What do I do?
This is the only question students ever ask about scenework.
Of course it comes encased in particulars like “how do I handle being named two different names” or “how much backstory does my character need” but the essential desire of all improv students is to be comfortable in the unfurling unknown. Ultimately we all want to be confident in our choices no matter the circumstances. Whatever happens in a scene, we want to know exactly how to respond at all times (and be right).
Ability in any craft is honed through trial and error. Improv, however, gives itself an extra level of difficulty in its immense variability. No two scenes are ever the same. It’s learning how to play an instrument except you can’t play anyone else’s perfectly good songs and you can’t play your own songs more than once. What this means is the only way to get better is a hell of a lot of trial and a hell of lot of error with material that is constantly unfamiliar. This is what makes improv so simultaneously exciting and frustrating to practice. Its excessive novelty keeps us engaged but can lead to chaos if insufficiently tamed.
So how do we answer the question when the material we’re working with is so variable? How do we successfully tame novelty on a regular basis if it is constantly changing? I think a distilled question deserves a distilled answer, and mine starts with the approach. What do I do? I do whatever the scene needs at that particular moment.
Truthfully this covers a vast spectrum of activity. One one end it could mean initiating with a robust premise, heavily endowing all our scene partners, driving a lot of the action, and doing most of the speaking. On the other end it could mean staying completely uninvolved just to edit when the scene’s over. Realistically we’re aiming for the middle of that spectrum as much as possible. After all, this is a collaborative endeavor and spending too much time playing near either extreme is a good way to annoy our teammates.
Still, being comfortable with playing at any point in the spectrum is important as long as we’re doing it for the right reasons. Are we making choices because they’re good for the scene or are we making them because they’re good for the individual? In a craft as personal and vulnerable as improv it can be easy to lose sight of the greater picture. Discomfort at any level can stimulate our natural self-preservation techniques of fight (arguing, negging) or flight (under-committing, timidity). Most of us have struggled with trusting our teammates at one point or another which can lead to these same issues as well as steamrolling and stage-hogging. All of these tendencies of discomfort serve the individual at the expense of the scene.
Remember that our job is to create a scene that is as compelling and joyful as possible. Our job is not to be funny - our job is to make the scene funny, which sometimes means being deathly serious. Our job is not to fight for our character’s needs - our job is to fight for the scene’s needs, which sometimes means deliberately harming or allowing harm to befall our own characters. Our job is not even to support our teammate’s characters - yes we want to support their choices, but sometimes the most supportive thing we can do is give them a challenge and be the villain to their hero. The question that should constantly run in the back of our minds is “who is this for?” Am I making this choice for the good of an individual (even if that individual is not me) or am I making this choice for the good of the scene?
What do I do?
Serve the scene.
It would be fair at this point to argue that this answer is unsatisfyingly vague. Yes it establishes an important foundation from which all our choices emanate, but it is not particularly helpful in terms of specific action. So let’s add a layer of particularity - if I’m going to be serving the scene, what does the scene actually need?
Clarity has to do with the nature of all the information in the scene. It’s all the facts - the basic contextual ones like who the characters are to each other and where they are in space and time, it’s character traits and backstory like appearance and role and personal history, and it’s all the specifics that arise over the course of the scene. But it’s not just the facts on their own, it’s also how they relate to each other.
A scene is much more likely to be successful if the audience is able to follow along completely unhindered. In order to laugh they need to be comfortable, and in order to be comfortable they need to know what’s happening or at least trust they will shortly. Foremost this means getting the information from our heads into theirs through verbal or nonverbal means. If we have an idea of where our characters are but we never say it out loud or indicate it through object work, the audience may make an educated guess but will never know for sure. We can get away with a bit of uncertainty here and there, but too much compounded over time creates confusion. Too much confusion leads to discomfort and the audience pulls back. Once they pull back, they aren’t as immersed in the scene as they should be and thus they tend to laugh less. There can be moments of confusion, certainly, as comedy is a tension/release game, but that tension of confusion has to eventually be released with Clarity.
For example let’s say we start a scene by learning that our character is a lawyer. We spend some time doing lawyer stuff when we learn that our character has a pet dolphin in their backyard swimming pool. Let’s say the audience finds that specific image surprising and funny. We might recognize that as a potential pattern and start listing other unusual animals that our character has as pets. What we’ll tend to find is that as the scene continues this sort of pattern has diminishing returns. We won’t have changed anything, but each new addition will get less and less of a reaction. What happened? A lack of Clarity.
It’s not that the pattern is messy, it’s that it’s untethered to the initial context of the scene. Remember that our character is a lawyer. When the dolphin revelation gets a laugh we have to resist our initial urge to immediately run with it as far as we can take it, instead we need to first look back to collect all prior information and carry that along too. What does a pet dolphin have to do with being a lawyer? Maybe we’re a lawyer for an oceanography company and the dolphin helps us interview animal victims of oil spills? The specific justification doesn’t matter as long as the connection is made. From there we can continue the pattern of strange pets as long as we continue to connect them back to lawyer. Whenever the connection between information is not immediately obvious we either need to make it obvious or create the connection.
Every detail matters.
Otherwise, why did it even come up? Clarity is making sure both the performers and the audience are aware of the relevant information in the scene AND understand how it all connects. We add Clarity by recognizing where there are gaps in understanding and filling them in.
A small disclaimer with Clarity is that sometimes an attempt to create it will result in more gaps of understanding - in our attempt to explain we may accidentally make things more confusing. Making connections between seemingly unrelated things can often be challenging and sometimes takes several steps. The trick is to keep going until the connection is complete, to lean in instead of bail out, and of course lots of practice.
Passion is all about the emotional energy in a scene, which is another important aspect of the immersive process. If Clarity helps the audience understand what’s happening, Passion is why they should care. It operates on two levels - Performer and Character.
At the Performer level, Passion means an aggressive commitment to our choices. Ironic detachment, bailing, and breaking are all symptoms of a lack of performative commitment that hinder immersion. It’s a peek behind the curtain and a reminder to the audience that they’re watching an improv show. Under-committing is something most improvisers have encountered due to discomfort or just plain laziness. A full investment not only takes a lot of energy, it often feels vulnerable. But it’s exactly that level of vulnerability that the scene is asking us for. Passion is being willing to step completely out of our ego and put all of our energy into doing whatever the scene needs of us, even if that means doing something that might fail or be embarrassing, and especially if we know we have friends or family or coworkers or a date in the audience. Passion at the Performer level is not being afraid to look silly.
At the Character level, Passion means our characters have a legitimate investment in what’s going on. They have big, honest emotional reactions to whatever is said or done and/or they are highly motivated to accomplish something. The idea is that a scene is a short glimpse into our characters’ lives - a life that extends back in time before the scene began and will extend forward in time after the scene is over (unless they die in the scene). Because we have chosen this specific short timespan in their lives to show to an audience, there must be some reason for it. Why is this particular few minutes what we’ve chosen? What makes these moments important? It might be that this particular window is a good example of their regular behavior; an introduction to a standard day of an interesting character. It might be that it’s a big moment of change in their lives that we watch them experience for the first time. It might be that they are a more or less unremarkable individual finding themselves in a particularly unusual situation. Whatever it is, it should be important enough to justify seeing it.
Bluntly, Passion is giving a shit. It’s Performers giving a shit about their choices, and it’s Characters giving a shit about what’s happening to them.
Caring is contagious.
If as Performers we’re really leaning into our choices and our Characters care about what’s happening, the audience will care too. The more they care, the more they’ll enjoy themselves.
With the exception of blackout-style scenes of less than 30 seconds, most scenes will need some sort of momentum to sustain them. There needs to be some sort of fluctuation of energy over time. There needs to be some sort of scenic Motion.
This could mean our characters are literally moving in physical space while doing some sort of activity. It could also mean the emotional energy of the scene fluctuates based on what happens and what characters do and say. Motion means the scene is in constant need of change. Change is a bit of a tricky area because it doesn’t mean any particular aspect of the the scene can suddenly be different, it means information is revealed which fits the already-established parameters of the scene but creates new circumstances. It’s an ever-expanding understanding of the characters and their world.
There are essentially two models of Motion that are effective for comedy - Heightening and On/Off. In a Heightening model the intensity of the energy increases over time. In an On/Off Model the intensity of the energy fluctuates between high and low.
These models apply to both emotion and action. For example if a scene is about a happy bunny, in a Heightening model the bunny would get increasingly happy over the course of the scene, whereas in an On/Off model the bunny would fluctuate between happy and neutral/sad based on changing circumstances. If the scene is about washing an elephant, in a Heightening model the elephant washing would slowly get increasingly vigorous, whereas in an On/Off model there would be multiple short-but-intense bursts of elephant washing followed by breaks of a similar duration.
In reality, Motion in most scenes won’t be as simple. Depending on the scene there could be multiple things in Motion at any given time which means juggling multiple models or having some combination of both. The type of model we use will also depend on the type of characters in the scene. Characters who use Heightening models tend to be more stable and stubborn. They’re more committed to their emotions and actions and are less willing or able to change their minds or attitude. Characters who use On/Off models tend to be more unstable or flexible. They have short attention spans or big mood swings or are willing to let go of stuff very quickly and move on. Their emotions and actions accelerate and decelerate rapidly. They are easily triggered and they are easily distracted. The type of Motion model we end up using in a scene will likely be inspired by these character tendencies.
An additional trick for On/Off models is to juggle two at once so one’s On is another’s Off. This is especially useful when using emotional and action-based models concurrently. So if the scene is about happy bunny washing an elephant, happy bunny could turn happy On when elephant washing is Off and vice versa. Of course this behavior would have to be justified - maybe playing is what makes bunny happy and elephant washing is hard work which makes it unhappy - but creating some sort of alternating pattern can be a super efficient way to maintain Motion in a scene because it means there’s always something else to do. When the emotional energy of happy bunny’s break time starts to wane, we can always inject some action energy by going back to elephant washing.
Make something happen.
This is especially true at the beginning of scenes for reasons we’ve previously discussed HERE.
An easy way to create Motion even when the scene is just a conversation and a good way to avoid “talking heads” scenes is to steal from a trick from Hollywood and have our characters doing something while they talk. Maybe they’re unloading boxes from a truck like in Law & Order or walking down a corridor like in The West Wing or driving around in a car like in True Detective. Plenty of those scenes are just people talking, but they’re also physically moving and interacting with their environment which makes the scene more dynamic and engaging. Even if what they’re doing isn’t super relevant to what they’re talking about, it’s a great way to add to the immersion. Interacting with their environment is also another great way to showcase character behavior beyond their words. For this reason their movements should be highly intentional - pacing back and forth is a good way to stay in Motion, but it’s not nearly as immersive and meaningful as moving with a purpose.
The simultaneous gift and curse of improvisation is its immense freedom. A scene can be about anything, go any direction at any pace, change direction at any time, and potentially last as long as we’re willing to let it continue. That all seems great except that doesn’t necessarily guarantee an enjoyable experience; it might mean complete chaos and confusion, both of which lead to discomfort and dissolve immersion onstage and off. In order to keep it focused and engaging, the scene needs Limits - some sort of structure or set of rules that keep it contained enough to manage internally by performers and consume externally by audiences.
The easiest and most common form of scene Limits is a Time Limit. For most shows this will mean an edit. In Harold, for example, the longest scenes will run around four minutes before they’re edited. But even an unedited show like the Monoscene uses time limits by having characters enter and exit. It’s not as clear an ending as a traditional edit, but it’s a way to delineate chapters of the show and keep them from getting stale or chaotic. Even if the subject matter is relatively unfocused and there are a lot of moving pieces, the scene likely won’t become too unwieldy or too hard to follow unless it lasts more than a few minutes. Past that threshold, however, and the scene starts needing additional structure to hold it together.
The longer the scene lasts, the more information is added. The more information there is, the more complex the scene becomes. For this reason longer scenes often need to Limit complexity, which can be done in a few ways:
1. Establishing Fixed Dynamics
2. Narrowing the Focus
3. Creating Character Patterns
4. Creating Scenic Patterns
Establishing Fixed Dynamics means defining how the characters interact and never letting it change. The way they feel about each other or how they behave toward each other stays the same the entire scene. Narrowing the Focus means the attention of the scene is consistently centered on a particular subject. It can stray here and there but it always comes back to the main idea that the characters care about the most. Character Patterns and Scenic Patterns are when elements of a character or scene reoccur within a certain framework and may or may not change or heighten.
This is all pretty vague, so let’s take a look at this set by NYC team Fuck That Shit for specific examples.
The first thing to note is that their show has a predetermined scenic Time Limit. They are clearly not doing a Monoscene; the plan is to end scenes at some point to continue exploring the characters and narrative they’ve built. Overall they do 8 scenes in 25 minutes and the length of those scenes generally shortens over time (7>5.5>5>3>.5>1>1>1) which creates a satisfying acceleration in show pace.
Let’s look specifically at their first scene, which lasts the longest and is therefore the most complex. Of all the scenes in the show this is the one that will need the most Limits. With the first line of “I want the traitor’s head” Aaron Jackson quickly establishes a role with status and a character desire with implications toward a larger narrative. Kate Zelensky responds with a strong affirmative while making a big character choice and making Jackson’s role more specific (“your majesty”), at which point Jackson labels her character “cartoonishly evil”. Three lines and 20 seconds into the scene is still a little early for Limits, but they’ve got some solid options. At this point the scene could explore a few different topics - the traitor, the relationship between the two characters, Zelensky’s character being “cartoonishly evil” - and it’s the latter that they jump on first. With the addition of “you have no eyebrows” and “you’re sweating blood” Jackson establishes the first Limit - a Character Pattern - at 30 seconds.
At 45 seconds and six lines in Zelensky gives her character a strong point of view toward Jackson’s character with “I love you. I’ve always loved you. I’ll kill this kitten for you.” which kickstarts the process of defining their Fixed Dynamic. Jackson further defines her character as a dark wizard in his employ and then completes his half of the Dynamic by making it clear he is not at all interested in her at all. Then at 2 minutes he Narrows the Focus by repeating his opening line again: “I want you to bring me the traitor’s head.” Thus far they’ve established a Fixed Dynamic between them and a Character Pattern for the wizard, but Jackson recognizes that they’ve strayed too far from what he thinks the scene should be about - the traitor, specifically the king’s desire for him to be decapitated. It makes sense to Narrow the Focus to this subject because it was the first line in the scene, and at this point Jackson is in position to make this decision because his character is higher status. If the wizard had status over the king, it’s entirely possible Zelensky would have Narrowed the Focus to drive her character desire (to sleep with the king), but she defers to her role as subservient employee and lets his desire be the focus of the scene. Jackson then uses the wizard singing as an opportunity to create the first Scenic Pattern by saying he also has a minstrel in his employ. Dru Johnston quickly enters as the minstrel and confirms the Pattern of the king having multiple employees. This is the scene’s first broad Pattern that isn’t specific to one character’s behavior. Instead it’s a scene-wide Limit that establishes how all the characters in the scene relate to each other. We are 2.5 minutes in.
The trio then spend the next two minutes jumping between playing with the Fixed Dynamic and the wizard’s Character Pattern while Johnston establishes for the minstrel a Character Pattern of singing 80’s music. Terry Withers continues the Scenic Pattern of introducing employees of the king by entering at minute 4.5 as the maid Esmerelda. They play a little more in the established Dynamic and Patterns, Jackson uses his status to Narrow the Focus again (the traitor’s head!) which sends Zelensky and Johnston out of the scene, and at 7 minutes we have our first Time Limit - a tag to edit the scene.
To recap, in order to sustain this scene for 7 minutes Fuck That Shit established 5 separate Limits: a Fixed Dynamic (the wizard loves the king who does not reciprocate), two Character Patterns (the wizard is cartoonishly evil, the minstrel sings 80’s music), a Scenic Pattern (everyone who enters the scene is an employee of the king), and a Narrow Focus (the king wants the traitor’s head). Establishing and then jumping back and forth within these Limits is what kept things simple enough to be comprehensible during what amounts to be a fairly long scene.
Repetition creates stability.
Every scene needs novelty, but it’s the familiar elements that keep it comfortable enough to be sustainable. Finding ways to solidify how characters interact, centering the scene around a particular subject, or creating patterns for characters or the scene at large can help us live more comfortably in scenes as they become longer and more complex. Other than that, we can simply end them with an edit.
The paradoxical thing about improv is it’s helpful to know this theoretical stuff but it’s not helpful to be thinking about it in the scene. Fuck That Shit is not thinking the whole time about all the Limits they have to create; they’re creating and playing with Patterns and bringing the Focus of the scene back to one subject because it feels like the right thing to do at the time. Despite all the rules and best practices we explore in improv rehearsal, improv performance should ultimately be intuitive.
Therefore while we know that the scene has four basic needs - Clarity, Passion, Motion, and Limits - we’re not intentionally seeking them out from the start. It’s only when something about the scene feels off, it’s only when our gut says the scene needs help, that we might recognize which of these elements is missing or underdeveloped and deliberately add them or increase their intensity. Because ultimately we don’t serve the scene by thinking so much; ultimately we serve the scene but letting our intuition lead the way and using our brains as a backup.
That’s easier said than done, but with a lot of practice and making sure our choices stem from doing what’s best for the scene instead of any individual, we’ll eventually get more comfortable living and thriving in the unfurling unknown. Eventually we’ll know exactly what to do (and be right).