The Flesch Test – Friend Or Foe To A Fiction Author?
I recently overheard a conversation between two writers discussing the use of the Flesch test in fiction. And I was intrigued and frankly disturbed to hear how much weight and trust they put in the hands of this algorithm.
What is the Flesch test?
If you’re anything like me the first time I discovered the Flesch test, you’re probably thinking: ‘What the hell is that, and how does it help me?’
So today, I’m going to be discussing the Flesch test’s use in fiction, and why, as a book editor, I encourage you to take its findings with a barrel of salt.
The Flesch test assesses the average level of education required to read a certain passage of text. In short, its ‘readability’.
The algorithm works on a scoring system, with a high score (100-90) indicating ‘easily understood’ writing, and a low score (50-30) signifying a more ‘difficult’ read.
Now, before I begin to discourage the use of the Flesch test, I do want to say this: the Flesch test offers us some useful guidelines. The test is designed to monitor and encourage you to write short, readable sentences that make your writing more widely accessible, and is popular with journalists because of this.
However, the test only factors in length of word per sentences ratio, meaning that a high (good) score can only be achieved by using short words and short sentences.
The test also doesn’t monitor the types of words used, the quality of grammar, or the syntax of the sentence. So a weak sentence may score well on the Flesch test because it is short and snappy, not because it is ‘readable’ or of quality.
So we can already begin to see the flaws of the Flesch test. But how can these flaws be further emphasised when utilising it in fiction?
I invite you to think of some of the most profound, and some would argue, beautiful passages of literature.
Do they stick to simple language and journalistic sentence structure?
Think of classics such Wuthering Heights, Gulliver’s Travels, The Picture of Dorian Gray, Moby Dick, A Tale of Two Cities, and Frankenstein. And think of more contemporary novels such as Milkman, the winner of the Man Booker Prize 2018.
Do you know what all these books have in common?
You guessed it. They all score lowly on the Flesch test, with an average group score of 43 (‘difficult to read’).
But the point I’m trying to make is: when you reflect on the books you love, do you think of how wordy or ‘difficult’ they were to access? Or do you think about how they made you feel when you came to understand them?
For me, it is the latter, and I passionately believe that the ‘difficult’ structure of the books above are a large part of what makes them so beautiful, devastating, and human.
Of course, simple writing can be beautiful too.
I don’t mean to argue that only writing that scores lowly (50-30) on the Flesch test is ‘beautiful’ or ‘moving’. On the contrary, one of my favourite childhood stories, Green Eggs and Ham, is one of the highest scoring (most accessible) pieces of writing that you can run through the Flesch tests.
To further play Devil’s Advocate, the chart below shows the average education level needed to read some of the bestselling books of contemporary times.
As you can see, the majority of the bestselling fiction on this chart requires a lower reading age than any other genre of writing.
But what I’m trying to illustrate by comparing examples of ‘difficult’ and ‘easy to read’ texts is that the sentence structure of your fiction is part of a larger picture, and sometimes, complexity is just as effective as simplicity.
So should you use the Flesch test when editing your book?
I have seen it advised in multiple writers’ forums that: ‘Novels should have a Flesch-Kincaid grade level of 5 (score 90.00-80.00).’
And indeed, in a country where 99% of people are literate but have an average literacy age of 9, ‘easily understood’ texts often prove more popular, and it is important to publish them.
However, as a book editor, I want to emphasise how unhelpful and restrictive this advice is for authors editing their novels.
Yes, good structure and accessibility is a crucial part of good writing (even if you’re going down the experimental route, in which case, your work still needs to be decipherable to a point). But you do not need to obey rigid guidelines such as those set by the Flesch test to ensure that your texts are enjoyed and understood, as I hope the examples I have given you above illustrate.
It is ok to acknowledge the rules, and then ignore them.
As George Orwell’s final and most important of his six elementary rules of writing states: ‘Break any of these rules sooner than say anything barbarous.’
Or as I like to paraphrase: ‘Break all the rules if they render your work boring or risk authenticity.’
Because that is essentially what the Flesch test is: a rule. And as we all know, rules are made to be broken.
So is the Flesch test a fiction author’s friend or foe?
I’d love to know what you think, but as for my opinion as a book editor goes, I believe that if we use machines to assess books, we might as well use machines to write them. And we’ve seen what happens when we leave a bot to write a book.
Good editing is not something that can be achieved with bots and algorithms. So if you're passionate about writing something authentic, unique, and fearless, then I'm the human for you. Please get in touch today to see how I can help you make your book the best it can be.