My colleagues Ross and Matt are meditating on the mounting pile of discredited faux-memoirs. Ross thinks the publishers need t take more responsibility; Matt asks, why bother? Does it matter if these things are true?

I'm reading Ross talking about another first-person non-fiction narrative that turns out to be B.S. and it's making me think of how a lot of old-school novels involve this pretense to accuracy. Often they'll begin with a narrator telling the "true" story of how he heard the story that makes up the heart of the plot. Or else the manuscript will be discovered somewhere. For reasons that I'm sure are well known to people who were paying more attention in lecture, early audiences seem to have been incapable of digesting something like "this is a story I made up because I thought you would get something out of reading it -- enjoy!" Instead, prose had to be true.

Meanwhile, contemporary fiction is pretty sharply bifurcated between crappy "genre" fiction and literary fiction that often seems very artsy-fartsy. For a well-crafted but basically straightforward story of people doing things and interacting with each other in a moderately realistic way, you need to turn to narrative non-fiction. You can tell people you've just been reading Bill Buford's Heat and hold your head high in sophisticated circles, it's not like copping to owning Tom Clancy's Op-Center: State of Siege.

But if you sold the story as fiction, I think it would be deemed inadequately literary. And yet the facticity of the narrative has nothing to do with anything. Do I actually care if Buford really sliced his finger dicing carrots that one time? Or if Dario the butcher really yelled at some restaurant owner in some other Tuscan town? To me it seems basically irrelevant. The verisimilitude of a lot of the mise en scène really is integral to the book's appeal, but the same could be said about Moby Dick and any number of other straightforwardly fictional works. The literal accuracy of the whole thing, by contrast, contributes very little to the actual work. What it does instead is alter the marketing possibilities and likely critical approaches, opening up space for a certain kind of narrative to be taken seriously. Which isn't to say that people should lie in their memoirs, but maybe there's something to be learned from the fact that there's such an appetite for made-up stories of a certain kind.

Matt has stumbled upon one of the few subjects I can plausibly claim to know well; my long-ago thesis was on the emergence of the novel, and I spent most of my senior year immersed in its early history. Novels didn't have to be sold as true stories, and frequently weren't; it's just that the man many regard as the first true English novelist happens to have adopted that form. Around the turn of the 18th century, the romance, which had previously dominated the market for prose fiction, sort of collapsed as a literary form; people were tired of reading about princes rescuing fair maidens. This, and a growing literate middle class, created a space for people to experiment with new fictional prose forms, combining some of the story-telling devices of a romance with other pre-existing forms, like the travel narrative and the picaresque. Defoe's best known works (out of the hundreds he published) are the novels Robinson Crusoe and Moll Flanders, which adopt many of the travel and picaresque conventions, including proclamations that they are true history--but it's not clear how many people actually believed those proclamations, rather than recognizing that they were a literary device, like Umberto Eco's footnotes in The Name of the Rose. Certainly no one thought that Don Quixote was real.

18th century readers had a more . . . er . . . nuanced relationship to the truth than we do. We separate fiction and non-fiction quite cleanly. They often sort of expected tall tales to creep into history (or real people to creep into fiction--this was the beginning of the rise of the Roman a Clef). Whatever people thought of Robinson Crusoe in the beginning, we know that fairly early in its life people recognized it as a fiction, yet it sold (and continues to sell) rather well. Think of the "True Confession" magazines of the 1940s--or perhaps Penthouse Letters, possibly the last place where this art form survives. Later novelists, like Melville and Dickens, were building on those conventions, which is why so many novels into the twentieth century resort to the device of the newly discovered manuscript or the personal tale retold.

I do think, though, that Matt has hit on something about our own time, though I'm not quite as down on contemporary fiction as he is. Since the modernists, all contemporary literary fiction--including narrative fiction--has focused less on certain aspects of telling a story. I understand that some cognitive scientists theorize that the reason we enjoy stories so much is that they activate the parts of our brain that deal with social cognition and learning. The reason that genre fiction, even though it is usually not a masterpiece of prose styling, can be so absorbing is that it provides this function. The fantasy of a space opera or a bodice-ripper is compelling because we're imagining ourselves as the hero--imagining ourselves as a better, more interesting version of ourselves. We're also exploring how we should/would act in certain (unlikely) situations; the novels that do best in these genres are the ones where the hero ultimately acts rightly, which is to say, producing the best result in some sense. This is possibly silly, even counterproductive--one sees women actually acting like heroines of romance novels, and wondering (though not in so many words) why men do not respond to them in the same way as in the book. But it's a deep element of most peoples' fantasy lives.

This is an itch that contemporary novels try very hard not to scratch. "The moral of the story . . . " is an archaism.

So for people who wouldn't be caught dead reading a bodice ripper, memoir fills that space. Having neatly separated fact and fiction, we now read only "fact" as a way to learn about correct behavior, where a hundred years ago people were perfectly accustomed to taking moral or social lessons out of obvious fiction (from whence the term "morality play"). Memoir alone do we permit ourselves to read for the (now conscious) purpose of obtaining information about how human beings behave in other situations than ours.

But for that, we require versimilitude; we're only interested in reading about being in rehab or growing up in a gang if that is what it is actually like. Otherwise the "compare and contrast" to our own lives seems meaningless. This is also, I think, why David Sedaris is suddenly less funny once you know he just made stuff up. Studies of laughter show that most of what people laugh at isn't really funny; it's a social bonding mechanism. It's hard to socially bond with something that didn't happen.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.