veliseraptor:

Man I love it when people ask me totally unsolicited questions! (Seriously, though, thank you. I need these reminders.)

For a few months now (roughly dating back to Infinity War, actually), I’ve been trying to figure out the difference between tragedies and downer endings – one of which I really like and one of which I…don’t. The conclusion I’ve come to is that it is primarily to do with generic mode.

What I mean by “generic mode” is basically the way that works of fiction (or, arguably, works of nonfiction) make use of conventions, tropes, narrative form, or style to set up expectations for a reader or viewer. Sometimes that is done in order to subvert or challenge those expectations in ways that can be jarring but still effective; sometimes it is a more straightforward kind of “shortcut” scene setting. The obvious example is the fairytale formula of “once upon a time” – but most things are more subtle than that. Very often, the generic mode of a work is a large part of what shapes the reaction to that work.

(Incidentally, generic mode is not wholly predetermined – I’d argue that, for instance, A Song of Ice and Fire reads differently to someone who has a lot of experience with the fantasy generic mode and someone who doesn’t – prior reading experience shapes how the text is perceived. But that’s irrelevant.)

What I’m saying comes down to something simple: the difference between a tragedy and a downer ending is what generic mode they’re written in.

Of course the experience setting up expectations in order to subvert them is a common and often very effective trick. It can make for very interesting stories and fresh perspectives. But – and this is the big but – it requires a skilled storyteller to make that pivot work, and, I’d argue, the seeds of that pivot need to be sown well in advance. The tendency – the possibility – needs to be there beforehand.

The thing about a tragedy is that it’s built on a specific framework – from the beginning, the tragedy is already there, inherent in the text even if it isn’t necessarily overt. Part of the tragedy of tragedy is the way in which you can see it coming – that stomach-plunge of recognition of how this is going to go. The narrative framework is vital to that experience, and to the catharsis that can result from a well-executed tragedy. That feeling of good pain. When I think of the sad endings that have worked for me – no matter how miserable they made me – it tends to be about the way the story was constructed to make the end feel inevitable. Maybe not in the moment – I might feel, initially, that’s not fair – and yet looking back, I can see how the pieces were set in place, laying the groundwork for what comes later.

What I don’t like is when the downer ending feels instead like a gotcha! moment – one that says “you expected this to work out, didn’t you, but it didn’t, surprise!” It doesn’t feel like a clever subversion of my expectations; doesn’t feel like pushing of generic boundaries or exploring new ground. It feels like a wrench thrown in the narrative gears – initially, a betrayal, and then just a disappointment. An anticlimax, even.

Maybe this would be different, if the world we lived in was different. Maybe I would feel differently if the constant drumbeat of misery weren’t so all-consuming, and a turn for the worse really did come as a surprise. Maybe I would feel differently if the market weren’t already saturated with stories that seem to take pleasure in setting their reader/viewer up only to knock them down, without understanding the structure that makes a tragedy.

But I don’t.

Basically: it’s not about needing a happy ending. I don’t always need things to be happy. I can enjoy when things aren’t – anyone who has read my writing knows that. (Not that there’s anything wrong with needing a happy ending, but that’s another five paragraph essay.) It’s about narrative structure, and the fact that there is a difference between writing a tragedy, and writing a downer ending. And that difference – if you’ll pardon the repetition – makes all the difference.