by K.J. Howe
Living in current times can feel like an episode of the Bionic Woman. Technology promises to make us “Better. Faster. Stronger.” Hi-tech and data-driven “life hacks” promise to save time and deliver uber-success. It’s like we’re on the verge of a technologically created utopia that will allow us to be our best selves with minimal effort. We are promised a successful career in only four hours a week, a cover model body in the same four hours. Precis of key books can be devoured in mere minutes.
With these amazing time savings available, it’s remarkable that I can find enough meaningful tasks to keep me busy. According to the new philosophy, I should be able to do my job, become super fit, and learn to be a gourmet chef by about Friday afternoon.
But what does this all mean for creative fields like writing or music? Are there life hacks available for us, ways to produce profound prose at a prolific rate with a minimum of effort? Are there short cuts and software to produce bestsellers in a shorter period of time? If we are to believe the hype, the short answer is yes! We’re offered everything from ways to write a Moorcockean thousand words an hour (no that is not a typo, it says “hour”) to algorithms that will increase our chances of creating a NYT bestseller. Are we at risk for Artificial Intelligence taking our jobs and churning out bestsellers based in microseconds or is this the dawning of a new age of milk and honey for writers where we can buy a couple of apps, study the data and dominate the market?
The short answer is neither.
Believing that a demanding and complex task like writing a novel can be reduced to a formula, or that the essence of what makes a work of art great (or even successful) can be calculated and emulated is seductive, but misleading. While machines are sensational at counting and organizing, they aren’t unable to understand the words they are counting. For instance, when an algorithm discerns that in paragraphs where J.K. Rowling uses the word Valdemorte there are also negative adjectives, adverbs or descriptors, it only identifies those words as “negative” because a human programmer has told it so. The machine doesn’t ascribe any meaning to good or evil, other than what its programmer tells it to, so the days of a machine writing a novel are a long, long way off.
There are also huge gaps in a computer’s knowledge base that reduce the tool’s effectiveness. For instance, when the algorithm tells us that “sex” is not a good topic for a book, this is because it has not been programmed with data from the thriving e-published erotic fiction genre. Its “diet” has been carefully selected to be NYT best sellers (itself a human curated and adjusted list, not pure data). Even when the software is fed a variety of books, those novels have already been screened and modified by human editors, once again importing significant human gatekeeper elements into a supposedly objective process.
The music world struggles with this theory of data driven product. Many experts have attempted to break down hit songs and discern the formula for popular music. Various musical structures and lengths have been calculated and discussed. Apparently, the perfect song length is 2:42, preferably under three minutes, and definitely not over 3:30. But like Mama Nature in Jurassic Park, music finds a way; giving us Bohemian Rhapsody at 5:55; Sympathy for the Devil at 6:18, Hotel California at 7:12, and Stairway to Heaven at 8:01. Formulas often lead to averages, and do we really want to write an average song or book?
Data-driven writing is also, by design, backwards looking. It looks at what has been successful in the past and quantifies it. Computers offer no guidance on innovation or originality. For example, if the bestsellers analysed were all from a time before vampire novels were popular, it would tell you that writing a vampire novel is not a good way to be successful. If the same calculation was completed after a few vampire bestsellers had dominated the market, the result would be totally different.
While there is clear value in learning structure or understanding the market, we must be very skeptical of promises like writing “under pressure” in your bladder will make you more productive (not kidding). Instead of recommending you use the word “need” between 160-180 times in your novel, or the ratio of adjectives to verbs should be 2.5 to 1, here is a simple formula to help you with your writing:
Creativity plus craft;
Passion plus discipline;
Butt plus chair;
Fingers plus keyboard;