• Go to 100 Scenes
• Podcast Feed

Note, 4/21/09: This project is currently on hold, as Dave works on another blog-based project, My Way To Canossa.

Note, 1/24/08: This project is becoming more and more interesting, even as it becomes less and less related to the original idea--writing a novel quickly.

"100 Scenes" is the working title of my semi-aleatory novel-in-progress. Right now, the working title describes the process and the structure rather than the content; it's entirely possible that once the novel is written, I'll give it a title based on what actually happens.

Note, 1/24/08: It's pretty obvious, at least to the author, what the title is going to be, but I'll keep the blog project with the original title for now. Okay, I'll tell the rest of the world--at least that tiny subset that looks at this page--the title is going to be "The Tangled Bank."

Much of what might be called the "macro" structure of the novel is determined by rules and random operations. The intent is to remove as much "artistic creativity" as possible from the creation of the overall structure--to generate a structure whereby the individual scenes, or monologues, present themselves defined by a set of givens, so that the imagination may work within those givens. I'm trying to let rules and random procedures create the a social world, leaving my imagination to create incidents and events within that world.

Although I am not trying to write "100 Scenes" in a month--more like a year--(Note, 1/24/08: Yeah, right! This project is going to take as long as it takes.) some of the process is inspired by Chris Baty's "National Novel Writing Month project (www.nanowrimo.org)--in particular, Chris's fairly radical idea that the writer ought to come up with a new idea during the month.

Well that got me thinking--about ways to generate fresh "ideas" for a novel. And pretty soon, I had somewhat idly figured out a method of producing a short novel of 50,000 to 60,000 words, out of 100 scenes of between 500 and 600 words each. Pretty soon, the idea lost any connection with NaNoWriMo, and took on a life of its own.

So here's the system--the randomization tools were very low tech--shuffling 3x5 index cards, rolling dice, and flipping coins..

1. The 100 scenes are to be generated out of all possible pairwise combinations of ten characters, as shown in the graph. The shaded squares represent moments of introspection. Otherwise, the first letter represents the point-of-view character and the second character represents an impact character, a sort of a mini-antagonist. For example, the square BC represents a scene from the POV of character B, describing a dramatic incident that occurred with character C. The square CD represents a scene from the POV of character C, involving an incident with character D.

 

2. There are to be same number of male and female characters. The "odd" characters (A,C,E,G,I) are male, and the "even" characters (B,D,F,H,J) are female. At this point, I created a set of index cards for the characters. All I knew about each character was his or her sex.

3. The characters are to be randomly matched with ages. In practice, this meant creating a set of index cards with an age in the middle of each decade of a human life (5, 15, 25, 35, 45, 55, 65, 75, 85, and 95). I shuffled the character cards, shuffled the age cards, and matched them up one by one. The progression of birthdays ten years apart seemed too monotonous, however, so I used dice--a single die, actually--to add or subtract up to six years from each age. I flipped a coin to decide whether to add or subtract. (In the case of the youngest character, I made a choice that I would only add; for the oldest character, I would only subtract.) I ended up with characters with the following ages: 8, 11, 30, 34, 43, 51, 61, 69, 81, and 89. Now I had characters with ages and sex.

4. I wanted the choice of character names to involve as little "creativity" as possible, so I used the Social Security System database of baby names given in each year to pick the first name of each character, with the additional rule that no two names could sound alike or start with the same letter. For example, for my female character born in 1946, I picked the name "Betty"--the 14th most popular baby name of 1946. Now I had characters with ages, sex, and names.

5. Next, I assigned each character one of the Ten Commandments. This, of course, is an homage to Krzysztof Kieslowski's immortal "Dekalog." I had thought about the Seven Deadly Sins--which are actually a much more humane categorization of human moral flaws--appetites gotten out of hand, rather than pronouncements from on high. But I was already working with the number ten--and I thought it would be an interesting challenge to me, as a non-religious person, raised in Catholicism, to find some serious contemporary moral issue in "Thou shalt keep holy the Lord's Day." (I also found a great conservative Catholic website, The Catholic Refresher Course, that presents 1950s-style strict Catholic teaching, to remind me of just how scared I was when the nuns taught me this stuff.) Although this part of the process is clearly derivative of Kieslowski, and of the medieval morality plays, it seems to me that this way of working has hardly exhausted its potential--the interaction between a traditional moral system and contemporary characters, the tradition illuminating the contemporary, and the contemporary redefining the tradition. The assignment of commandment to character involved another set of index cards, and another shuffle. Now I had ten characters, each with an age, sex, name, and sin--or moral issue.

6. I decided that the scenes would proceed in randomly generated rounds. For the first round, this involved the creation of a second set of character cards, shuffling them, and then picking one from each "deck" to create the first round matchups. In subsequent rounds, I'll have to remove cards from the second deck, to prevent the re-selection of previously-used combinations. For example (to go back to using the the above grid), if the combination FB shows up in the first round, I'll remove the B card from the second deck when I'm picking the impact character for F. Following this procedure, each round will become less and less random--there will be fewer combinations remaining to be picked. It also means that each character will appear in 10 scenes (once per round) as the primary character, and 9 additional scenes as the secondary character.

7. I decided that each chacter would face a major moral choice--in their 7th, 8th, or 9th appearance as a primary character. In some cases, this will mean simply that they will "commit the sin" suggested by their assigned commandmant. In other cases, it means that they will confront a serious moral issue, inspired by a contemporary re-imagining of their commandment. The point of this rule is simple: this is the narrative arc. In those penultimate rounds, the characters will kill (or not kill), lie (or not lie), keep holy (or violate). That, then, is what I know about each character--this person is moving toward a moral crisis inspired by their assigned commandment.

8. It seemed to me that a year, rather than a month, would be good time-frame for the project. That means, approximately, writing one scene every three days, or two in most weeks. I decided that each scene would take place on the actual day I wrote it--or in the previous few days. (Note, 1/24/08: Yeah, right, again. This project is going to take as long as it takes.)

9. Finally--and here I moved out of rule-governed and random creation techniques--I looked over the list of ages and sexes, and sketched out a few family relationships. This person looks like this person's mother, this person could be married to this person, and so on.

10. Then I sat down to write. The first set of givens were waiting for me: Christopher, age 30, wife of Melissa, age 34, and father of Jacob, age 8, needed to have some conflict with Samantha, age 11, daughter of Lisa, age 45. I knew that Christopher was moving toward a moral crisis based on the 2nd commandment, and that Samantha was moving toward a crisis based on the 1st commandment. I wrote the first scene in Christopher's voice, and I decided that this was a project worth doing.

• Go to 100 Scenes

Podcast Feed