixth sense”, “Intuition” or some other synonym), but would a statement like, “You’re not sure why but there is something about this woman that trips your lizard brain, and that lizard is telling you she is dangerous” could be considered too intrusive.
I also wonder if complete objectivity comes at the cost of atmosphere: describing a place a “gloomy” as opposed to “dimly lit” has particular connotations (sadness, spookiness), and it would be a disservice to the narrative to describe a successful seduction as, “The conventionally attractive barman consents to engage in sexual relations”, unless, of course, you’re trying to make a point about how mechanical and passionless the encounter is. That’s what I meant when I suggested that a narrative completely devoid of emotion might not be desirable.
You have brought us around full circle to the original post. Good work, Agent Bunny. It has been a pleasure talking about talking with you. 🙂