Subtext Revisited

sub 2Over the course of the last few reviews [and thanks to a linguistic tip about adjacency pairs from site contributor Joel Dorland] my thoughts about dialogue formation have sharpened. I still think all the tips established in Bowdlerizing Kant are wildly relevant… even under cross-examination by these new ideas. I just want to broaden the discussion more than I did in those first attempts. Basically, I’m trying to add tricks to the screenwriting tool belt. I hope these ideas will prove useful in the effort to create great dialogue, laden with subtext, and character individuation.

Although this will be [primarily] a linguistic exercise, I’ll admit I am not trained in this branch of social science. It is likely that I will repeat ideas which ought to be attributed to other people, and [quite often] make mistakes in how I apply and use these ideas. I’m also likely to rename basic linguistic concepts based on my lack of familiarity with the subject. Unfortunately, I do not have the time for rigorousness. Life is short and heuristics must be deployed.

I don’t think we will accomplish anything new with this exercise, but I do think focusing creative writers on the building blocks of their medium is likely to IMPROVE creative writing. The reward for writers outweighs the cost to science [with its insistence on rigorousness].

I will be using my intuitions as my guide. As is usual, whenever I begin to think about something, I will start with the idealized version first before making any attempt to support derived concepts with empirical examples.

Lastly, I’ll note that a proper study of these ideas would begin at the beginning—with fundamental signifiers. I’d spend the rest of my life thinking about these atomic pieces of language and never produce one thing that might help a writer. The subject of this inquiry, then, will concern “chunks” of dialogue.

So, yes, I am still chasing refabrication.

I’ll begin [as is most likely proper] with a definition.

A conversation is a social construct designed to force people to trust one another.

Society uses conversation in this way because trusting someone else, especially someone you don’t know, is not in our nature. If humans do not trust each other, then there is no society. [I’d have to do a lot of “black box” hypothesizing to come up with all the premises that show those last two sentences are true. I think that would take us far away from what we are trying to do in this article. For now, I’m just going to assume my almost-syllogism is valid and sound. I’ll go through the effort of proving it when I have more time.]

We will not be able to tread too far into these waters without hypothesizing that all conversations have a goal. [We’ll see later this goal is connected with trust.] Speakers, no matter what they may think they are doing, engage in conversation for a reason. They don’t have to have a good reason, but they, necessarily, have something they want to gain.

You might object that not every time someone wants to share something does there arise a corresponding desire in another person to receive something. This objection is not an impediment. In such situations the recipient of the unwanted conversation could want nothing more than that the conversation end. The point is, whenever someone speaks to you, they do it for a reason. They have a goal, and societal structures being what they are, you are required to participate. (1) When you do, you also have a goal. You also have something you want to gain.

As screenwriters we can use this information to construct our dialogue:

1. Conversations have a goal.
2. Participants in conversations are trying to gain something. [An inference based on the truth of one.]

In thinking about two we can see that if the participants in a conversation are each trying to gain something, then they have unwittingly entered into a game. This brings us to point three:

3. There are always conversational winners and losers.

No doubt someone is now thinking to herself, ‘this is false… there are no winners and losers, for instance, when people make “small talk”’. I, however, think statement 3 stands no matter the type of conversation in which you engage.

It might be easier to see this if we generalized conversations by their types. Toward this end, I will offer [from my own imagination] that there are two main categories:

Category A: Cooperative Conversations
Category B: Confrontational Conversations

The goal, when in a cooperative conversation, is to relay information. Small talk, on the job training, doctor’s visits, are all examples of A. The participants want to share general facts about the world, or about the context which brought them together. It is a property of these events that both parties leave the dialogue enhanced. Rawls would be proud because there does not have to be equality in the enhancement experienced by the participants. It’s just that both boats have to rise at least a little. [It turns out to be a fact about these conversations that both boats DO NOT rise the same amount. As necessarily true as it is that there will be enhancement, it is just as necessarily true that this enhancement will not be equal. [I hope there is time in the full presentation of this article to expand on this idea.]

The goal, in confrontational conversation, is to manipulate information to disadvantage one of the participants. This manipulation can be justified by the context, or it can be unjustified by the context. Whether justified or not, the manipulation will [almost always] center on facts about one of the participants. Examples relevant to screenwriting are: police interviews, fights with spouses, or meetings with bosses about poor results.

I take it as granted that [psychologically normal] participants in cooperative conversations enter their dialogue believing the other person is being as honest as they are. When you signal your speech is cooperative, what you’re really doing is inviting your conversational partner to trust you.

I also take it as granted that participants [whether they are psychologically normal or not] signaling they are engaging in confrontational conversation enter their dialogue with zero idea how honest they want to be. The truth values of the things said in a confrontation have fluid value. The individual statements fluctuate in veracity depending on what the aggressor in the confrontation hopes to accomplish, and how willing the recipient of the aggression is to allow the aggressor to “win”.

Bringing this back to screenwriting, imagine how different a scene looks when a husband begins a scene with:

I tried to call you at lunch time today.

Versus beginning the scene with:

Bill in accounting said he saw your car parked
On Elm Street around lunch time.

What is fascinating to me about analyzing lines like these is the second line is more interesting [at least to begin with], but it immediately sets the husband up as an aggressor. The husband is signaling he wants to find out where his wife actually was when, ordinarily, she would have been eating lunch. If you’re goal with your scene is to immediately ratchet the tension up to eleven so that it might be a great accompaniment for Christoher Guest’s amps, then you want to go with the aggressive dialogue.

However, the first attempt may be even better. It has the potential to bloom from cooperative into confrontational conversation. Is the potential for flowering I see here the biggest piece in the “How to Write Great Dialogue” jigsaw? Let’s sketch this hypothetical scene out a little more, and see how illuminating it can become:

I tried to call you at lunch time today.

Oh sorry, I was working on the Newman account
With Helen.

You two ordered in?

Ate at our desks with all phones silenced.

That’s odd.

We wanted to get a head start on Monday’s presentation
To the board.

Bill said he saw your car parked on Elm Street
Around noon when he went to pick up his
Daughter from preschool.

The wife says nothing.

He said it was still parked there at 12:45
When he came back to work.

The wife begins to object.

Those vanity plates, they are such vanity.

First of all, I’m not trying to pretend this is any sort of magnificent example of dialogue. I wrote this pretend scene in two minutes. What it does show [I think] is how starting your scene in the cooperative frame and then moving it into the confrontational frame is THE method by which dialogue moves from exposition to subtext. Also, isn’t this exactly what we humans do when we talk to one another? Depending on what our goals are, don’t we manipulate conversations by starting out as a cooperator and becoming an aggressor?

In my pretend scene of nine spoken lines, there is a fair amount of exposition:

1. We learn Bill saw the wife’s car at 12 and 12:45
2. The Wife has a presentation about the Newman account on Monday. [She’s under work stress.]
3. Bill has a preschooler.
4. The Wife’s car has vanity plates. This makes the car easily recognizable.

The most important thing we learn, however, is:

5. NOTHIING about the Husband.

This is exactly what we predicted above when we said, in a confrontation, the focus will be [as exclusively as possible] on the recipient of the aggression.

Now let’s summarize what this scene [written in two minutes], hides up its subtextual sleeve.

1. The Husband thinks the wife is having an affair.

2. The Wife is lying to the husband about something. What that thing is, we don’t yet know. Most importantly, the Wife does NOT have to be having an affair. This could be the central conceit of this completed script. The Husband thinks the lies are told to cover up an affair when really the Wife is lying to protect the Husband from some group that wants to harm him. Whatever it is the Wife is hiding, in a good screenwriter’s hands, it won’t be the thing the Husband believes it to be.

3. The Husband, tacitly or not, is spying on the Wife.

4. The Husband thinks the Wife is “vain”. The line about the plates being the most important in the exchange. Not only do they give us the exposition that the Wife’s car is easily identifiable, his subtextual message here tells us that, right now, he really does not like his wife. He wants to find her guilty. We now know he is looking for evidence that convicts her. Evidence that might exonerate her will, likely, be discarded.

Again, the pretend scene is not well-written but, by following the few short rules laid in these opening remarks, this scene is interesting [with re-writes it WILL be good]. The interest the lines generate naturally arises because the scene changes conversational frames. By moving from cooperative to confrontational, the exposition strikes the ear as natural, and subtext blooms.

We screenwriters can turn this into a rule for designing dialogue in our most important scenes:

Subtext naturally arises whenever a character changes from a cooperative to a confrontational frame.

You can find my initial thoughts on subtext here.

***I will continue to expand on these ideas over the next few weeks.


1. I’m uninclined to take the time to prove this is true. You’re a human, you know it is true.


9 responses to “Subtext Revisited

  1. I am really excited about this continuing analysis.

    I agree 100% with what you have so far, and you articulated it far better than I ever could.

    • Joel,

      Thanks for the encouragement. I see this as a three or four part article. We’ll see how it goes. A little encouragement might help me get to the finish line :)

      Definitely owe you another debt for pointing me in the direction of adjacency pairs. They helped to further mechanize the responses I was having to the dialogue in the new batch of scripts.

  2. Very interesting ideas. Your ideas on subtext from previous articles have already helped me tremendously. This article is no different.

    • Thank you, Alex. Very happy the articles have been useful to you. I know I am enjoying writing them.

  3. One of my favourite dialogue scenes ever, useful for this analysis going forward. From THE CONTENDER:

  4. Bleh. Well a search of “gary oldman the contender part 4” on youtube does the trick. It’s the scene that opens with oldman eating steak. Perfect 3 minute scene.

    • That is a great scene. When I wrap the article up with the script posts, I’ll use this scene as one of the examples.

  5. I like the two different types of conversation, coop v confrontational. But I have to disagree about your example, I think “I tried to call you at lunch today,” is an accusation. It’s a great opening line, but it starts right off as a confrontation in my reading of it.

  6. I could see that interpretation. It does depend on what you know about the motivations at the time of speaking.

    You are right that it is leaning confrontational, I would say both parties would recognize this as soon as it is said. The audience, on the other hand would be limited to the information they had prior to the line being spoken.

    It is still possible that the husband has no confrontational intentions. Although, the line does tip its hand more than other lines might.

    A nice observation.

    For greater subtextual impact, I would move this line toward the middle of the scene and draw out the beginning with straightforwardly cooperative dialogue.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s