It is really quite simple. First, engage in world building and create a “world model” filled with “if this goes on” speculation about science, society and the interaction between the two. Next, create some compelling characters and plot how those characters interact with this world model. Finally, write the novel.
Write the first word. Write the second word. Write the next word. Keep doing that until you have enough words for a novel.
How do we select the words? Now we are getting to the crux of the problem. I suppose we can have something like a new reality series, The Wordsmith, in which the words flirt with the novelist in an effort to be the next word to be written. It sounds dreadful but it is probably not much worse than what is available right now.
If we have constructed the world model with some care, that model provides everything that we need to select the words. All we have to do is pick out one idea from the hundreds or thousands of decisions that we have made as we built this world model and write it down. From there, it is just a matter of following the connections from that one notion to the next notion. Keep repeating that action until we have written words for all of the ideas and plot points from our world model. How hard could that be?
Here is the problem: In this world model there are multiple dimensions. There is time, space expressed as multiple locations, cultures, technologies (real and imagined), and multiple dimensions of imagination. The novel, with its “one word follows another word” pattern, really only has one dimension. The trick in writing a novel is to somehow make that single stream of words in that single dimension evoke the notion of all those multiple dimensions in the mind of the reader.
In the area of software development, which forms the vast majority of my background, we call that mismatch between the world model and the expression of that world model in the form of a novel an “impedance mismatch.” In other words, the two notions do not get along with each other. For example, we might have a website that allows you to order things online and have them shipped to you. Sounds simple enough, but the needs of the user interface as it creates and manipulates the order are much different from the needs of the database that is to store the details of that order for later processing. The user interface thrives on repeating redundant information and the database strives to eliminate the redundancy altogether. We have an impedance mismatch.
This is a serious problem. The engineer who designs the database is going to think about the order data in a significantly different way from the way that the person who designs the webpage. Successful software projects will typically employ different people to fulfill each of these different roles. If the person who designs the database is also permitted to design the user interface, the user is subjected to an interface that exposes all the plumbing. It is possible to accomplish things using such a system, but it is awkward and unpleasant. Likewise if the person who designs the user interface is permitted to design the database, the result is equally awkward and unpleasant, but of course in a much different way. Experience shows that having a third software engineer write “adapter” software to mediate between the two different views of the data improves the quality of the system greatly.
Here is my problem. I am writing a science fiction novel. I am designing the model of the world. I am exposing that model in the form of the words of the novel. And I am the one who is responsible for deciding how to mediate between the model of the world and its expression as a novel. My experience in software development says that this is not going to work very well. I am going to be better at one of these activities and that will twist the novel in unpleasant ways.
When I wrote the first draft of the novel, I had a fairly vague idea of how the world model for the novel actually worked. I am an engineer and realized fairly early on that there were a lot of omissions and downright inconsistencies in how I had expressed that world in the words of the novel. So I did what any good engineer would do: I did a formal design of the world. I wrote a lot of material about how time travel worked. I did a lot of work on making sure that the dates which span several millennia were coherent. Specifications here, details there, exactitude all around.
I then revised many of the existing chapters and wrote new chapters based upon my maturing understanding of the world of the Arisa. The more closely that these chapters were tied to the world building that I had done, the more awkward the expression of those ideas were. I had a strong tendency to expose the plumbing to the reader. In my defense, I thought I’d done a pretty decent job of world building. I was proud of it. I wanted to show people. I wanted to strut my stuff. The problem is that the reader is not interested in the plumbing. Indeed, the reader is strongly averse to knowing that the plumbing even exists much less how it works. I eventually got to that “it just works” state but only after many rewrites and revisions.
In software development we have a role called the user experience (or UX) designer. While this individual, or team of individuals, is aware of how the system might internally do the work, the focus of the UX person is to understand in detail how the system is to be used. Using our example from above of a website to order products, the UX designer figures out in detail the steps that an individual must go through in order to purchase a product. The intent is to make that purchase process as smooth and painless and frictionless as possible. The goal is to make the gap between the desire to buy and the completion of the order as small as possible: “I want this item and suddenly it is being shipped to me.” What could be better than that?
To a large extent, the same is true of the relationship between a reader and the text of the novel. The reader has to understand that the novel involves time travel and that there are certain constraints on how time travel works. The details of the proposed mechanism for time travel are irrelevant to the story and if I, as the author am foolish enough to include them, I have only cluttered up the narrative. In other words, no matter how proud I am of my world building, it is best not to strut my stuff while the reader is watching. In fact, the less that I say the better.
Less is not a bad thing. I am writing a science fiction novel. It is almost certain that this is not the reader’s first science fiction novel. This means that the reader will consume the novel against the context of the genre of which it is a part. An offhanded reference to a time travel portal should be sufficient to get the message across to this audience.
There is an essential difference here, however. The UX designer will typically develop a set of closely related scenarios. For example, consider the following scenarios for purchasing something online: one, I want to order something and have it shipped to my home address and charged to the credit card number that the website already knows about; this covers 80 to 90% of the purchases that might be made on the website. Two, I want to purchase something with my stored credit card and have it shipped to a different address; maybe 5%. Three, I want to purchase something with my stored credit card and have it shipped to a different address along with a gift card; another 5%. Four, I want to preorder a product and be notified when it is available. The UX designer might identify a dozen more possible scenarios, but there will be just a few that are optimized to achieve that frictionless buying experience.
The UX designer has an advantage here over the novelist. The webpage has the ability to be actively dynamic. If the “this is a gift” indicator is set, say via a checkbox on the webpage, then, and only then, does the data associated with handling gifts need to be retrieved and processed. The novelist does not have any option here. Once the path has been picked, the novelist has no choice but to follow that path to its very end. The novelist must make a hard, inflexible decision for the reader whereas the UX designer can allow the individual ordering the product to make decisions and affect the path that is ultimately taken.
“With great power comes great responsibility!” What this all means is that the novelist is the prosecutor, judge, jury, and executioner; there is no one else to claim the credit when it goes right and no one else to blame when it does not. The responsibility is to pick a path that satisfies the needs of most readers. And the truth is that, no matter how much support the author gets from editors or beta readers, no one path is going to satisfy all of the potential readers. While it is possible to write the story from multiple points of view against a common set of events, it is still a case of the linear nature of the novel trying to emulate the nonlinear nature of the world model.
As an inveterate technologist, I do dream of the possibilities of writing and implementing an interactive novel. But this all comes dangerously close to the notion of producing a game rather than a novel. Plus the expectation is that games will have graphics and sound and music, all of which are well beyond my capabilities. There is a temptation to learn too much from web-based technology. The result would be interesting but increasingly far afield from what a novel is all about.
I cannot claim that any of this is useful to anyone, with the possible exception of myself. All this is merely an attempt to capture the inner dialogue that I have had with myself about this thing called “writing a novel.” I’m trying to learn how to do this properly. And like any effort to learn something new, I seek out parallels and similarities with the old things that I already know. Sometimes it gives me an insight that is very useful and sometimes it leads me astray. Either way I am compelled to think about the details of what I am learning. And for me, to think is to write and to write is to think.