Skip to Content

Books

The computer that spots hit novels

Robert Douglas-Fairhurst is unconvinced by Jodie Archer and Matthew L. Jockers’s The Bestseller Code

17 September 2016

9:00 AM

17 September 2016

9:00 AM

The Bestseller Code: Anatomy of the Blockbuster Novel Jodie Archer and Matthew L. Jockers

Allen Lane, pp.242, £20

A few years ago, a reporter from the Chicago Tribune stumbled upon what was widely reported as ‘the Holy Grail of chicken’: a version of Colonel Sanders’s secret recipe that his second wife had scribbled in an album. Anyone hoping that it would contain exotic ingredients such as powdered lark’s tongue or virgin snow from Kilimanjaro was in for a disappointment. Those famous 11 herbs and spices turned out to be sadly humdrum: salt, pepper, oregano, thyme, and so on. It sounded like the kind of thing someone might come up with by dropping a spice rack on the floor and then adding a bag of flour. But none of that mattered to modern fans of KFC. Now they could recreate their favourite fast food much more slowly at home.

In The Bestseller Code, Jodie Archer and Matthew L. Jockers promise to do something similar for popular fiction, although they caution that ‘This is not a prescriptive “how to” book and comes attached to no guarantee.’ (The few tips they do offer tend to be snappy and practical — for example,   ‘endless adverbs and strings of adjectives are like sticking fat tyres and flashing rims on a vintage Jaguar’.) Instead, they summarise some complex analysis they performed on New York Times bestsellers by the likes of John Grisham and Dan Brown — hence the winking allusion to The Da Vinci Code in their title.

Their ‘bold claim’ is that the success of these novels is ‘not random but predictable’. Feed any novel into a computer that has been programmed to notice details that usually slip beneath a reader’s radar, such as the fact that around 30 per cent of a bestseller’s pages are devoted to just one or two topics, and it becomes possible to predict how likely it is to succeed. For Andy Weir’s The Martian it is 93.4 per cent likely; for Michael Connelly’s The Lincoln Lawyer it is a whopping 99.2 per cent.


Some of Archer and Jockers’s claims are eye-grabbing: for example, that their algorithm could guess with 82 per cent accuracy whether a piece of writing was by a male or a female author, or that it could detect ‘with a reasonable degree of certainty’ whether s/he was British or American simply based on how often s/he used the word ‘the’. Other conclusions are also highly plausible. For example, although EL James’s Fifty Shades of Grey might seem an unlikely candidate for literary success — sex rarely sells in fiction — Archer and Jockers point out that it deals with many of the same subjects as more mainstream bestsellers, albeit with a few bolt-on extras like furry manacles and nipple clamps. In fact, with its wholly traditional story arc, and its reliance on themes such as ‘closeness and intimacy’, there was always a 90 per cent chance that it would sell well, albeit perhaps not 125 million copies in four years.

Slightly more surprising is how keen the authors of this study are to gush over James’s prose style as ‘the compelling, living, breathing side of narrative’, claiming that ‘the reader feels the thrum of her words in their bodies like the effect of club music’. Well, maybe. Or maybe the real SM relationship at the heart of Fifty Shades of Grey is the one it establishes with its readers: every time they read one of her clunky sentences it gives them pain, and yet they keep coming back for more.

Either way, Archer and Jockers’s eloquent defence does not really fit with the rest of their book, which makes it clear that the question of quality isn’t one they find terribly interesting. Instead, authors are ‘brands’ who aim to improve their ‘positioning in the market’, while in one highly significant shift they refer to ‘the best writers — or those who will achieve the most readers’. The idea that a novel might reveal its greatness precisely because it disturbs their algorithms is never mentioned. Instead, it seems, every novelist should aim to write like Danielle Steele or Stephen King. After all, as Archer and Jockers point out with wide-eyed enthusiasm, ‘Hit the lists and stay there for while’ and you too might be rewarded with ‘the fancy cars and diamond tiaras of the literary domain’.

What they don’t say is that even authors who stay in the bestseller lists for years sometimes end up being only a footnote in literary history. Who now, apart from scholars of Victorian literature, reads Charles Monroe Sheldon’s 1895 bestseller In His Steps? That shifted over 30 million copies. Or Hall Caine’s million-selling 1901 murder mystery The Eternal City? A bestseller might get you a fancy car; it might even let you join the literary set with ‘houses in the Hamptons’. But when it comes to the sort of fame most authors crave, it might as well be written in dust.


Show comments
Close