Now that we've done the history of "Spoiler Alert," let's discuss appropriate and/or civically obligatory uses.
The (previously, of course) definitive guide to spoiler alert usage was written by Awl contributor Dan Kois for New York magazine's Vulture blog in 2008. The whole guide is worth reading, as is the accompanying manifesto calling for a return to a "water cooler culture," in which people who really care about a show or book or movie make an effort to read or watch it as soon as they can, so they can then discuss it with their co-workers (or whomever) in person.
Spoilers, in this account, are allowed after a brief but reasonable interval that allows for anyone who truly cares to watch or read the show or movie or book in question. Spoilers are allowed in the text of articles more quickly than in headlines, because people can simply choose not to read articles for a few days to avoid spoilers, if they must, but it can be hard to miss a headline.
Furthermore, the rules vary for different media: you should give people a few extra hours to watch a TV show, a few days to see a movie, and a few months to read a book. Reality shows can be spoiled immediately upon conclusion, as they are essentially sporting events. Operas are never, under any circumstances, to be spoiled. (I'm pretty sure that last one is a joke.)
All of which is well and good and probably necessary to lay out on a blog such as Vulture where people are writing about shows and movies and books and the reactions thereto several times every weekday. And the accompanying manifesto really is terrific; you should read it (after you finish this; or at least come right back).
But the guidelines, I regret to say, are flawed. They are both too severe and not severe enough. That is because they ignore a crucial factor: artistic ambition.
That probably sounds snobbish, and I suspect (though Dan is welcome to correct me) that an aversion to snobbishness is at least partly responsible for this oversight-just as an assumption of snobbishness is probably the reason Ron Rosenbaum also failed to make this distinction (in his own pro-spoiler blog post from 2006, cited by Kois in his manifesto). Put simply, a truly ambitious and successful work of narrative art is spoiler-proof. If a show or movie or book is really, truly great, you can watch it again and again and again, well after you know what's going to happen, and the aesthetic pleasure you derive therefrom will not diminish. It may even increase. This is an essential part of the work's greatness.
Consider this: Alfred Hitchcock knew as much about creating suspense as perhaps any narrative artist of the past century; and when he made what is, hands down, his most artistically ambitious movie, Vertigo, he went out of his way to spoil the mystery halfway through. Vertigo is the story of one woman pretending to be another in an effort to deceive a man, and Hitchcock easily could have preserved the mystery of that woman's identity until the end of the film.
But the pleasures and satisfactions of Vertigo don't depend on not knowing a basic aspect of the plot. They derive from the movie's brilliant illustration of love and desire and the ways we idealize and romanticize particular human beings and then become disappointed or even disgusted by their simple, physical humanity. It's the best thing Hitchcock ever did, and knowing who is actually who doesn't change that.
On the other hand you have The Usual Suspects, which, after you have learned the identity of Keyser Soze, really isn't very good.
(By the way: Hitchcock's deliberate avoidance of narrative suspense in Vertigo is one of the reasons it is better than the truly excellent but not greatest-ever-made film Citizen Kane-no matter what some fancy poll says-which employs the narrative crutch of withheld knowledge and then bestows that knowledge in a corny and not very satisfactory way at the end.)
So if you're discussing something like The Usual Suspects, you should not try not to reveal the ending unless absolutely necessary (even now), and, if you must, a warning is in order. If you're talking about, say, Eternal Sunshine of the Spotless Mind, the rules are different. It's actually more okay to spoil something the better it is-and this rule comes with good cultural consequences: should we really spend that much time talking about The Usual Suspects? Let's talk about the things that can't be spoiled, no matter how many plot points we give away.
Your own artistic ambitions as a critic are relevant here, too: if you're writing a blog post you consider more or less ephemeral, then, what, you can't be bothered to throw in the silly but really rather simple phrase "spoiler alert"? But if you're aiming for something more lasting, then yeah, it's not really fair for readers to get mad at you for not using what is honestly kind of an embarrassing cliche. Editors can follow this principle as well: is your publication for the next three days or "forever"? Edit accordingly. Readers could then approach a publication with the appropriate degree of caution.
With that established, the other question: Why is imperviousness to spoilers an essential aspect of truly great narrative art? I'm not really sure. I have a theory, though, one that is at present about quarter-baked at best and will probably sound even more pretentious than everything I've written so far (which, considering the repeated use of the phrase "narrative art" and the appearance of both "thereto"and "therefrom" is, I imagine, saying something). I think it has to do with life and death and the way the former leads inevitably to the latter. That is: Life is not a mystery. We know how it ends. And if a work of art can be "spoiled" when we know the ending, it can't really have that much to say about life, can it?
David Haglund is the managing editor of PEN America and has recently written for Bookforum, Slate and The National. He is on the board of the National Book Critics Circle.