The Bestseller Code by Jodie Archer and Matthew Jockers is a fabulous book if you’re a geeky writer like me. It has loads of delicious graphs and images, mind-numbing academic findings, and basically blows the field of “how can computers help writers?” wide open. But is it a game-changer?
Here’s a clue: I started reading it on Kindle. But I need to scribble on it. Lots. So I bought the paperback, too. Sensitive literary souls – look away now.
Who’s it for?
Writers interested in the future of writing; publishers interested in the future of publishing; data geeks; futurists; curious and open-minded writers; anyone teaching students about writing or literature.
Who’s it not for?
Writers looking for a “how-to” recipe for a bestseller, or a quick fix to improve their writing technique. This book isn’t about that. It’s about a scientific experiment to reveal the inner workings of beststelling books. You may be able to use some of the findings, but you’ll have to dig deep.
What’s it about?
The authors – experts in computational linguistics – crunched 5,000 New York Times bestsellers through vast computers to understand what they had in common.
The scale of data mining allowed them to see patterns that a human could only discern in a lifetime of reading.
Imagine being able to absorb the distilled essence of 5,000 best-selling novels. Like a shot of pure fiction! Sure, you’d miss a lot of pleasure, but you’d also understand a helluva lot about what makes writing tick.
It takes a long time for novelists to learn their craft. As well as practising like demons, they need to devour books, feel story shapes, immerse themselves in different techniques and voices.
Archer and Jocker’s experiment massively speeds up the ability to tune into the hidden codes in bestselling books.
The DNA of writing, if you like.
What are the findings?
The authors narrowed down criteria that bestsellers have in common, including aspects of story, stylistics and the tiny tics that make up authors’ voices.
Some findings aren’t that surprising – about story structure, for example. Story gurus from Aristotle to Vogler to McKee to Goins have always banged on about journeys, transformations, reversals, and proven ways to reach the audience’s hearts. In some ways, this just gives scientific backup to an existing writerly understanding.
But if you’re interested in the close-up minutiae of style and stylistics, you’ll find some fascinating nuggets. Here are some that jumped out for me:
- Want to know whether an author is male or female? Look at how often they use the.
- Do MFA courses train you in a recognisable writing style? Yes!
- Does a journalism background make your writing more engaging? Yes!
- Do bestsellers need loads of sex, drugs’n’ roll? No! They need heroes and heroines with solid day jobs.
Some findings affirm points I already teach as a creative writing tutor:
- If you don’t give your character a face, it’s hard for readers to engage
- Active verbs help momentum
- Use simple vocabulary for pace
The Bestseller Code puts an objective, quantifiable spin on some of these style basics – very useful, if you’re trying to convince a reluctant student.
Reverse-engineering on speed?
Why I’m really excited about this book: like most professionals, I’m constantly striving to be a better writer, to up my game. It’s a life-time’s work and passion.
I once spent a summer dissecting Shakespeare to understand his dramatic techniques, so that I could use them in my plays. I struggled to handle fictional time, so I filleted some fiction by Ray Bradbury and Stephen King, to get a better handle on how they did it.
Recently, I’ve discovered that other pro writers do this, too. Screenwriters, poets, copywriters. It’s a kind of reverse-engineering.
Take the writing to bits, to see how it works.
Then practise, using your new-found tools.
This kind of deep study and deliberate practice is the fast-track to better writing.
And by doing the heavy lifting, computational linguistics can help us understand the patterns faster, and improve more quickly.
Where it grinds to a halt
This book offers tantalising glimpses of what’s ahead, but there are no detailed findings from the actual research. No concrete takeaways, unless you extrapolate your own. It stops short of being actively useful.
The reason is easy to guess. The commercial potential of this research is huge. It’ll be extremely attractive to booksellers, authors and publishers, including the authors’ own house, Penguin.
I hope Archer and Jockers are busy working up Book 2, and not keeping all this fascinating information to themselves.
PS. An uncanny plot twist: after all the research, the book that came out with the highest best-seller score was… no spoilers! Go and read The Bestseller Code.