Background, for those who aren't aware: The Hugo Awards are nominated and voted on by the members of WSFS, the World Science Fiction Society, which is typically mostly the people attending the annual Worldcon, but which also includes people who spend $50 on a Supporting Membership. The number of Supporting Memberships has grown over recent years, since publishers began allowing the Worldcon to distribute voting packets with electronic copies of nominated works. (Not all works end up in the packet, but most do.) Each member can nominate up to five works in each category, and there are five finalists.
In 2014, a group called the Sad Puppies, led by someone who was angry that he hadn't been nominated for a Hugo after having been nominated for Best New Writer and losing, tried to push the idea of slate voting, where they'd come up with a list of nominees and encourage their followers to all vote for them. This effort had limited success, it was widely condemned, and most or all of the slated works (deservedly) finished below No Award.
In 2015, they tried again, this time dividing into Sad Puppies and Rabid Puppies (the latter led by an open white supremacist who thinks SF awards pander to woman and minorities, by sometimes giving them awards), pushing a nomination slate of five works in most categories, and recruiting enough followers that the slate swept most categories, leaving no non-slate nominees in those categories. Buying memberships just to vote for a slate of specific works is an obvious attack on the Hugos, and it's been tried a few times before, but nobody has ever tried to knock all legitimate nominees off the ballot before. Social pressure shamed a number of Puppy nominees into withdrawing (and, to be fair, a number withdrew of their own volition and didn't need to be shamed), but there were still eight categories with only Puppy nominees. No Award placed first in the final voting in all eight of those categories.
There was a lot of discussion after the nominations came out, trying to figure out what to do about this. My immediate suggestion was that limiting the number of nominations each person could make to four, and increasing the number of finalists to six, would at least give us a choice of two non-slate works in each category. The obvious flaw in this analysis is that if there are two slates (as there in fact were) they could still sweep the ballot. Still, it would at least make it a bit harder for them to dominate, and when someone else on the forum said he wanted to submit it as a formal amendment I agreed to co-sponsor it.
He also wanted to get rid of a rule that had been causing problems in the short story category, which said that finalists had to appear on at least 5% of nominating ballots. In 2011, 2013, and 2014 this had led to just three or four short stories qualifying as finalists. He wanted to combine both proposals into a single amendment, but I convinced him to make them separate amendments.
In the months between this and the convention, a more elaborate proposal called E Pluribus Hugo was developed and submitted as an amendment. This changed the counting procedure in a clever way to dilute the power of slates. Basically, each voter got one point in each category, divided among all their nominations. Everything that was nominated would be ranked by points, the two works with the fewest points would be compared, the one that appeared on the smaller number of nominating ballots would be eliminated (even if it had more points), and the points for all the works would be re-computed and the process repeated. The idea is that with normal voting, the final ballot would be the five works that had the most nominations, just like the old system; but if there was a slate, then they would have relatively few points compared to non-slate works, eventually two works on the slate would have the least amount of points, and one would get knocked off the ballot. The hope was that this algorithm would knock at least two or three slate works off the ballot in the more popular categories, and at least one or two in the low-turnout categories, giving voters a real choice. I thought this was a superior proposal to 4 and 6, but as I was already co-sponsor of a rival proposal I elected not to co-sponsor EPH as well.
At the Worldcon in Spokane last year, all three of these amendments received initial passage. They needed to be ratified at this year's Worldcon to take effect next year. There is no way to change the rules any faster than that.
Note that EPH and 4 and 6 are compatible with each other, so there's no reason both couldn't pass. However, if EPH is in effect, it has a perverse effect on 4 and 6: the 6 part is good, since the extra finalist will usually be a non-slate work, but the 4 part hurts regular voters more than it hurts slates. There was a non-binding ruling at the Business Meeting that if 4 and 6 were changed to either 4 and 5 or 5 and 6, then that would be what is called a "lesser change" (basically, closer to the current version of the Constitution), and it wouldn't delay ratification. If either number were moved further away from 5, though, such as 3 and 6 or 4 and 7, that would be a "greater change", and the amendment would have to be passed on to the next Worldcon, which could ratify it.
This year, the Sad Puppies effectively ceased to be a force that mattered, and the Rabid Puppies decided to nominate a mix of works that would likely have made the ballot anyway (such as a Lois McMaster Bujold novella, and a Neil Gaiman graphic novel) and works that were designed to be simply offensive (including a few that were basically mean-spirited invective directed at SF personalities they despise). An amendment called Three-Stage Voting was introduced to deal with the latter case: it would have an intermediate stage in which the voters could consider a long-list of the top 15 nominees in each category, and vote to toss any of them with a super-majority consisting of 60% of the people voting on that item, including at least 600 people voting to reject. The hope is that anything that would finish below No Award on the final ballot anyway would get thrown out, and that the offensive garbage would be replaced with worthy nominees. (I'm skeptical that this will be as effective as its proponents think, but I don't think it will do any harm. At worst it's a waste of time, but it shouldn't be actively pernicious.)
There was also a refinement of EPH proposed, called EPH+, which changes the divisors EPH uses to make it more effective: if you have two nominations remaining in a category, each gets 1/3 of a point instead of 1/2, if you have three each gets 1/5 of a point instead of 1/3, with four each gets 1/7 instead of 1/4, etc. Theoretically, this should make EPH more effective.
There were a few other measures proposed to fight slate voting, but they either failed or were small changes, so I won't discuss them.
Whew. Okay, enough background, on to the meeting.
The WSFS Business Meeting was scheduled for four three-hour sessions. Most years, we run out of business before the fourth session and it isn't held, but that's also the only session after the Hugo Award ceremony, which is when the nominating data for the Hugos is released. Pretty much everyone agreed that it would be best to consider EPH, 4 and 6, and EPH+ on Sunday.
The night before the first session, I run into David Wallace, one of the co-sponsors of EPH. He mentions an amendment that Kevin Standlee, one of the sponsors of 3SV and a major power in WSFS, wants to make which would require re-ratification every year between now and 2022. (There's already a sunset clause requiring re-ratification in 2022. I don't think there's every been a sunset clause requiring more than one re-ratification.) We're both really strongly against making it that easy to get rid of EPH; attendance by reformers at the Business Meeting has been high the last couple of years, but people are going to get tired of going to Business Meetings, and it would be much easier for die-hard opponents to pack a meeting and win one vote than to win two consecutive votes. I mention a counter-proposal I'd thought up, which would allow a Business Meeting to suspend EPH for the following year, which would still allow the Business Meeting to get rid of EPH immediately if they wanted, but would mean that they'd still need two years to get rid of it permanently. David likes the idea, and plans to talk to Jameson Quinn (one of the principal movers behind EPH) about it, and see if Kevin will consider that proposal instead. (I hear later that Kevin says he's too fried to consider the merits of our proposal, and it would be out of order to try to amend his amendment, but it would be in order to suggest it as an alternative during debate.)
The first session is devoted to setting the agenda for the remaining meetings, including the order that items were considered and the debate time for each item. It's also possible to kill items without debate with a 2/3 supermajority. 3SV survived one of those motions, but another Hugo reform proposal was killed. My main objective for the preliminary Business Meeting was to make sure that 4 and 6 was considered after EPH, since if EPH passed I wanted to amend 4 and 6 to make it 5 and 6. I had e-mailed the lead sponsor of EPH well in advance of the Worldcon to make sure he supported considering EPH first, then e-mailed the chair of the Business Meeting to make sure the two items were initially put on the agenda in that order. However, someone at the meeting did move to re-order the agenda to put 4 and 6 before EPH, so I had to make a quick argument against. Basically, I said you could sensibly consider EPH without knowing whether or not 4 and 6 would pass, but the reverse wasn't true, and I said that if EPH passed I intended to propose an amendment changing it to 5 and 6. The attempt to put 4 and 6 before EPH failed, and I don't think it hurt any to put the idea of 5 and 6 into people's heads.
An incredible amount of time is spent voting on how long to debate each amendment. Much, much more time is spent voting on debate times than is saved by shortening the chair's suggested debate times.
The 5% Solution came up for ratification on the second day of the Business Meeting. I'd prepared pretty thoroughly for my speech in favor. I looked at all of the short story nomination longlists for 2007 through 2015, plus the longlists for 1980 (when the 5% rule was adopted by WSFS) and 1984, looking not just at how many works there were above the 5% cutoff, but where each of the stories was published. Turns out that in 1980 and 1984 three-quarters of the stories on the longlist appeared in just two or three magazines. Not the same magazines, but both years a small number of magazines dominated. From about 2010 onward, there was occasionally one magazine (Clarkesworld) that had several entries on the longlist, but basically nominations were much more widely dispersed among markets than in the 1980's. And that was basically my statement: when this rule was put in place, the short story market was smaller and everyone was basically reading the same stuff. Now there's a lot more good fiction being published, nobody can read it all, and it's become much harder for even outstanding work to clear the 5% bar.
If I remember correctly, the only real argument against The 5% Solution came from Ben Yalow, who had in 1980 pushed to adopt the rule in the first place. His argument was basically, well, it may be a good idea in theory even if it's not a good idea in practice, and besides, how can we be sure it's not a good idea in practice, maybe the stuff that's getting eliminated should be eliminated. Not very compelling, but I wish I hadn't cut one detail from my speech for time: In 2014, the fourth-place short story had 1/4 of a nomination less than 5% of the ballots cast. If the Hugo administrators hadn't rounded the cutoff down to a whole number, it would not have been a finalist, and if it hadn't been a finalist, it obviously would not have gone on to win the Hugo, as it did.
There's an effort to add a sunset clause, requiring re-ratification in three years. It fails.
The 5% Solution passes easily.
There is a huge time-consuming debate over whether there is a parliamentarily permissible method for postponing a ratification vote for one way, i.e., taking something that's up for ratification at this year's Worldcon and vote to ratify it at next year's Worldcon instead. Various methods are suggested, and eventually the Nitpicking and Flyspecking Committee (which is a real, highly respected standing committee) is instructed to report back the next day. Have I mentioned how much time is wasted on pointless wrangling over rules at these meetings?
Day 3, none of my amendments were up for discussion, but I attended anyway. The Nitpicking and Flyspecking Committee decides that the preferred method for delaying ratification by one year is to add a sunrise clause delaying its effect to the year after next, and a sunset clause repealing the amendment if not ratified at the following Worldcon. Sunset clauses have commonly been used, but sunrise clauses haven't ever been put in the Constitution. This report will have unforeseen consequences, as various people attempt to add the hitherto unprecedented "sunrise clause" to various amendments.
A terrible, but badly written, amendment was proposed to 3SV. It was referred to committee to clean up the wording of the amendment, to report back on Sunday. The rest of the amendments were disposed of one way or the other, leaving us with four items of business to deal with on Sunday. Unlike the first three meetings, the Sunday meeting could continue as long as necessary past the three hours in the schedule, although it would be difficult to stay past 2:30 since one of the presiding officers had a plane to catch.
Of note: an amendment defining North America received first passage, which explicitly states that Washington, DC is part of North America. If the amendment is ratified next year, a Worldcon held in my apartment will not trigger a NASFiC.
Later that day, I learn that a report showing what effect EPH would have had on the 2014 and 2015 results is up on the Business Meeting page of MidAmeriCon's website. I go over it before going to bed, noting that 5 and 6 would put Puppy works that EPH eliminated back on the ballot with distressing frequency. Still, 75% of the time it adds a non-slate work. EPH itself adds an average of about one non-slate work per category, which is a smaller effect than people had hoped for.
Sunday. The terrible amendment to 3SV is defeated. A pointless amendment, which would literally have had no effect whatsoever, is withdrawn after that fact is drummed into the maker's head. After much debate, 3SV receives first passage. Still needs to be ratified next year, and I'm not sure it will be.
EPH comes up. Jameson proposes, at the request of this year's Hugo administrator, a technical change regarding how to handle finalists that are withdrawn from the ballot by the authors. It passes, after the chair rules that it's a lesser change, and an appeal of the chair's ruling is defeated. Kevin Standlee makes his motion to add the multi-year sunset clause; David, during debate, describes the alternative I suggested and says he'll move it as an amendment if the sunrise clause is defeated; the sunrise clause fails, with 96 votes in favor, 99 votes against. David proposes the suspension clause, it passes.
There's a report showing what effect EPH would have on the 2016 nominations, for both the 2016 Hugos and the 1941 Retro-Hugos, but only five paper copies. They're working on getting it up on the website. I find the links on the website, but both links go to the report for 1941. Fortunately, at some point while the Hugo administrator is answering questions, they get the link fixed, and I manage to pull the 2016 report up on my cell phone. 5 and 6 puts some really, really awful stuff back on the ballot, but 80% of the time it adds a non-slate work. This is data I need.
EPH passes. A resolution to request that the Hugo administrators report next year to the Business Meeting on the effect of EPH passes. (The Business Meeting doesn't have the power to require that the Hugo administrators do anything.)
On to 4 and 6. Before I can speak, someone makes a motion to add a suspension clause to 4 and 6, like the one in EPH. I think it's unnecessary, but it passes pretty easily. I don't think it will ever be invoked, though. As maker of the motion, I am then given priority to speak in favor of the amendment.
I begin by making a Point of Parliamentary Inquiry, asking if changing 4 to 5 would be a lesser or greater change. On being informed that it's a lesser change, I so move, and am immediately and enthusiastically seconded.
I begin my speech by comparing 5 and 6 to 5 and 5, the status quo. I argue that the change can't reduce the number of non-slate works on the ballot, and, based on the 2015/2016 data, it will add a non-slate work 75-80% of the time. 20-25% of the time it will put a slate work that EPH got rid of back on the ballot, but I argue that the benefit of giving the voters more non-slate options outweighs the harm of adding a few more slate works to the ballot. Combined with EPH, therefore, 5 and 6 is an improvement on the status quo, in terms of fighting slates. I give a condensed argument that 5 and 6 is better under EPH than 4 and 6, referring to some charts Jameson had prepared for the agenda comparing various systems, which I hope people will either follow, or take on faith. Finally, I say when changing the Constitution, we should also consider the effect that the amendment will have when the slate voters finally go aware, and I argue that with the increasing diversity of the field—stylistically, geographically, in terms of content and subgenres—moving to six nominees allows us to better reflect the breadth of what's being published, and a modest expansion in the number of finalists is a good thing not just in terms of fighting slates, but for its own sake.
I've used up just about all the debate time in favor of the amendment. There are several speeches against, during which I am reminded of the dictum "Debate need not be factual" (or sensible). The vote is called, and after a show of hands the chair rules that the amendment is ... defeated. Ooof. I honestly don't understand why anyone is voting against this. (Against the underlying amendment, I would understand, but why does anyone prefer 4 and 6 to 5 and 6?)
There's a motion for a serpentine division, which is a precise count in which those voting on each side count off in sequence. 85 votes in favor. I note, when the count against reaches me, that it's running slightly behind the yes vote. Votes against peter out at 61. The chair was wrong. My amendment to change from 4 and 6 to 5 and 6 passes.
I'm asked if I want to speak in favor of the underlying motion, and decline. Nobody wants to make any further arguments against the amendment either. We vote, and it passes.
Someone tries to move to adjourn immediately, without considering EPH+. The chair rules that this is out of order; since EPH+ was moved specifically to this point in the agenda, we can't adjourn until we deal with it. He then moves to suspend the rules and adjourn immediately. He needs a 2/3 majority to prevail, and there is at least 2/3 against him. Everyone is exhausted. After a bare minimum of debate, someone moves to call the question, the meeting votes to end debate, we vote, EPH+ passes. A motion to adjourn is passed enthusiastically. The Business Meeting is over. It took almost exactly three hours, which largely by coincidence is the time it was scheduled for.
Side note. I didn't support these amendments in the expectation of glory—in fact, when I agreed to co-sponsor them, I didn't think I would even need to speak at the Business Meeting, but it turned out that the lead sponsor was an indifferent orator and was agreeable to letting me make the opening arguments—but after the meeting, several people came up to me, thanked me, and congratulated me on my success. I think I'm better known in the Worldcon community than I was, and I have gained a significant amount of respect. Which is probably the way things ought to work, but it did take me a bit by surprise.
I was gratified that everything I sponsored passed in pretty much the form I wanted, and glad to have made a suggestion that helped EPH.