Word play and family secrets

For Enid Blyton, it was ‘woof’. For Nabokov, ‘mauve’ And for Ian Fleming? Er, ‘lavatory’. Those were the writers’ most-used words, claims an author who used computer science in a bid to decode the literary greats – and left our critic blowing a fuse

It might just as easily have been called Enid Blyton’s Favourite Word Is Woof, or Agatha Christie’s Favourite Word Is Inquest, or Joseph Conrad’s Favourite Word Is Poop. The author, a literary geek, had what must have seemed like the bright idea of putting hundreds of famous books through a computer and letting it work out each author’s most distinctive words and phrases. Having fed all of Ian Fleming’s James Bond novels into his computer, he emerged with the unlikely news that Fleming’s favourite word is ‘lavatory’.

Oddly enough, Blatt says that Ian McEwan’s favourite word is also ‘lavatory’. I wonder if his computer is overdue a servicing?

While some authors’ favourite words are reassuringly predictable – Tolkien’s top three are apparently ‘elves’, ‘goblins’ and ‘wizards’, and JK Rowling’s ‘wand’, ‘wizard’ and ‘potion’ – others are so farout as to raise alarm bells.

For instance, could it really be true that, of all the words in the world, the favourite of both John Updike and Zadie Smith is ‘rimmed’? Or that Hemingway’s favourite word is ‘concierge’ and Virginia Woolf ’s is ‘flushing’?

At this point, I begin to suspect Mr Blatt’s methodology. Let’s take the claim in the book’s title. He argues that Nabokov’s favourite word is ‘mauve’ because, according to his computer, it is ‘used at least once in all of his eight books’. In fact, Nabokov wrote three times that number of books, but Blatt only got round to feeding eight of them into his computer. Even so, the appearance of mauve ‘at least once’ per book could hardly be said to make it Nabokov’s favourite word. As it happens, mauve makes the grade only because other authors use it so little, which is not quite the same as being his own favourite.

Blatt’s central thesis is that every author, good or bad, upmarket or downmarket, bestseller or worst-seller, has a distinctive literary style, a ‘fingerprint’ that can be most readily uncovered by ‘computer-aided text processing’. Clearly, he’s right about the ‘fingerprint’ – it goes without saying that no two writers are exactly the same – but I am less convinced by his faith in the detective skills of the computer. More often than not, his computer comes up with stuff that would be obvious to even the most backward child. A chapter on the difference between male and female writers promises a lot, but delivers very little. On page 49, we learn that ‘male writers make their female characters scream more often than their male characters’, while on page 53, we learn that ‘both men and women describe men as killing more often than women’. Neither Blatt nor his computer are afraid to state the obvious.

Nor does he shy away from PC sermonising. At one point, he raps virtually every writer who ever lived over the knuckles for using the word ‘he’ four times as often as the word ‘she’. ‘If you are a reader and every book you read doesn’t even achieve a one-in-four ratio, chances are you’re not getting a true reflection of, or gaining a true appreciation for, how other people act in the world,’ he says.

Again, he devotes a whole chapter to measuring the size of authors’ names on the covers of their books before and after they become famous. He goes to great lengths to work out the exact proportion of nameto-cover for Stephen King before (7%) and after (47%) he became a bestselling author, and then concludes that: ‘The fact that King’s first book was the one where his name was the smallest reveals how book marketers think. If you haven’t sold any books, your name is not a selling point, so the publisher keeps it small. If you’ve already had a bestseller and your name would bring those book buyers back around, then they might as well put that name in large print.’ Well, just fancy that! There are numerous other examples of Blatt’s computer using all its resources to come up with a statement of the blindingly obvious. He devotes one chapter to analysing the lengths of the most famous opening sentences ever written – ‘Call me Ishmael’, ‘It is a truth universally acknowledged’… and so forth. First, he devotes a lot of space to explaining how he selected the particular sentences, then he writes them all out for us. Next, he notes that ‘the variation is very high, with several ultra-short sentences and several winding ones, such as The Catcher In The Rye’s 63-word opener’. And then his computer works out that 60% of these opening sentences ‘were shorter than the average sentence in their respective books’. After all this, Blatt concludes that ‘what each of the best openers have in common is not length but a certain originality