Kids In Need of Direction (Kind) utilised its Digicel Foundation EPIC grant of $33,750 to equip its computer lab with seven desktop computers and a multimedia projector which will assist with its...
You are here
What makes a bestseller?
Literary criticism is over 300 years old yet, in the 21st century, all literary criticism still boils down to one sentence: “This is my opinion.”
If enough critics agree, then a book becomes part of the canon by self-fulfilling prophecy. On the other hand, editors of publishing houses are tested by the market—either the books they choose to publish sell or don’t.
Now, however, an algorithm developed by English professor Matthew Jockers and editor/writer Jodie Archer may change the face of publishing.
“The bold claim of this book is that the novels that hit the New York Times bestseller lists are not random, and the market is not in fact as unknowable as others suggest,” they write.
“Regardless of genre, bestsellers share an uncanny number of latent features that give us new insights into what we read and why.”
Publishing is a hit-or-miss industry.
Harry Potter, Archer and Jockers note, was turned down by 12 publishers. In the United States alone, 50,000 to 55,000 works of fiction are published every year; 220 to 220 of those reach the bestseller lists of the New York Times; and of that half per cent, just a few authors manage to stay on the list for over 10 weeks, with just three or four selling a million or more copies within a year.
Jockers and Archer used a computer model which could identify thousands of distinct features in thousands of books and find patterns, winnowing an original list of 20,000 traits down to 2,800 key ones.
The model was 80 per cent accurate in identifying bestsellers, a success rate far higher than any literary critic or publisher editor could hope to attain. Good writing, Jockers and Archer say, depends on “having the right words in the right order.”
The books they used in their sample included Great American Novels, Pulitzer Prize winners, and mass market hits. For each one, their algorithm identifies the intrinsic traits of a good book.
“The term ‘bestseller’ should carry no intrinsic comment on quality or type of book and is not a synonym for either ‘genre’ or ‘popular fiction’,” they write.
Indeed, even as they were completing their project, a serious challenge was thrown in their lap with the publication of porn bestseller Fifty Shades of Grey, which even readers who liked it agreed had horrible prose.
But the algorithm was up to the challenge, identifying the fundamental structural features that Fifty Shades had in common with, say, The Old Man and the Sea. (As an aside, Jockers and Archer note that erotica does not in itself sell books, having only a niche market.)
If publishers do eventually start using this algorithm to choose manuscripts for publication, this will be good news for Caribbean writers.
“We are interested in the potential to launch new authors,” Archer and Jockers say.
“We care about bringing people who don’t have the right contacts in New York to a readership.”
In other words, unknown writers won’t have to depend on networking or the literary biases of an editor. They will just to have to write a good book.
The Bestseller Code Jodie Archer & Matthew L Jockers.
St Martin’s Press, 2016