Towards a Theory of Plausibility.
I am interested in implausible stories. Every author strives to write a plausible story and no reader will ever enjoy an implausible story, in fact plausibility is often the key target of criticism and many devices are in place to ensure stories are plausible. For example in movie production there are people who specialize in what is called "continuity", ensuring that each scene in plausibly related to the one before. These are the guys who make sure that for each take of a scene all the papers are correctly positioned on a table, the bed is made the same way, and nobody's costume suddenly changes. I have heard that Alfred Hitchcock would not introduce a train schedule into his story unless there actually was a train that kept that schedule, demonstrating an extreme attention to plausibility. Every author make great efforts to avoid implausible and acausal activity in their plots.
However inadvertently implausibility may still appear. The most common form of accidental implausibility in plotline comes in the form of Literary Second Order Confluence (LSOC). This is a form of implausibility that the author does not even realize is present in her work, but nonetheless infects the entire story. Here is an easy illustration of LSOC. Read the two very short stories below. Which one seems like a plausible story?
Story #1: A man just wins a lottery. Then he wins a second lottery! This guy is lucky. A mobster hears about this and kidnaps the man and drives him to Vegas. Under threats the man is forced to roll dice at craps because the mobster feels that with all that luck this guy will be a big winner. Of course our "lucky" man loses and the mobster plans to shoot him. The now "unlucky man" discovers courage that he did not know he had, and escapes. The End.
Story #2 A man wins a lottery. Being a bit irresponsible he goes to Vegas and gambles it all away. In the meantime a mobster kidnaps the man's sister hoping to collect ransom based on the man's lottery winnings. Now the man is in big trouble because the mobster does not believe that the man lost the money in Vegas. What is he to do? Well, he goes to the corner store and buys another lottery ticket. He wins again! My goodness! Now he has the money to pay the mobster. The End.
Both stories are about very similar things. A man, a lottery, and a mobster. Both stories are about a man that wins TWO lotteries. But certainly the second story is not plausible while the first one is. LSOC is my attempt to provide some rigor as to the reason why story number two is implausible. Note that story number two is not technically impossible, that is not the problem. In the first story the man wins two lotteries and the story begins. In the second story the man wins a lottery and the story advances and then he wins a second lottery. It is something about the sequence of the lottery winnings, and about the fact that the lotteries are, by assumption, independent that makes the second story implausible. The hallmark of LSOC is the appearance of two unlikely, independent and necessary events or relationships in the body of a story without any special explanation. In the cases described above the necessary independent and unlikely events are the two lotteries. In the first story our character winning two lotteries provides motive for the mobster. In the second story the first winning is necessary to provide motive for the mobster, and the second one is necessary because that is how the hero gets money for the ransom. This gives us a clue as to why the second story is not plausible and the first one is: each independent lottery winning in the second story is associated with a separate literary purpose while in the first story they are associated with the same literary purpose. The missing explanation of the second story is "Why is this guy able to win a second lottery just when he needs to?" Notice that there are other unlikely events in both stories that are not in any way objectionable. The fact that a mobster decides to get involved is also unlikely and critical to the plot. Note, however, that the involvement of the mobster is NOT independent of the lottery winnings and we shall see that this fact makes the mobsters entrance into the story plausible. As we shall see below independence is critical for LSOC.
Literary Second Order Confluence. Plausibility is part of every writer's intuitive understanding of plot. Only children write blatantly implausible stories and that is because they actually think those stories are plausible. My interest is when an implausible story ends up being written despite the author's best efforts to do a good job. The way this typically happens is through the vehicle of Literary Second Order Confluence (LSOC).
First let's get a handle on a "First Order Confluence" (FOC). FOC is any confluence of occurrences that give a story its direction or critically changes the direction of a story. It is usually an improbable event that provides motive: Everything that subsequently happens in a story will logically follow or even be casually connected to a FOC. Finding money in a trash can is an FOC because of the unlikely confluence of our character with, say, the secret agents who were doing a money drop-off. Stowing away on a pirate ship is an FOC because we are watching the confluence of a character with criminals on a sea voyage. Being a super agent of a top secret organization is a FOC that lets fantastic adventures that do not happen to normal people occur. FOC is what makes a story worth writing, it is what makes the story interesting. In a nutshell, LSOC is a story with two independent FOCs that have two separate literary functions.
LSOC has two key elements:
1) Two FOCs must be present that are statistically independent.
2) Each FOC must have a different literary purpose related to the same story.
 | What is "Independence"? In our usage here, "independence" means that there is no causal connection between events. The fact that the Raider's scored a touchdown in the 4th quarter is entirely independent from the fact that there was a mechanical failure in the starboard engine of a certain passenger airplane. Not only would the engine failure have occurred regardless of the touchdown, but the likelihood of the engine failure would be unaffected by the ability of the Oakland Raiders to score. LSOC demands that the relevant FOCs be independent. In the language of mathematics we need to conclusively show that the probability of the second FOC is unaffected by the fact that the first FOC occurred. P(FOC#2 | FOC#1) =P(FOC#2). If this statement is true, then FOC#1 and FOC#2 are independent. In all cases of LSOC this condition is absolute, you must be able to conclude independence. This is what makes the appearance of LSOC an objective issue, not a matter of opinion. |
| What is "literary purpose"? This is sometimes called "author intent". An author may want to establish a relationship between characters, build tension or create a diversion. All these are literary purposes and when an author uses two independent FOCs for two separate literary purposes, he is guilty of LSOC. |
| What do you mean by the "same story"? Some stories are really many separate stories told in parallel with each other. Two FOCs in two separate stories are by definition independent. However once the stories join, all subsequent FOCs must be plausibly related ( that is, not independent as defined above). |
| Why the terms "First Order" and "Second Order"? A confluence means "a coming or flowing together" and I am using the word to explain that important moments of a story can be described in terms of confluences. All stories are loaded with confluences and usually one confluence leads to another in a plausible, connected way. "First Order" is a way of explicitly, and perhaps redundantly, saying that two aspects are coming together in a simple single confluence. " Second Order " would be the confluence of two confluences which is just to unusual to ignore. |
Going back to the examples at the top of this page, Story #1 has two FOCs, two lottery winnings, but they both support the same literary purpose: motivate the mobster to kidnap the protagonist (The confluence of an ordinary man with a state lottery pick is the particular confluence). The second story has two FOCs that support different literary purposes: the first to motivate the mobster, the second to provide means for a conclusion, i.e. pay the ransom. In both cases the lottery winnings are independent. Notice that the total number of FOCs in a story is not important and there are usually several (perhaps hundreds in a normal size novel). The only thing that matters is the number of INDEPENDENT FOCs. If there are two independent FOCs we have a case of LSOC. In other words, the confluence of two independent confluences is a violation of plausibility. If there are three, then we have a case of literary THIRD order Confluence, "LTOC". This is my best attempt to point out what should be obvious yet seems to crop up over and over in scripts. The best way to fully understand the LSOC is to examine the examples that I have collected from extensive television and movie watching. When you notice an LSOC the feeling is one of confusion. The audience says to themselves, "Hey, these characters are going on about their business as though nothing weird just happened!"
Plausibility vs. Probability. Each event in our day to day lives, no matter how boring, is relatively unlikely. Every day we have a string of unlikely experiences. How likely was it that a red Honda Civic would be in front of you at the stoplight? How likely is it that you caught exactly three lights on your way to work? Isn't it unlikely that of all the homeless people in the city begging for money, the one that you pass by today has a yellow hat and combat boots? The reason we are not shocked by the low probability of any single event in our lives is because however unlikely these events are, they are entirely "plausible". A car WILL be in front of you at the light, you will be stopped by a few lights on the way to work and some homeless person always stands on the corner you typically walk past, so why not the one with the yellow hat and combat boots? Nothing is particularly implausible.
My explicit example of plausibility vs. probability will use the example of a character, call him Hunter, who is tossing a coin in a park. My story is about him tossing this coin ten times and commenting on whether each flip is "heads" or "tails". Now in order to write this story I, as the author, must choose the overall sequence of coin tosses Hunter observes during the course of the story. Say I choose HTHHTHTTTT. This seems ok: it appears to be a plausible sequence that might result from tossing a coin ten times. Now I can write my story. I could also have chosen HHTHTHHTT. That seems plausible also. But What if I chose HHHHHHHHHH? Suddenly, as an author I need to do some explaining. All heads, though just as PROBABLE as any other sequence (any given sequence has the same probability 1:210 or 1:1024), somehow does not seem plausible for a story about a guy who is flipping a fair coin. The reason the first two sequences seem plausible and the third does not has to do with the number of heads and tails that show up: we expect about as many heads as tails and since the third sequence does not meet this expectation it seems implausible. This is not just a trick of human perception: the expectation of a similar number of heads and tails is justified. Technically speaking, the number of sequences that have a similar occurrence of heads and tails is overwhelmingly larger than the number of sequences that have all heads. When we "dip" into all possible sequences we expect to pull one out with a similar number of heads and tails. Interestingly this point is of the greatest physical profundity and leads to the construction of an entire branch of physics called "statistical mechanics" of which thermodynamics is a subset. If I choose HHHHHHHHHH as the sequence and I include an explanation (Hunter discovers that his coin is a magic coin, or that he is living in a weird time wormhole that replays his coin toss over and over or that both sides of the coin are heads or.....etc. etc.) then I have eliminated the independence of each toss and plausibility is restored. Alternatively I could comment that this was a very unusual occurrence , one chance in over a thousand, and this fact motivates Hunter to do something. Either way the author can not afford to simply ignore the odd, implausible outcome of the coin flips.