Like most readers, I really hate it when my favorite character dies, whether it’s in mid-story or right at the end (though the longer the character has been around, the more I’m invested and the more I hate losing them). But there are some stories that I love in spite of the fact that my favorite died, and others where the author kills off a character and my reaction is, “Oh, just no, nope, I’m done.”
This past week, I’ve read a couple of interviews with writers who discussed why they’ve killed off a likeable reader-favorite character in the course of their story. One of the most common reasons was the one-scene viewpoint, where the likeable character is only present in the story as a victim viewpoint, to “show” the reader the horror of being attacked by a serial killer or a werewolf, or to provide a first-hand look at the Evil Overlord’s destruction of a town. This almost always makes me want to pitch the book across the room, because it is almost never truly necessary and it feels to me like cheap manipulation. The only saving grace is that it’s only one scene, so I’ll keep reading for a while if the author is good enough. If the author does it a second time, though, I’m gone. Even if it’s in a different book.
A couple of writers said they killed off likeable characters deliberately, to demonstrate to the reader that nobody was safe; that anybody could die at any moment. For me, that’s simply an advanced form of the one-shot victim-viewpoint, because in most of these cases, the likeable character was only present in the story in order to be killed off and shake up the reader. None of the characters who had a real function in the story died, only the ones whose “function” was, like the serial-killer victim, to demonstrate that death and violence could happen anywhere, to anybody.
The problem with using character deaths to “prove” that a serial killer is dangerous/horrible or that nobody is safe is that the writer doesn’t actually need to kill off a character on-stage to do either of these things…and in fact it is frequently more effective not to kill a character directly on-stage. As evidence, I offer the first chapter or so of Thomas Harris’ Red Dragon, in which two detectives are looking at evidence collected from a crime scene. By the time I finished that chapter, I wanted to get up and make sure all my doors and windows were locked (and it was the middle of the afternoon!)…and not only were the killings not onstage and not of any characters we’d met, the stuff the detectives were looking at was never described in detail.
And then there are the writers who kill off characters in the final battle “because it isn’t plausible for all of the main characters to come through something like that alive.” It’s funny how many of these are one-half of a couple who just got engaged the night before…
One guy actually bragged about rolling dice to decide which characters would die at certain points in the story “because I wanted it to feel real, and real-life violence is random.” There are a bunch of things wrong with that statement, but I’ll stick to the writing-related ones, the biggest of which is that fiction is a model of real life, not a newspaper story. Randomly killing off characters doesn’t necessarily make a story feel realistic; in fact, it almost always has the opposite effect on me. It makes me think the writer doesn’t know what he/she is doing.
Another couple of writers killed off characters near the ends of their stories, not because they’d planned to all along, but because they suddenly realized that by doing so they could provide a situational parallel to the opening of the story, or because they decided their protagonist needed that final revenge motivation or angst. Sometimes this does work for me – but only when it doesn’t feel forced.
What does work for me are the character deaths that have an actual story-related reason behind them. The first sort are the ones where the character earns it – that is, the character’s death comes about because they did something that is totally in character, something that they couldn’t not-do without turning to cardboard or becoming someone else. Greg pushes his beloved brother out of the way of the deadly curse; Janice runs back into the burning building to rescue her father; John and Annabel make the suicide run to destroy the incoming asteroid and save the world.
The second sort that works for me are the “random events” that have significant story-related consequences. By that I mean more than “Susan got hit by a truck and everybody grieves;” I mean that the seemingly random event has ongoing, significant ripples through the plot and characters. Because Susan dies, Sam goes out and gets drunk and spills important plot-beans, Meghan is at the funeral home making arrangements instead of being at home when the letter bomb explodes, and Joanna takes over Susan’s job and finds the heroin hidden in her office supplies, none of which would otherwise have happened. The death of Christian in Dorothy Dunnett’s The Game of Kings is an example of this type; it is shocking, random, complicates the plot wonderfully, and provides for a bunch of characterization and motivation for other characters. And it’s the result of a bunch of different factors that all make story-sense, no one or two of which would have resulted in her death, but all of which together made it happen.
The telling phrase, for me, is when a writer says, “I didn’t want to kill off X, but I had to” – in other words, the story required it. By “the story required it” I do not mean that the plot has Jennifer dying in Chapter 13, so off she goes – plots can and do change radically as things move along, and forcing any plot development to happen simply because one decided on it six months and 60,000 words ago is usually a mistake. I mean that no matter how shocking or unexpected the death, something in the story will not work without it. I still hate it when a character I love dies, but I don’t feel manipulated and I love the stories in spite of it all.
The “single-scene redshirt POV” character getting killed doesn’t bother me quite so much. Part of it, I suspect, is that I pick up on the “you are not supposed to get attached to this character” cues. So I read it as a genre or sub-genre convention.
The number of additional characters killed in the background also makes a difference to me. If the redshirt is the only one who gets killed, or is one of only a handful, it’s more likely to bug me than if the redshirt is shown dying in an attack that kills dozens or hundreds or even more. I don’t know why, but there it is.
Having more-developed characters die “to show that anyone can die at any time” does annoy me; I see that as something worse than just an advanced version of the one-scene victim-viewpoint. Likewise random death “because death is random,” or major characters dying in the final battle because “it’s implausible for them all to survive.” Bleah.
I suspect these writers underestimate just how heavily they’re salting their stories with death. I’ve see something similar in tabletop roleplaying games, where the GM doesn’t seem to recognize just how deadly the game system or his setting is. This may be a parallel development, or an example of cross-fertilization from RPGs to written stories. Or both.
I’ve taken a break from trying to GM because players always seem to die when I do. It got to the point where I had someone else setting up the battles and even then players would die.
Admittedly, it was occasionally funny. I once had someone who’d tweaked their character into a resistant-to-everything vampire. The party ran into corpse worms. Everyone else took a few points of insignificant damage. The vampire died in two rounds.
One of the things I don’t like so much about Scott Lynch’s otherwise excellent Locke Lamora books is his tendency to kill people off. Mostly it does make sense for the plot, but it can get to be a bit much.
When I met him at Fourth Street Fantasy, I informed him that if he ever kills Jean Tannen (my favorite character), I’m dropping the series right there. He said, “I can neither confirm nor deny that. It’s an uncertain world.” :-p
When it comes to inventing three-dimensional, sympathetic, even sometimes likeable characters, and then killing them horribly off, there’s nobody to beat George R. R. Martin. I don’t read him any more.
That’s the main reason I never started reading him in the first place!
David Weber’s series on Honor Harrington kills off a LOT of people – but does so in a very realistic story about war and politics. Makes the political maneuvering so much more intense when you know a bunch of people can and will die if someone stomps off in a huff. Ditto the battle scenes when you know that the missiles hitting the ships WILL kill off people you got to know in the book.
The most recent book managed to annoy me thoroughly with a couple of deaths – but mostly because other plot issues. (It was a ton of short stories taking place simultaneously – so he chopped them up into chronological order and called it a novel.) I spent so much time feeling lost and confused that the random deaths annoyed me far more then normal.
I was pantsing a story at one point and had a character unintentionally die–50+ foot fall while sneaking into the main bad guy’s lair. It bothered me a lot because the death, while a logical result of actions, was fairly meaningless. I eventually went back and saved the character by having him captured. I did need the justification that he doesn’t have to die if he ends up in a much worse situation than death.
I don’t like unnecessary death. If the story would work without it, I feel like the author should try to make it work. But hey, I’ve never tried writing a story that required death in any way…I think the key phrase is “necessary”.
The death that bothered me the most was in one of G. G. Kay’s novels. The character’s demise was logical, meaningful, and had repercussions through the plotline; there was even a period of morning by the other principals that tugged appropriately at my heart strings.
Then the character was inexplicably brought back into the story, having managed to save herself off camera. It felt cheap and clunky and wholely unnecessary, and as a reader I felt both manipulated and cheated.
A number of people cite this as one of their books of all time, so they clearly did not have the(multiple other) problems with it that I did, so there is that.
George R. R. Martin kills major characters on a regular basis, but is still extremely popular. What is he doing that makes it work?
I had a character walk into my series of Cynthia stories (someday to appear online as _The Witch of Syracuse), as my characters sometimes do. He became Cynthia’s second of three husbands, and I knew as soon as I saw him that he was going to die, since his appearance had caused a plot sequence to bloom that was inevitably going to kill him. But he was a lovely guy and I didn’t wanna. I managed to get one more story out of him en route, and then I got stuck. I tried to think of some plot element that would delay his death, and I couldn’t, and finally my husband said, “You’d better kill him and get it over with.” So I did, and his death motivated Cynthia for the whole rest of the cycle, till the last story in which I stole from the Odyssey, the Aeneid, the Pearl, The Empire Strikes Back, and Dante’s Purgatorio, and I read it out at Greyhaven and had Diana Paxson in tears/ Sometimes you just have to kill them.