What Bad Books Can Tell Us About Good Writing

Today I want to discuss what bad books can tell us about good writing. To do this I’m going to discuss the history of an intentionally bad book—Atlanta Nights—and tell you about something surprising (well, surprising to me) I’ve discovered.

Atlanta Nights

First, Atlanta Nights. This book was created as part of a sting operation against notorious vanity publisher, Publish America. James D. Macdonald organized a group of science fiction and fantasy authors to pull off this travesty, each author taking a chapter (one was computer generated!), with the goal of creating a story so bad only a vanity publisher would accept it. Appropriately, the group pseudonym they adopted was: Travis Tea. (See: Atlanta Nights)

Happily, Publish America accepted the book allowing everyone to ask a very good question: Why on earth would a respectable publisher, one who made their money from book sales, accept such absolute dreck? After all, it was so bad the publisher couldn’t possibly hope to make money on it. Unless, that is, Publish America wasn’t a respectable publisher at all. (If you’d like to read more about Publish America and the controversy swirling around it, head over to Absolute Write.)

What Atlanta Nights Can Tell Us About Good Writing

Here we have a rarity, an intentionally horrible book. It turns out (and this is the surprising bit I’ll go into in more depth at the end of this post) that writing a bad book isn’t easy.

Story vs Prose

Here’s how I look at things, writing—good or bad—is composed of two things, the prose itself and the story the prose expresses. I agree wholeheartedly that the story expressed by the prose in Atlanta Nights is awful, horrible, irredeemable. But the prose itself, it’s actually not that bad. It’s not good, but it’s certainly nowhere near as bad as the story it expresses.

From the outset, I’d like to make one thing perfectly, vividly, clear: Atlanta Nights IS a bad book. I know that’s not a technical way of putting things, saying just that something is ‘bad’ isn’t descriptive. So I’ll let you judge for yourself. What follows is a quotation from Chapter Two of Atlanta Nights:

The Atlanta sun slanted low in the west, rain showers predicted for later that afternoon, then clearing. Bruce Lucent looked from the side window of his friend's shiny Maserati sports car as they wheeled their way westward against the afternoon traffic.

"I'm glad you could give me a ride," Bruce Lucent muttered, his pain-worn face reddened by the yellow sunlight. "What with my new car all smashed and all."

His old friend, Isadore, shook his massive head at him. "We know how it must be to have a lot of money but no working car," he said, the harsh Macon County drawl of his voice softened by his years in Atlanta high society. "It's my pleasure to bring you back to your fancy apartment, and we're all so happy that y'all is still alive. Y'all could have been killed in that dreadful wreck." Isadore paused to put on the turn signal before making a safe turn across rush-hour traffic into the parking lot of Bruce Lucent's luxury apartment building. "Y'all'll gets a new car on Monday."

"I don't know how I'll be able to drive it with my arm in a cast," Bruce Lucent shoots back. "It's lucky I wasn't killed outright like so many people are when they have horrid automobile wrecks." (Atlanta Nights—this link leads you to a free pdf of the story; it’s on the website of Andrew Burt, one of the authors.)

This is certainly NOT good writing, and intentionally so. (This bit was excerpted from the chapter penned by James D. Macdonald.) I’d say the authors collectively called Travis Tea did a fabulous job creating a story no respectable publisher would buy.

But, as I said, there’s a problem. It turns out that while we all intuitively recognize this writing as bad, that, in one respect, it’s ... okay.

Let me explain.

I’ve been creating a program, a writing analysis program, that has the ability to analyze a book and compare it to other books along various dimensions.

For example, my program will look at how many “-ly” adverbs, wh-adverbs, how many superlative adjectives, how many verbs ending in “-ing,” and so on, a book contains. Based on this my program will generate a score for the book.

One thing I was curious about was how close my generated score (a score generated from objective and quantifiable characteristics) would align with the subjective scores I had assigned each book.

The Results

It turns out that the score generated by my program and the subjective scores I’ve assigned to each of the books are strongly correlated. 

So far so good. 

But there is a problem. It turns out that while my program generated scores are quite close to the user defined scores for the higher scoring books that the generated scores are off when it comes to one low-scoring book.

That book is Atlanta Nights.

It turns out that although humans have no trouble identifying Atlanta Nights as bad, it throws my program for a loop. While it should put Atlanta Nights in the same group of books as The Eye of Argon, my program consistently puts it closer to James Patterson’s books (and, while Patterson’s books aren’t everyone’s cup of tea, they certainly aren’t bad in the same way Atlanta Nights is bad.).

In the beginning, this caused me no end of concern. I thought something had to be seriously wrong with my program since it scored Atlanta Nights high.

But, what I’ve come to suspect, is that the writers of Atlanta Nights did one thing well and one thing not so well. What they did well was telling an awful story. What they did not so well was WRITING an awful story. That is, they couldn’t help themselves, their prose itself (as opposed to the concepts expressed by that prose) wasn’t in the same badness category as, say, The Eye of Argon. Don’t get me wrong, I’m not saying it was good! Far from it. But it wasn’t horrible. 

Now, I’m not at all trying to cast aspersions on any of the writers involved in the creation of Atlanta Nights. I’m just saying that, in a way, they failed. Their prose (as opposed to the story expressed by their prose) wasn’t all that bad. Or, rather, wasn’t as bad as some books that have gotten published by traditional, non-vanity, publishers (case in point: The Eye of Argon).

This seems to point to something truly interesting, and the reason I wrote this post: It’s possible that one’s prose style is built up over a long period of time—years—and becomes ingrained, like one’s accent or culinary cravings.

It’s possible that we, as writers, aren’t even completely conscious of our prose style and so find it very difficult to change, even when we want to!

What do you think? Whatever your opinion, I invite you to create a truly terrible microstory of 100 words or less.