When I first wrote this post, I knew that it would bring to the forefront the author of the original article and their opinion on the matter. Sue Grafton has caught flack on the issue and when you publicize your opinions and thoughts, inevitably there will be comment and confrontation. But the inspiration for this post wasn’t drawn from her comment out of a personal outrage that I felt at the statement. Rather, it reminded me of a thought that has been brewing in my head for some time. She is not the first, nor will she be the last commentator on the quality of self-publishing. And rather than focus on the person, I would far rather focus on the question. I believe that since it has risen so highly in the discourse on the topic, it deserves a serious analysis. And I hope that through this process I can come to some kind of answer, if only for my own personal edification.
And that brings us to the question posited in the first post in this series: Is there a high level of low quality material published through self-published models. And if so, does it matter?
In this post, I will be focusing primarily on the first part of the question. The second part will be addressed in a following post.
Is there an easy answer? Ideally, I would love to see data that illuminates the answer and unfortunately, I am not finding that. There is no pie chart out there with the percentage of badly written self-published books in relation to those that are mediocre or even great. So my first question is, “Where are we getting the notion that there are a lot of badly written self-published books?”
I imagine it comes from firsthand knowledge in some cases and repetitive recitation of what others have observed in other cases. So, what does that mean? If I randomly choose a self-published book or three and find they are awful, is that an appropriate test? Can I say that the majority of self-published books fall into the same bucket?
I hope the scientist in you shivers, because this is hardly a representative sample. So first we need to collect a sample group to test (We can’t read ALL the books out there, so we have to select a number based on criteria that will hopefully show the proportions of quality in the whole). So how do we determine a sample group? Are the books read limited to a specific genre? How are they selected? If I were to simply read the self-published books found on the Amazon Top 100 lists I would be wary to claim I had significantly seen the level of work being produced. There are so many factors that contribute to a book “making” it to those lists, that I have found they do not necessarily bring us the “best”. In fact, I have had better luck finding great self-published work through word of mouth and serendipitous discovery. (Imagine that, it works just like the traditional model)
Are we limiting our search to Amazon? Or do we include the other options, like Barnes & Noble, Smashwords, Lulu, etc.? And once we have selected where we will pull the books from, how do we then select them? There isn’t a list of self-published books that we can skim to choose books. We could go to Smashwords or others that only cater to self-published authors, but it isn’t a complete look at what is out there as so many skip that as a publishing option. In addition, a number of self-published books look just like every traditionally published book out there. In many cases you have to dig to really decide if it was self-published. And if we miss these kinds of books are we not filtering out a set of the sample that could be quite telling?
This is a significant problem. And I am starting to see something quite interesting, but hang on I will get there.
Let’s pretend we are able to put together a good representative sample. (Honestly, I think it is possible— work intensive, but possible. Anyone looking for a dissertation topic?) Once we have a sample, then we would need to establish the criteria that determine the quality of a book. Are we looking at basic grammatical standards? At what level? Newspapers are supposed to be written at an eighth grade level— is that the standard? What level of errors is considered negligible to not matter? Spelling, commas, sentence structure. Where does it stop?
And are we limiting the definition to just basic writing rules. Or are we looking at storytelling as well? Are we talking about criteria set by the literary types in universities across the globe? Or perhaps criteria that the reading public determine? Hold on. We don’t know what the reading public likes. Why do people flock to Twilight and Fifty Shades when so many cry that it is bad writing? What made Harry Potter an international phenomenon? Why would Transformers 3 make millions? (Wait, sorry about that last one—wrong industry).
I am getting a little snarky about the process, but let me rein that in and state that yes, we do know that there are basic criteria that determine a book’s quality. Good story and editing that ensures a reader never falls out of a book is a sign of good writing. But the specific details that make one book stand above the rest often elude us. The Writer’s Digest recently posted an article on the 21 key traits of bestselling fiction. And these are great, but they aren’t necessarily quantifiable. If I follow every rule, can I guarantee a bestseller? No. Not at all.
Perhaps if we cannot quantify the criteria that make a good book, we should utilize a sample of readers instead. So my proposal is to have thousands of readers read the sample of thousands of self-published books, provide their opinions and then evaluate the opinions over that broad spectrum. If we could only tap into this unutilized source of data…
Actually, this is already there. We have Amazon and Barnes & Noble, and GoodReads, and Smashwords. Thousands of self-published books are being uploaded, sold and read daily. The answers are there. Do we know if the collective whole is bad? If you sort the searches in Amazon by the ratings, you get self-published and traditionally published books running through the list from top to bottom. Some of those books that rise to the top of the list could be characterized as awful writing. But consider this: It is at the top for a reason? Why are people reading it? Why are they loving it? Why are bad books making money? Maybe the real question is: Do you really know what makes a good book?
If you hold up two books and cannot tell the difference between traditionally and self-published, then perhaps it is not that distinction that is important. Perhaps it is not the model at all that we are observing as the precipitating factor, but rather the work, effort, and experience being put into the finished product. Perhaps the issue is that there are a number of books out there that are not meeting an accepted standard. Because the self-publishing model is relatively easy to break into, it allows for individuals to produce works that are less likely to be found released through the traditional model.
Yes, as a self-publisher, I can slap stick-figures on a book cover, type nonsense onto a page and throw it on the web. And I am not going to argue that there are not a significant number of badly written books (by whatever criteria) out there. And while I have not been able to do the actual math, the self-publishing model does provide the opportunity to put up just about anything. And inevitably, the quality can range from phenomenal to not so great.
I went through a lot of words going over how we can tell if there is there a high level of low quality material published through self-published models. And ultimately, it was an exercise in futility, because I didn’t give you a definitive answer. I would love to see actual numbers and know how those numbers were reached, because as you can see, there are so many variables it really needs to be a well designed study. But, the truth is, this isn’t the important part. And that is what I will look at tomorrow.
Does it matter?
If there are a ton of badly written self-published books out there, does it matter? What implications does it have on the industry and what does it mean to the traditional, the self-published and the budding author?
And so… until tomorrow.