When I heard people were apparently upset about the gender balance1 of this year’s Hugo winners, I thought I could give the records a quick eyeball and fill the empty abyss of daily existence for a short time establish once and for all whether or not this year was particularly atypical. If there’s one thing known about human nature, it is that concrete numbers resolve all arguments.
Because I don’t want to offend any gods that are lurking about with the sin of excessive perfection, I only looked at the prose fiction categories. Still, even a quick perusal reveals an astounding trend.
The longer data sets are in the end note (because I am pretty sure a footnote of that length would break Tor.com’s footnote system). Here is the Coles Notes version:
Of 65 years in which the Best Novel Hugos were issued, 45 (69%) had finalist ballots significantly skewed in a particular direction. Of 52 years in which the Best Novella Hugos were issued, 36 (69%) had finalist ballots significantly skewed in a particular direction. Of the 54 years in which Best Novelette was offered, 40 (74%) saw finalist ballots dominated by a particular gender. Of the 64 years in which a Hugo for Best Short Story was offered, 50 (78%) saw finalists predominantly of a particular gender.
For some reason, which I do not have time to pursue at this time, Hugo ballots almost never had comparable numbers of men and women. How pronounced this was varies from decade to decade but the trend is quite consistent.
It’s rather odd, therefore, that would be much fuss over gender balance in this particular year rather than some previous year, since until now people seemed perfectly happy with ballots dominated by one gender or the other. No doubt some subtle factor has changed… some esoteric element my quick perusal of the records failed to reveal. Perhaps it is as simple as a sudden embrace of egalitarianism! Feel free to offer other explanations in comments.
End Note One; or the lengthy bean counting section.
First, a couple of disclaimers: I am going to say “finalists” over and over but we only have the identity of the winners from the first few years of the Hugos. As well, the Hugos are a living tree: categories change from year to year and not all of the fiction categories have had awards every year the Hugos were handed out. Oh, and if a single person had multiple works nominated, they get counted for each nomination. Books written by a man and a woman would count as a book by a woman and a book by a man. And I’m not including the Retro-Hugos. Also, see footnote 1.
Best Novel
Years in which no or only one novel by a woman was nominated for the Best Hugo:
- 1953
- 1955
- 1956
- 1958
- 1959
- 1960
- 1961
- 1962
- 1963
- 1964
- 1965
- 1966
- 1967
- 1968
- 1969
- 1970
- 1971
- 1973
- 1974
- 1975
- 1976
- 1977
- 1978
- 1980
- 1981
- 1983
- 1985
- 1986
- 1987
- 1988
- 1990
- 1991
- 1994
- 1996
- 1998
- 2003
- 2004
- 2005
- 2006
- 2007
- 2008
- 2009
Years in which no novels by men were nominated for the Best Novel Hugo:
None that I can see.
Years in which only one novel by a man was nominated for the Best Novel Hugo:
- 1979
- 2011
- 2019
Of 65 years in which the Best Novel Hugos were issued, 45 had finalist ballots significantly skewed in a particular direction.
Best Novella
Best Novella Hugo Awards have been given out since 1968.
Years in which no women or only one woman was nominated for Best Novella:
- 1968
- 1969
- 1970
- 1971
- 1972
- 1973
- 1974
- 1975
- 1977
- 1978
- 1979
- 1980
- 1981
- 1983
- 1984
- 1985
- 1987
- 1988
- 1989
- 1993
- 1994
- 1995
- 1998
- 1999
- 2002
- 2003
- 2005
- 2007
- 2009
- 2015
Years in which no or only one novella by a man (solo or collab) was nominated for the Best Novella Hugo:
- 1990
- 1992
- 2012
- 2018
- 2019
If we combine years where very few women or very few men were nominated, we get 36 years of 52 in which one gender was more represented than the other.
Best Novelette
From Wikipedia: “The Hugo Award for Best Novelette was first awarded in 1955, and was subsequently awarded in 1956, 1958, and 1959, lapsing in 1960. The category was reinstated for 1967 through 1969, before lapsing again in 1970; after returning in 1973, it has remained to date.”
Years in which no or only one story by women was nominated for Best Novelette:
- 1955
- 1956
- 1958
- 1967
- 1968
- 1969
- 1973
- 1974
- 1975
- 1976
- 1977
- 1979
- 1980
- 1981
- 1982
- 1984
- 1986
- 1987
- 1989
- 1991
- 1994
- 1996
- 1998
- 2000
- 2001
- 2002
- 2004
- 2005
- 2006
- 2007
- 2008
- 2009
- 2011
- 2015
- 2016
Years in which no or only one novelette by men was nominated for Best Novelette:
- 1955
- 1993
- 2013
- 2017
- 2019
Of the 54 years in which Best Novelette was offered, 40 saw finalists predominantly of a particular gender.
Short Stories
Years in which no or only one story by women2 was nominated for a Hugo for Best Short Story:
- 1955
- 1956
- 1959
- 1960
- 1961
- 1962
- 1963
- 1964
- 1965
- 1966
- 1967
- 1968
- 1969
- 1970
- 1971
- 1972
- 1973
- 1975
- 1976
- 1977
- 1978
- 1980
- 1981
- 1982
- 1983
- 1984
- 1985
- 1986
- 1987
- 1989
- 1991
- 1995
- 1997
- 1998
- 1999
- 2000
- 2001
- 2002
- 2003
- 2004
- 2005
- 2006,
- 2007
- 2008
- 2015
Years in which no or only one story by men was nominated for Best Short Story:
- 2011
- 2013
- 2017
- 2018
- 2019
Of the 64 years in which a Hugo for Best Short Story was offered, 50 saw finalists predominantly of a particular gender.
In the words of Wikipedia editor TexasAndroid, prolific book reviewer and perennial Darwin Award nominee James Davis Nicoll is of “questionable notability.” His work has appeared in Publishers Weekly and Romantic Times as well as on his own websites, James Nicoll Reviews and Young People Read Old SFF (where he is assisted by editor Karen Lofstrom and web person Adrienne L. Travis). He was a finalist for the 2019 Best Fan Writer Hugo Award, and is surprisingly flammable.
[1]Gender is of course a complicated thing. I realize that by dividing finalists into gender binaries, this analysis is incomplete, and that there have been Hugo finalists who did not neatly fall into one group or the other [see footnote 2].
[2]James Tiptree, Jr. was believed to be ineluctably masculine for a fair chunk of her active career, so for the purposes of this piece I am not counting any nominations received before she was revealed to be a woman.
Thanks for all this diligent work, James.
You may wish to double check the titles of the two tables at the beginning of End Note One. It appears to me that you have the two genders swapped.
Or I’m reading it very wrong.
Good work!
I thing the gender labels on HugoNoms-Gender.jpg are flipped.
Interestingly 2019 *is* unique – it appears to be the first time one gender got exactly one nom for each category.
Sorry if I sound thick but sometimes when I write something, what I intended to put down overwrites what I actually put down and I literally cannot see the error other people see. What I intended was one table that tracks years in which men got no or one nomination in various categories, which is on the left, and one that does the same for women, on the right. Because there were a lot of years in which women got few noms, the one on the right has lot more xes. What did I actually put down?
Awesome, as always. Simple arithmetic speaks such volumes.
thats quite a list, thanks for all your effort![.](//s04.flagcounter.com/mini/Jkzv/bg_FFFFFF/txt_dddddd/border_ffffff/flags_0/.jpg" width="1"/>)
The charts are labeled correctly, but to try and clear up the confusion: James has placed an X in the boxes for years in which none or ONLY one of the finalists in a given fiction category were male/female. For example, in 1979 only one male author was nominated for Best Novel. In the same year, there was exactly one female author nominated for Best Novella, and zero female authors appeared on the ballot for Best Novelette.
Relatedly, you shouldn’t expect to see an even balance of nominees even in the absence of bias. The chance of getting exactly three men and three women on a six-name shortlist is 0.31; most years, there will be more men or more women.
(Similarly, the next time there’s an #OscarsSoWhite storm, bear in mind that you’d expect that to happen one year in four even in a colourblind world…)
Egalitarianism would be the positive way to look at it. The rise of outrage culture would be the cynical one.
I confess, I had not noticed a furore, but I wasn’t particularly paying attention.
The observation that there have been imbalances most years strikes me as not entirely useful. It is certainly true that it used to be a boys only playground. It is also true that pretty much since the puppies were seen off, women have dominated the nominations and indeed the awards.
Not saying that this is a bead thing, but it certainly is a sea change in the Hugo voting patterns.
Thank you for this cogent analysis.
— Michael A. Burstein
Well, the obvious elephant aside, the fact that there’s a strong celebratory narrative of “Look, we’ve finally Fix The Hugos!” being bruited about in certain places probably makes people more inclined to gripe about the fact that what we’ve actually got is just an inverse of the previous skew rather than something which intuitively appears equitable.
Bravo for doing the number crunching though, that’s some A+ work.
ajay @@@@@ 9:
Similarly, the next time there’s an #OscarsSoWhite storm, bear in mind that you’d expect that to happen one year in four even in a colourblind world…
I think you’ve got the math wrong there. Even if we restrict ourselves to a single category (say, Best Actress), the probability of an all-white selection in a genuinely colorblind setting would be about 12%, which means it would happen about one year in eight (assuming that we’re talking about US demographics, with about 65% of the population being “non-Hispanic white” — thus, the odds of getting an all-white set of five people chosen at random would be 0.65^5).
But the #OscarsSoWhite complaint was about a year (2015) in which all four of the acting categories were all-white. The odds of all twenty nominees being white in a colorblind world is roughly 0.65^20 = 0.02%. (2016 was a repeat of 2015 in this respect.)
James, the problem with your approach to this — treating the entire past going back to the 1950s as equivalent to 2019 — is that you run the risk of (unintentionally) normalizing gender discrimination, in that you’re implying a (hypothetical) imbalance now against men would be just fine, because there was plenty of imbalance against women in the past. Which implies that gender discrimination is not really a bad thing in and of itself, as long as it’s somehow balanced over time (in other words, past injustice can be balanced by complementary future injustice).
A better thing to do would be (following on from ajay’s point) to look at how often “extreme” nomination results would happen if gender were irrelevant. For example, having 5 or 6 of 6 nominees being all one gender, as was the case in 2019, has a probability of 12% (making the simplifying assumption that there are only two genders). Having this happen in all four prose categories has a probability of only 0.02%[*], so perhaps there is something a little odd going on.
Of course, weird things happen as well. The sort of naive, every-nomination-is-a-random-draw model I’ve been assuming would probably give a very low probability for two authors being each nominated in two different categories (as happened for Brooke Bolander and P. Djèlí Clark), unless the pool of nominatable authors is really small. (Maybe it is.)
[*] Which looks the same as the #OscarsSoWhite probability I calculated in my previous comment, which is basically just an odd coincidence (more accurate probabilities would be 0.18% for the all-white in all four acting categories case and 0.21% for the 5-or-6-of-one-gender in all four Hugo categories case, If anyone cares about the details…)
“World” awards that are only American (or at least English-language) don’t make sense anyway. If you use numbers for a real world award you get different probabilities for a fair distribution.
birgit @@@@@ 16:
If you use numbers for a real world award you get different probabilities for a fair distribution.
True, but we are talking about English-language science fiction. Assuming the whole world speaks and writes in English wouldn’t really make sense. (And, of course, a “real-world” award wouldn’t have a different gender distribution….)
Why don’t we just set quotas and be done with it?
Peter Erwin @17 wrote: “And, of course, a “real-world” award wouldn’t have a different gender distribution….”
I would expect it to, if gender correlated in any way with human factors such as – in particular – the interests and preferences of readers and writers. Personally I would find it very astonishing if it did not. Of course, interests have always been less skewed gender-wise in SFF compared to genres like Romance or War; then again, SFF is more of a portmanteau genre.
Gerry_Quinn @@@@@ 19
I’ll admit I was assuming that a “world” award implied the population of the world in general, which basically has the same gender distribution as the English speaking/reading/writing world.
But now that I think about it, it’s certainly conceivable that there might be some real-world variations from country to country in terms of the gender distribution of who’s actually writing SFF.
Could it be that this year those are the best books? Is anyone claiming a poor set of nominees? Any obvious snubs? Otherwise I don’t care about sample size.
@18 – I’m with you. Someone please just set the rules so that there’s an exact breakdown of nominees according to racial and gender lines and call it done. I’m tired of the argument.
So, why are the charts so confusing? I feel like I am on the verge of learning something useful.
Generally people are used to counting things, rather than counting the absence or near-absence of such things.
I’m more of a ‘graph’ rather than a ‘table’ guy. Maybe this can add to the discussion.
Hugo award for Best Novel (including Retros). Hope I had been faithful to the genders.
Men are in blue, women in pink.
Hugo Best Novel – Gender Graph
The tables look like there are many women and few men. I also had to look twice to understand what they are actually supposed to mean.
thanks for this info!
@25
thanks for the graph, it’s very helpful.
princessroxana @18, they could do what the Olympics, the Oscars and Wimbledon have done forever, and have “Best Male” and “Best Female” awards.
@@@@@ 16,17
I still don’t understand how works that come from different languages are valid candidates for the Hugos. Like, Cixin Liu wrote “The Three Body Problem” in magazines in 2006, but it’s nominated when it’s translated into English, in 2014.
28: Two problems with that: it immediately excludes authors like Raphael Carter, who don’t fall into one set or the other.
The women’s award will immediately be relegated to second class status.
15: I don’t see why the same explanations used to normalize the situation as it was in, say, 2007 should not now apply to the current situation. Surely people twelve years ago were arguing in good faith, not merely searching for plausible sounding sophistry to support a situation that at that time favoured them?
An interesting stat:
44 male winners/252 male finalists = 17%
22 female winners/76 female finalists = 29%
Women have been traditionally less likely to be finalists, but if they _were_ finalists, more likely to win.
@25 Nice work on the graph, no offence to James, but I find that a more intuitive way of displaying it.
It does beg the question though, what happened in 1978 (or 79?) that there was 80% female nominations?
Data is fun!
For some reason there were only four FIVE four nominees that year. For ome
Otherwise, it was a pretty good year:
Vonda N. McIntyre’s Dreamsnake, Anne McCaffrey’s The White Dragon, C. J. Cherryh’s The Faded Sun: Kesrith, James Tiptree, Jr.’s Up The Walls of the World (withdrawn) and Tom Reamy’s Blind Voices.
You know, I bet Jo Walton’s old series talks about this year.
I kind of agree with @21 – dantrimble. A choice of a “best” really is just that, these were voted the best by those that are eligible to vote. Winning “best” doesn’t mean you’re a fantastic example, it just means you’ve been voted best of that year for that role. If you don’t like the choice then work to join the voting population and change the outcomes of the future as we can’t change the past.
I’m not disputing the fact that it’s obvious that women were skipped over. You can’t have that much of a skew without a true effort to straight up ignore female authors. However, we can all scream about it from the sidelines and say “I’ll boycott the Hugos!!!”, but we all know we’ll just read another article the following year on the same subject. If you want to change it then join up, make your voice heard and vote for what you really believe were the best works, not just to fill a quota
Says you. I’ve just worked out how to ensure Ted Cruz isn’t elected President in 2016. All I need is to find the right Jurassic butterfly to step on.
I mean, obviously you can change the past. The past exists in records, in physical evidence, and in people’s minds. If you change or destroy those, you change the past.
Or by adding new context. A positive example: I read James White’s “Nuisance Value” differently once I discovered where he was from.
EdgarSF @@@@@ 25:
Thats a really nice graph! And it makes it clear how “noisy” the year-to-year values are.
It makes me think that small-number effects are probably pretty important. One could argue that there are only a small number of writers regularly capable of Hugo-nominatable work, and they’re not going to be producing their best work at a constant rate, so the results could be a lot more variable than the naive statistical model would predict.
James @@@@@ 31:
I don’t see why the same explanations used to normalize the situation as it was in, say, 2007 should not now apply to the current situation.
I was thinking more in terms of typical results from, say, the 1950s and 1960s. (The long strings of years favoring men versus the short strings of years favoring women in your listings tend to suggest historical bias is the main point.)
But you’re right that 2007 was basically just as skewed (in the opposite direction) as 2019.
I have no memory of what discussions might have taken place in 2007 vis-a-vis the gender distribution of nominees. Were there people who complained about that distribution (as opposed to trying to normalize it)? Were they as wrong to do so as you suggest people complaining about the 2019 distribution are?
The statistics here are not exactly defective, but they suffer from a rather basic technical fallacy.
In order to assess whether the process is “skewed” in a particular direction, you need to know the makeup of the underlying population. Here, the real key question is not (simplifying slightly) whether more men than women were nominated for/won the Best Novel award, but whether a novel published by a man had a greater chance of winning/being nominated for the award than a novel published by a woman. If in a given year 95% of the st novels were published by men, a 5/0 split in nominations would not only be unremarkable but statistically likely. In a year where 50% of the novels were published by men, such an outcome would be unlikely, if not quite remarkable. (1 in 32 for a single year is on the statistical significance borderline.)
So the question that should be looked at initially is: In a given time period, what percentage of st novels (or other lengths) were published by men? What percentage of the awards/nominations went to novels published by men? If there isn’t a significant disparity between those figures, then there’s likely little skew/bias in the awards. (Which is not to say that there might not be subtler effects at work; eg, both Heinlein and Norton were less likely to be nominated for genre juveniles.) But ignoring this factor would lead, reductio ad absurdum, to the argument that since no woman has ever been named Super Bowl MVP . . . .
Statistical curiosity: during his lifetime, roughly half the Hugo-nominated fiction written by women was edited/published/purchased by the much-recently-maligned John W. Campbell, a clearly disproportionate share. Does this indicate that Campbell was less biased against female writers than the other editors of his day?
You’re going to make me go through 1950s Galaxy, F&SF, Amazing etc, aren’t you?
What it could reflect is that his biases were such that a story by a woman had to be much, much better than a story by a man for him to consider it, so if it managed to get into Astounding/Analog at all, it was more likely to be Hugo-quality than, say, any given story by male mainstays like Christopher Anvil or John Berryman.
See, for example, the ratio of male winners/male finalists and the ratio of female winners/female winners. For novels, those are
44 male winners/252 male finalists = 17%
22 female winners/76 female finalists = 29%
So women were less likely to make it onto the ballot in the first place but if they got there, almost twice as likely to actually win. Now, that is less surprising if you know “if they got there” often meant “if they happened to be Ursula Le Guin or Lois McMaster Bujold.”
A minor mystery I have not solved yet: in the 1970s I was very aware that SF works by women were high grade ore for the sort of material I like. Having gone to the trouble of checking magazine and best of TOCs, I have no idea where I was encountering them; time and multiple concussions have burnished off the edges of my memory.
You’re misreading my comment. Roughly half the Hugo-nominated fiction written by women was published by Campbell (in Analog). All you need to look at is the Hugo-nominated fiction published by women. (He also published Leigh Brackett’s debut, if I remember correctly.) The point is simply that he was much more likely to publish high-quality sf by women, as messed by Hugo recognition, than any other editor in the field. As measured by raw numbers.
As it happens I know his stats from the end of his career: four out of 71 stories were by women in 1972 (Bova then slowly increased the number of stories by women over his tenure to a peak of 13.5 of 74 in 1978). By comparison in 1972, Galaxy had 2 stories by women of 36 total, and Worlds of If had 4 of 37. Apparently I never bothered to look at F&SF….
Well….
A quick skim says F&F had 84 stories and poems in 1972, 15 of which were by women. 2 were by Tiptree so Ferman would have thought he was buying from a man.
I sincerely hope that all awards are based purely on merit and are not determined by gender, creed, race or anything else. Just quality of the work.
And I’m not exactly lurking …
I have to second what ‘God of Thunder’ has said. In a truly equal world we wouldn’t even be having this conversation, because gender, race, religion, would not matter.
If it were truly a competition between equals, based on the merit of the work, none of us would be concerned with ensuring we had an equivalent number of competitors from each gender. We wouldn’t have a best male category, we wouldn’t have a best female category, we would simply have a ‘best drama’, or ‘best short story’.
In the end gender shouldn’t matter.
I did a gender analysis of Science Fiction & Fantasy Poetry Association Rhysling Award winners from its inception in 1978 through 2017. There has been a similar shift from predominantly male nominees and winners to female. For the decade 1978–1987, 20% of the winners and 24% of the nominees were women; for the decade 2008–2017, 82% of the winners and 58% of the nominees were women, with the ratio changing gradually but consistently over the intervening 30 years. For the entire span 1978–2017, 36% of the winners and 48% of the nominees were women. The winner and nominees are linked to at http://www.sfpoetry.com/ra/rhysarchive.html; in 2018, 5 of the 6 winners were women (the male in question was Neil Gaiman), and in 2019, 6 of 7 winners (including one tie) were women. It’s an interesting trend.
I too noted this, didn’t think much of it, & didn’t know it was a controversy, but also wasn’t paying much attention (there’s a lot to get worked up over in world happenings this year, Hugo folderol just didn’t make the cut this year in my mind space).
One element not mentioned: the Hugo is a fan based award, there is no curating, expert committee, procedure for balance or reorientation or even defining what merit is or is not, it is, in theory what fans like and/or simply what they want to see. There’s room in the world for both fan and blue ribbon committee awards. So, if – to name the elephant on the table – this year was some kind of nefarious backlash woman biased block voting- it is surely what fans want. End of story. Most importantly word in that previous sentence: IF.
Now, I must say – this was one year – wake me up if this happens 8 years in a row, then, maybe, just maybe, I’ll Spock-raise one eyebrow. Now, it’s just one year- chill.
Instead, my complaint this year is the rather poor choices offered to us on the short list, in the shorter fiction categories… but that’s just one fan’s — minority of one — opinion. Shrug.
Just to be clear, I don’t think there was anything wrong with this year’s selection or results.
Having served on award juries, there’s the exact same potential for statistically interesting results for those too. Maybe more, because you’re dealing with the tastes of much smaller groups of people.
I assume some of the winners this year are good writers. Sadly, identity politics cheapens the win for those who actually deserved it.
Weird how identity politics only cheapens awards when it’s women or POC winning but never when its SWM.
I prefer Jerry Seinfeld’s take on awards:
“You don’t give awards to comedians”. First of all, comedians don’t need awards, awards are for people that are looking for work, we’re not looking for work. If you’re any good as a comedian, you’ve got tons of work. We’ve all got wrinkled suits and smelly shirts from packing and unpacking and schlepping all over the goddamn country doing 10 million different kinds of gigs.
And secondly and even more important is your whole career as a comedian is about making fun of pretentious, high minded, self-congratulatory B.S. events like this one. The whole feeling in this room of reverence and honoring is the exact opposite of everything I have wanted my life to be about. I – I – I really don’t want to be up here. I want to be in the back over there – somewhere over there saying something funny to somebody about what a crook this whole thing is.
And I don’t want to give you the wrong impression. I don’t want you to think that I’m not honored by this, because I’m, I feel very, very honored, and it’s – but it’s just that awards are stupid. Every real estate office has some framed, five-diamond president’s award thing by the desk, every hotel check-in has some gold circle service thing; every car salesman is a platinum jubilee winner. It’s all a big jerk off. It is, the hotel sucks, the real estate person is stupid, and the only thing the car salesman is good at is ripping you off.
And didja ever notice that the people who say “X shouldn’t matter” in a discussion about current differences about X more often than not are in the demographic that has benefited from the difference in X?
@52, yeah, good for Jerry Seinfeld. Oddly enough, he did not make a fuss about the 10 Emmys, 3 Golden Globes, 3 SAGs, and 3 Peoples Choice Awards.
And it is nice to receive recognition for your work. Doing the work in order to receive an award, yeah, cheesy, but being recognized for doing it really well? Nothing wrong with that at all.
I’ve participated in Hugo nominations for a number of years, and always posted the books and stories I liked best, without any real notice as to the possible identities of the writers. I don’t remember any gender or race quotas being imposed, but I probably read slightly more books by women than men, so my nominations might reflect that–not greatly because many of my favorite writers are guys. Since I particularly enjoy action and adventure in my scifi and fantasy reading, some books by men are always on my nomination form. I do try to objectively select for quality in writing and never nominate or vote for anything I haven’t actually read.
44 male winners/252 male finalists = 17%
22 female winners/76 female finalists = 29%
You would get cleaner results for this particular stat if you threw out years where the ballot is all-male or all-female because those won’t tell you anything about the relative win rates.
The difference should be much starker if you only look at mixed-gender ballots.