A couple years ago when reading some television criticism something stuck out at me that had always just floated below the surface for so long: that TV episodes and movies were structured around three acts. And everywhere I went I’d read about how this needed to happen in this act or the other, about how a particular episode conformed or didn’t, how act three was too lethargic or too quickly wrapped up.
Of course some of this was unreflected use of technical terminology. Sometimes it veered toward jargon. It was almost a shibboleth, and I was no part of the filmic tribe.
I.
A.
Before I began thinking critically about television or movies, in particular about their structure, I thought about stage plays, and while my ‘professional career’ was dedicated for years to slightly older examples of the form, I have to conclude that my familiarity with drama started not with Goethe or Lessing or Brecht but with Shakespeare.
In the fourth grade it was “A Midsummer Night’s Dream”; in high school it was a play or two a year. In college there were two Shakespeare seminars, and I was not an English major. This year I got around to “Coriolanus”. But what Shakespeare always provided was a five-act structure, each act composed of scenes. I’d find the same, later, in much of the Goethe I read, the Lessing, the Schiller, Hebbel, Kleist, even some Hofmannsthal, and so on. As long as I sanded down the rough edges, the use of formula went like this: between acts location and time changes could occur (e.g. later that day, the next day, years later; from one end of the country to another, elsewhere in the town, a different part of the house); scenes marked the coming and going of characters.
In general the first act set the stage, introduced characters, and at the end proposed a conflict. The second developed that conflict, the third brought it out in the open and perhaps brought with it a turning or tipping point. In the fourth plots began their march toward resolution, and in the fifth all was resolved, for good or bad. It was in the 3rd act that Macbeth sends murderers after Banquo, when murder comes out in the open and it’s no longer just about achieving the crown. In Act 3 of “Hamlet” we get “To be, or not to be” (scene 1), the play-with-in-the-play confirming Claudius’ guilt (scene 2), Hamlet failing to kill Claudius (scene 3), and Hamlet killing Polonius (scene 4); there is no turning back. In Act 3 of “A Midsummer Night’s Dream” Puck transform’s Bottom’s head into that of an ass (scene 1) and the various ‘groups’ interact fantastically and comically; this is the act dominated by Puck’s mayhem.
“And so on” I would claim, even though I’ve cherry-picked my evidence. And even though much of the five-act structure found in Shakespeare is a relic of later editors and editions making his work more ‘classical’.
Five acts naturally structured how I thought about plays; Goethe’s “Faust, Part II” was ‘classical’ because it used those five acts; his Storm-and-Stress “Götz” had acts but eschewed balance altogether, consisting of myriad scenes in series in the play but skipping from location to location within an act. The lack of act breaks was a sign that “Faust, Part I” either had Storm-and-Stress echoes or was not really meant for the stage. They were aberrations from an accepted norm.
B.
Three acts, however, dominated film and television discussion.
My first thoughts, back then, were “Why did they simplify?” … of course they began with five and reduced to three!
The three-act structure was a formula: if you wanted to write a screenplay, you needed three acts. This of course made no sense to me, this idea that good writing was three-act writing. Having a weak act this or that was a structural fault. Finding examples of it everywhere was not exactly an endorsement, only evidence of its dominance.
And yet there’s a simple, naive justification for a three-act structure as well. Temporally we have (or want?) a beginning, a middle, and an end. In terms of entertaining viewers, we need to hook them, we need to keep them interested, and we need to provide a payoff. In terms of plot logistics, we need to introduce characters and the situation, we need to develop it, and we need to resolve it.
We are building, building, building until the climax, which is less turning point than summit and coming together of all conflicts and tensions. From that point on questions are answered, tensions resolved.
Yves Lavandier (“Writing Drama“) makes it slightly more centered around an action, which has a before, a during, an an after. The action itself is contained only in the ‘during’ and this elevates that middle segment to prominence; if that is the middle act (of three), it follows that it’s the largest act. The first is mainly prelude, the third wrap-up. The climax is then in the second.
What seems to be the case in three-act works is an almost constant building toward a conflict at the end; final battles are indeed final, as they occur toward the end. Conflict is privileged not by being made central, but by resolution being delayed, by the plot always being a matter of increasing accumulation. It’s hard not to see a justification for excess here … a bigger, better climax.
A denouement is an afterthought.
C.
George Bernard Shaw wrote one-act plays in addition to his larger works. Schnitzler’s “Anatol” consists of seven single-act works, and other short pieces, like “Gallant Cassian” and “Literature” also consist of one formal act.
“Glenngarry Glen Ross” consists of two acts. The same is true of “The Dining Room” and various works by Sam Shepard (e.g. “La Turista”, “The Tooth of the Crime”, “Geography of a Horse Dreamer”, and “Operation Sidewinder”).
“August: Osage County” has three acts and a prologue. “Who’s Afraid of Virginia Woolf?” has three acts, each titled (“Fun and Games”, “Walpurgisnacht”, “The Exorcism”). “The Importance of Being Earnest” has three but began life with four.
Chekhov’s “The Seagull” has four acts, as do “The Crucible” and “Long Day’s Journey Into Night”.
“The Hairy Ape” has no acts, but has eight scenes. Goethe’s “Faust, Part I” has various scenes named after locations (or times), but no acts; his “Götz” has even more scenes, though they are wrapped in a five-act structure. Schnitzler’s “Reigen” is a sequence of two-person encounters, each sharing one member of the previous encounter until it all loops back around.
II.
A.
James Bonnet asks, “What’s Wrong With The Three Act Structure“, and his commonsensical rant-critique-essay raises any number of relevant points:
- The act-structure (whatever number) is a relic of the theatre, the need for intermissions, and/or a way to change scenery.
- The modern act-structure in television seems tied to the insertion of commercial breaks.
- Myths, legends an the like do not naturally break along act lines.
- Look at the structure of a movie or play, etc., through the problem to be solved, not through the lens of an arbitrary three-act division.
An so on.
After criticizing the unnaturalness of the three-act structure in modern screenplay writing, Bonnet sneaks in a crypto-five-act structure by way of Aristotle. Bonnet’s argument that stories should be structured by the problem to be solved, and the general structure to solving problems was provided by Aristotle: “All of the structures you might find in the act are already built into the problem solving action that encounters resistance, namely: conflict, complications, crises (turning points) climax and resolution. It is, in fact, the structure of any problem solving action (real or fiction) that encounters resistance.” Our conflict arises in the first act, complications are rising action, and so on. We have one through five.
B.
Any number of sources will regurgitate the workings of the three-act structure:
- Three-act Structure (Robin Franson Pruter, College of DuPage)
- Three Act Structure (tvtropes)
- Three-Act Narrative Structure (Indiana University)
- Story Structure & Plot – Novel Writing Tips (Peder Hill)
It’s much of the same; for more insight, though, one could read excerpts from Lavandier’s “Writing Drama” or read a brief review of it.
But for another counterpoint I might bring up Alex Epstein’s “Craft Screenwriting“, which points out that, indeed, there’s a bit of ‘myth’ surrounding the three-act structure. As he rightly asks: “Where are the act breaks in Hard Day’s Night? All that Jazz? How about Spartacus? Forrest Gump? Apollo 13? Annie Hall? Or the superbly written Wild Things, which has about five or six major twists?” He continues with “The Wizard of Oz” and “The Fugitive”, wondering where the third and second acts begin, respectively. In his critique of the structure he asks, “Suppose you could decide where the third act begins. How would that help you understand how the story works?”
Like Bonnet Epstein focuses on “story” rather than structure (for Bonnet story reduced to “problem” (to be solved)).
C.
Part of the context to both Bonnet’s and Eptein’s criticisms is that the three-act form has become mere formula. As Epstein mocks near the end of the cited page:
Just tell a good story that keeps people interested.
Note, however, that if you are turning in an outline to a producer, he will probably want to know where the act breaks are. Pick some plausible page numbers or events and humor him.
Good writers know what they’re doing. Producers are pencil-pushers and bean-counters. Robots following rules they do not understand, obstacles to be overcome.
Form has become formula.
Plug in the values. Spit out more mediocrity. Good scriptwriting breaks, abandons, or ignores the supposed rules!
Implicit, too, in both critiques, is a separation between content and form, a story to be told and how the story is told.
Epstein, whose blog I quite enjoy, is at times a bit fuzzy on the issue, using the term ‘story’ in a broad sense to cover whatever the writer wants to get across; Bonnet uses it more narrowly but seems to conflate it with plot. But in this regard they’re not really working outside the terminology of the three-act structure, which is also plot-centric, focusing on isolating the most important events and decisions, and also turning plot — and story — into a matter of discreet, often agential units.
III.
A.
Mapping five acts to requisite story elements in certain kinds of narratives? It’s a bit circular. One the one hand we should be amazed that Aristotle’s five aspects or parts of problem-solving map so nicely to five acts; then we might wonder whether the five-act division was modeled upon Aristotle and Horace.
The three-act structure is similar as soon as we assume the temporal progression beginning-middle-end. “Real life” has it (though only for well-defined phenomena … which may be part of the definition of phenomena, but I digress), but should we assume that kind of mimetic fidelity and tether art, at least drama or narrative, to ‘reality’?
We can likewise justify the one-act by way of naive realism: no breaks, one location, but characters can come and go within reason. Let’s be a bit normative: do not overcomplicate.
B.
We have ‘accidental’ formalism: acts to provide intermissions, acts breaks for scenery change or commercial interruption.
An underlying assumption to the one-, three-, and five-act structures as in some way ‘essential’ is a level of mimesis: perhaps a strong unity of time and place, perhaps a fidelity to a notion of temporal progression and classification, perhaps a representation of how a problem is to be solved or obstacle to be overcome.
They all take this realism as a given, perhaps as a purpose, possibly only as necessary or ‘best’. This is the way things work.
Hardly.
And thus why not radical formalism as a way of exploding realism, at least the naive variety. To quote Georgia O’Keeffe again:
Nothing is less real than realism. Details are confusing. It is only by selection, by elimination, by emphasis, that we get at the real meaning of things.
Or then there’s Hegel: “The real is the rational, the rational is the real.”
But why not a formalism that reminds us that what we’re seeing is not real, not naively so, not in an unfiltered or unmediated fashion, and that any attempts to better mimic nature through greater “realism” — better props, better effects, better CGI, better acting and affect, more ‘natural’ dialogue — often only result in greater illusion, greater misleading, a greater veil over the real, not a revelation of it. The formal reminds us that what we’re doing is art and artifice.
Let us assume that a one, three, five, or similar number of acts better represents reality; but why ought art represent reality? And if this is achieved mainly through better imitating that which is found in nature, does this not reduce the artist to slavish re-creator, a tool or cog?
C.
I think of Lavandier’s “Writing Drama” and I think of Lessing and “Laokoon”, of genre poetics, not just as an empirical and practical matter (what works, what doesn’t, measured against experience), but also of analysis and principles; what are the tools available to a given medium and what sorts of (re)presentations are most easily or convincingly produced? What can drama (or film) do?
But I also think of the surface-depth, form-content, representation-story ‘debate’ or ‘division’ and ever since first reading Schiller about twenty years ago I’ve found it to a great extent nonsense, mainly insofar as often surface is content, what is real, and structure and syntax are (or at least contribute to) the meaning. They are costumes, not clothes; they are more than packaging to be discarded in an endeavor to get inside, get at the thing you really want.
It’s a matter of internal and external structures; the mapping of the plot points (see also: Aristotle, Horace) to a five-act presentation is not obvious or natural, but is one schema. What we readily accept is that there are multiple external structures, such as one-, two-, three-, four-, or five-act; one of our problems is in assuming that it’s a matter of mapping a singular narrative structure to a handful of presentational structures, when in fact there’s no reason we can’t have multiple ‘internal’ structures in need of mapping/(re)presentation.
It’s really all quite semi-Kantian. We have a manifold, a myriad, a multitude of bits and pieces, experiences and sensations, sounds and images, emotions and notions. We also have structures and conventions, empty forms reusable time and time again. Story is not found in either, and there is no naked story devoid of structure, no story to be stripped or unwrapped, laid out linearly, divorced of confounding and confusing form; story is the dynamic, schematized synthesis of the manifold and the narrative categories.
Afterword
I began this short piece after first watching “Wyvern”; noticing its almost exemplary B-movie structure; considering parallels with other good, bad, and ugly movies; and writing up some reflections about “Wyvern” and “Naked Fear”.
But I also knew that given the prominence of the three-act structure there was something perverse, perhaps inaccurate, and in the case of “Naked Fear” even a bit unfair, about bringing in a five-act structure, something that seemed so off the mark with regard to modern script writing. And so I decided to fill in a few blanks about the three-act structure, and its limitations and the like led me to Bonnet and Epstein among others. But in the end I found myself back where I often find myself … with Kant.
For better or for worse, through Immanuel is how I tend to schematize the pop cultural world.
This, however, is not exactly a stopping point, a conclusion or resolution, or resignation. Lessing is the superior critic; Schiller employs Kantian terminology and, frequently, a Kantian framework and is the superior critic compared with his mentor. When Kant talks about art he likes and the like, even in 1790 he seems stuck in the Rococo and he has no deep insights. Yet … yet.
There is a parallel between a potential Kantian approach and one provided by Lessing in “Laokoon”, in the introduction to which Lessing considered three ways to consider the work: as a connoisseur (Liebhaber), as a philosopher, and as a critic (Kunstrichter). The first deals with likes and desires and pleasures, how the work appeals; the second analyzes, finds sources and concepts; the third evaluates, judges, compares and contrast, both within a work and between media and genres. When dealing with the aesthetic Kant would in a sense reject both the connoisseur and the philosopher, the former dealing merely empirically and the latter rationally; his third term, however, would be — somewhat!? — different than Lessing’s. At times I like to imagine that in creating a Kantian critic as a kind of Frankenstein’s monster we would not focus pariticularly on Kant’s analysis of the beautiful and sublime, but on aesthetic judgment more broadly as applied to particular works, not merely as ‘artifacts’ but as works that … ‘work’. That affect and effect, that stand in relation to the observer, and that because they necessarily embody a tension between natural products (the materials and rules by which they were made) and the willing subject (and frequently but not always a willing creator) must also be judged by how this synthesis is achieved. Not by mere components, and not by ideas or intent, but by presentation and realization.
I love “Laokoon”, but in describing how poetry and painting work differently, Lessing seems to conflate constraints and predispositions with absolute limits, and with is and ought; Lessing notes how transgressing these observations — almost transfigured into rules –, leads to certain results, often failure, in specific works. He has criteria for judging one work better or worse than another. But from my more formalist perspective I see these descriptions and limitations as challenges, to be tested and redefined, and I see that a Kantian Criticism can embrace this approach.
I’ve failed go into other uses and interpretations of a three-act structure divorced from the plot-temporal, but that can come later.
And yet there’s a simple, naive justification for a three-act structure as well. Temporally we have (or want?) a beginning, a middle, and an end. In terms of entertaining viewers, we need to hook them, we need to keep them interested, and we need to provide a payoff. In terms of plot logistics, we need to introduce characters and the situation, we need to develop it, and we need to resolve it.
So I wrote. And everything written there is an assumption. An unwarranted assumption.
