Locus published their annual Recommended Reading List earlier this week. I was looking at it and became curious about its composition. Then I became curious about possible trends in the Reading List and looked at the lists for the last five years, 2011 to 2015. I took the data from the Locus website and copied it into Excel so I could manipulate it.
I want to preface this by saying that I believe that the Locus staff works very hard on this list and intends for it to be as comprehensive as they can make it. I know how hard it can be to stay on top of the flood of fiction and other affiliated works that are produced each year.
But I also believe that Locus has a responsibility to think about their biases so that lists of these type don’t inadvertently perpetuate structural inequalities–as our field’s magazine of record, this Reading List is published around the same time that Hugo nominations open and while qualified members of SFWA are filling out their Nebula nomination ballots. The Locus Recommended Reading List is a logical place for many people to start looking for works to populate their ballots.
A note about my methodology. Human identities are fuzzy around the edges. They have to be–none of us is any one thing, we contain multitudes. Quantitative data analysis doesn’t handle fuzzy edges very well. People will often classify themselves as belonging to multiple categories when given the opportunity to do so. I may have chosen incorrect categories for some individuals: I am more than happy to make changes as necessary.
As Locus is a U.S. publication, I defaulted to using U.S. standards of race–this means I put people I believe to be Ashkenazi Jewish in the white category; as far as I know there were no Jews of color on these lists. I understand that many Jewish people do not perceive themselves as white, but I also believed that it would cause confusion if I included them in the POC category. I also felt that it would be problematic to separate people by ethnic or religious identity.
Gender identity is also fluid. I did use a non-binary category for people who identify in that way, however I categorized trans binary people as either male or female. I know that this is not a perfect solution, however the number of people who I know to be non-binary or trans in this dataset is extremely small and it would not make a statistical difference to separate binary people into cis and trans categories.
If you would like a change made to your data, please contact me at eilatan at gmail.com.
Between 2011 and 2015, 1,401 separate works were included on the Locus Recommended Reading list. This includes fiction ranging from short stories to novels, collections, anthologies, non-fiction, and art books.
When there were multiple authors, I counted all the authors and editors, not just the first listed author or editor. You will see gender categories which reflect multi-author/editor works when the gender composition was mixed. There were some single-gender collaborative works, mostly two men working together, but occasionally two women as well.
I am going to present this data as a combination of bar graphs and data tables.
So the first visual is a simple bar graph showing how many works appear on the list each year, from a low of 267 to a high of 307. The breakdown by category is as follows:
I would say this is a fairly even distribution–some years are heavier in one category than in other years, but overall, I can deduce that Locus has a target total for works in each category.
So now we get to where it really starts to get interesting: the breakdown by gender and race. First, I’ll show an overall graph for each, and then I’ll break that down into tables which separate fiction works from non-fiction, anthologies, and art books.
The majority of the authors or editors of the works included on the Locus list are male–over 50% each year. Female authors or editors come in second in the 35-40% range. Mixed gender collaborations are next, followed by non-binary authors and editors.
On to the category breakouts!
As you can see, male authors tend to dominate most of the fiction categories, with two notable exceptions: First Novel and Young Adult Novel. The only category which has non-binary representation is Short Stories, and most works of fiction are single-author works.
This division becomes more pronounced when you look at the other categories, which are curated anthologies, art books, and non-fiction.
The overwhelming majority of editors and authors included are male. Particularly striking is the male dominance of the Non-Fiction and Art Book categories, but I find the dominance of male editors in the anthology categories also worrisome.
While the proportion of POC to white authors/editors is increasing on a year over year basis–which is a positive thing!–the representation this year still only comes to less than 17% of the total. And that’s counting all categories but white together.
Category breakouts, fiction first:
That’s pretty concerning, in my opinion. And now non-fiction, anthologies, and art books.
The overwhelming majority of the works in the anthology, art, and non-fiction categories was written, edited, or curated by white people.
Another axis I wanted to investigate was repeat appearances on the list. There were names that I kept seeing year after year–there are some people who have multiple works on the list each year.
There are 676 individual and collaborative authors and editors on this list of 1,401 works. Of that 676, 255 had multiple works. Those authors are responsible for 980 of the 1,401 works recommended over five years, or 69.95%.
The way I interpret this data is that once you appear on the Locus Recommended Reading List, your chances of appearing in subsequent years goes up. This is not wholly surprising–you would expect to see repetition, but this seems like a lot of repetition.
When broken down by race and gender, this becomes even more concerning.
First up is gender. I included multi-author works with the gender of the first listed author or editor.
And now let’s look at race. As with gender, I combined multi-author works based on the first listed author’s race.
And then I decided to pivot these two sets of data together to see what I could see from the intersection of gender and race.
So what did I learn from all this?
I learned that my initial read of the information was correct: the Locus Recommended Reading List is predominantly white and male. That the same handful of POC writers appear year over year and that many more male writers are much more likely to have repeat appearances. I also learned that POC and women and non-binary people are more likely to appear in the shorter fiction categories, perhaps because there is less financial risk to publishers in those markets.
I learned that we need more POC and women and non-binary editors curating anthologies–it is shameful that there is not a single Reprint/Best Of anthology with a POC editor in five years of data. I know there are POC writing non-fiction and making art, why aren’t they represented? Are their works not sent to the Locus staff for consideration?
This is by no means an exhaustive analysis of the data. I would love to see other people’s perspectives on this. I would love to see an analysis of age distribution or country of origin or works in translation or any number of other things. However, I have spent well over fifteen hours on this and I’m out of gas.
Here’s the dataset. Have fun. Let me know what you find.
I would love for this to open up a wider conversation.
2016-02-06: Please read this comment before leaving a comment complaining about my work and how you believe I should have done it.