WARNING: This post contains major spoilers for Hugh Howey’s Silo series and Machine Learning.
Hugh Howey’s Silo series started as a series of short stories which grew as fans fell in love with it. The series started in 2011 with the short story “Wool”, which was later published together with four sequel novellas as a novel with the same name. Along with Wool, the series consists of Shift, Dust, three short stories, and Wool: The Graphic Novel.
The Silo series is set in a post-apocalyptic world where humanity clings to survival in the Silo, a self-sustaining subterranean city with 144 floors. No records of the time before the Silo remain. All residents of the Silo are taught that the outside world is toxic and deadly, and the Silo’s cardinal rule is that anyone who expresses a desire to go outside must be sent there to clean the external sensors with a wool cloth. Those sent outdoors invariably clean the sensors as instructed but die within minutes, reaffirming to the Silo residents that the outside is uninhabitable.
In this post, I will examine the premise of the three short stories within the anthology titled Machine Learning set in the silo universe. “In the Air”, “In the Mountains”, and “In the Woods”. Juliette is killed off at the end of “In the Woods”. In the afterward, Howey makes no apology for killing off the much-loved character. His reasoning seems to be that he is just being real. So what is the big deal about that? Early in Wool, we see four main characters killed off (Holston, Allison, Jahns and Marnes). We don’t feel cheated when that happens. It adds to the tension. We don’t know who is safe for the rest of the series. And the main hook is the mystery of the Silo. Why is it there? Who built it? What is outside? Why would anyone want to leave it when the outside is clearly not safe? So we keep reading.
Yet with the three short stories in Machine Learning, there is not the same payoff, and the killing of Juliette is disappointing. I’m not saying that Howey should not have killed off Juliette. That was not what was disappointing to me. Her end could be tragic, just like Holston, Allison, Jahns and Marnes. It is not what happened but the contrived way it happened. When I finished those short stories, I wasn’t left with fascination about the world of the stories or wondering what would happen next. I was left with a feeling of the arbitrary nature of those stories. It just seemed to be a contrived way to kill off Juliette.
Two people from “before” survive in deep freeze in a bunker in Colorado for five hundred years. They feel cheated and angry and want revenge against the people who caused the apocalypse. So they go in search of those who are responsible. They meet members of Silo Seventeen and ask who their leader is. They are taken to meet Juliette and they kill her in cold blood without asking for explanations or giving anyone a chance to explain. They don’t even ask if the people responsible are still alive. That doesn’t feel earned to me. Judging by Howey’s explanation in the afterward, he was expecting some pushback. So that leads me to wonder why the author of the Silo series, who could kill off characters that we had begun to identify with so successfully in Wool, could not have built the plot around Juliette’s death more satisfyingly.
That reminds me of The Terminator movie series. James Cameron built a captivating world that worked well for three movies. The high point was the second movie. An outstanding achievement considering how hard it is to make a sequel that is better than the original movie. I remember wanting to know what the world would look like after the war with AI started. I now wish that I didn’t know. Sometimes a monster in a movie is more frightening when you can’t see it. After Terminator Three, the mystery was gone. I can now see why some directors resent making sequels. I am now glad there never was a Back to the Future Four. I hope the Apple+ TV Silo series doesn’t go for one season too long. That is unless the mystery can be revived and the ending is earned.