SO, YOU KNOW HOW there are two things nobody can avoid? Novelists have four. Death and taxes, same as the rest of us. Plus first pages and narrators. Can’t live with ’em, can’t live without ’em. And usually, they hang out together.
Perhaps the novelist’s hardest task is deciding who should tell the story. Main character? Secondary character? And what’s his attitude toward others in the tale and the events therein? Is she speaking from within the events as they happen or long after? An outside voice? Limited to a single character or omniscient? Telling all or holding back? Amused, bemused, deeply involved, or bored with it all? It is sometimes said that a change of viewpoint is a change of religion, that an age without firm belief in an all-powerful God cannot utilize the omniscient narrator. While that may or may not be true, this is: the course of the novel is set once the viewpoint is established. Moreover, readers’ relationships to the action and characters of the book rest on this critical decision. We cannot imagine a Gatsby narrated by anyone but Nick Carraway, a Moby Dick without the voice of Ishmael, or a The Fellowship of the Ring told by Frodo. We will be spending anywhere from thirty thousand to two hundred thousand words with this voice, this creature; some of us will even share our bed. Small wonder that the choice of narrator and point of view is never undertaken lightly.
And all this angst over a very small range of possibilities. Here’s pretty much the whole list.
Third person omniscient (sometimes listed as simply “omniscient”). This is the “godlike” option of immense popularity in the nineteenth century. This narrator can be everywhere in his creation at once, so he always knows what everyone is thinking and doing. The Victorian versions often had loads of personality, but it existed outside the story; if they used “I,” it was in the context of talking directly to the reader.
Third person limited. Like the omniscient narrator, this one is an outsider to the action, usually unidentified as anything other than a voice. This one, however, only identifies with one character, going where she goes and seeing what she sees, as well as recording her thoughts. It provides a fairly one-sided view of the action, although this is not the impediment it might seem.
Third person objective. This one sees everything from the outside, thereby offering only external hints at the characters’ interior lives. Since this is pretty much the state we find ourselves in regarding everyone we know, it’s not all that inviting. I don’t need a book to be clueless about other people, thanks.
Stream of consciousness. Not exactly a narrator at all, more an extractor that goes into characters’ heads to pull out their own narration of their existence. More anon, in a separate chapter.
Second person. A true rare bird. You can count the novels you’ll encounter in second-person in a normal reading life on one hand, not use the thumb, and still have digits left free.
First person central. The main character makes his own excuses. Think of Huck Finn or David Copperfield. Probably more popular in novels of growing up than in any other subgenre, with the possible exception of hard-boiled detective novels.
First person secondary. Oh, like you need me to explain this one to you. The sidekick, the second banana, the minor player, the guy standing right next to the hero when he took the bullet—you get the idea. The levels of subtlety and subterfuge are almost limitless here, which explains its perennial popularity with novelists.
That’s the lot. Doesn’t seem like such a big deal, does it? Of course, there are endless combinatory possibilities when the novelist starts dragging in reports, depositions, letters, statements from the involved, and birthday cards from Aunt Maude, but they still fall under one or another of these few headings.
On the other hand, maybe the scarcity of options is precisely why the choice is a big deal: in such a limited universe, how can I make my novel stand out? Will my Bildungsroman (a German jawbreaker applied to those novels about growing up from childhood to adulthood) look like every other Bildungsroman if I use a first-person central narration? On the other hand, will it be recognizable as what it is if I don’t tell the story that way? After all, every novel ever written helps to define the universe of the Novel, and sometimes definition involves limitation. It’s hard to write a novel about a fraud who comes to a bad end for a phony dream in the first-person secondary without seeming to ape The Great Gatsby.
I don’t, however, think that’s the main issue. Certainly novelists are aware of how their new creation may fit into that universe, but concern over narrative viewpoint is more basic than that. It will determine everything, from how well we know the main character to how much we can trust what we’re told to how long the book winds up being.
Everything?
Just about. Consider this: what are the longest novels you’ve ever read? Vanity Fair. Middlemarch. Bleak House. Tom Jones. The Bonfire of the Vanities. What element do those eight-hundred-page monstrosities have in common, aside from chest-crushing weight? Omniscience. Third-person omniscient narration is uniquely well suited to novels of great length. Indeed, it almost invariably causes them. Why? Nowhere to hide. If the novelist plays fair, and if he sets up a universe in which his narrator can know everything, then that figure must show what he knows. He can’t have someone moving surreptitiously in the shadows in another part of town and pretend not to know who is moving, what they’re doing, and why they’re doing it. He’s omniscient, for crying out loud. All-knowing. Can’t very well say, simply because it serves his purpose, oh, I don’t know this one thing. That’s not playing the game. That’s also why you don’t see a lot of omniscience in mysteries. They require secrets. And narrators who can see into everyone’s mind can’t pretend there are secrets.
So what does work for a mystery? Third person limited or objective shows up a good bit. Either we only know the detective’s thoughts or we know no one’s. First person is good. British mystery novels incline more toward first person secondary, American toward first person central. Why? Because of the stories being told. British mysteries tend to be ratiocinative (from the Latin, meaning having to do with working things out by means of reason). They rely on the brilliance of the detective. If he tells the story, we’ll see where his brilliance is taking us and lose all surprise. So instead, the story is told by someone who is, well, slightly dim. Sherlock Holmes has his Dr. Watson, Hercule Poirot his Captain Hastings (sometimes) or another, delegated-just-for-this-once civilian. These narrators aren’t stupid, or no more so than ourselves, and we’re not stupid. But they’re only ordinarily intelligent, whereas Holmes or Poirot are brilliant in a way that’s not entirely human. American mysteries, on the other hand, tend toward the hard-boiled (from the American, meaning I’m tougher than you and will beat or shoot my way to the truth). The detectives aren’t particularly brilliant, but they are tough, tenacious, cocky, and often good company. So they tell their own stories, from Raymond Chandler’s Philip Marlowe to Mickey Spillane’s Mike Hammer to Sue Grafton’s Kinsey Millhone. For people in a tight-lipped line of work, they prove to be surprisingly loquacious, joking and threatening their way through life. The choice of first-person narration helps establish their personality, particularly when, as in the case of Linda Barnes’s Carlotta Carlyle or Millhone, they ironically undercut their position with self-effacing wisecracks. The other really useful thing this viewpoint accomplishes in these novels is to let the main character make mistakes. One of the hallmarks of the American hard-boiled detective story is that periodically the baddies get the upper hand and the hero gets captured, misled, beaten up, shot, or otherwise snookered. When this happens in a third-person narration, as when Sam Spade gets drugged in The Maltese Falcon, it seems like he’s letting down the side. After all, we spend most of the novel watching him be nearly infallible, and then this happens. Incredible. On the flip side, Kinsey Millhone splits her time between telling us how professional she is and noting her shortcomings, so we’re not surprised when she gets herself in a pickle.
Okay, then, so what can first-person narration do and what else is it good for? It creates the illusion of immediacy. And we might as well stop here for a digression on illusion and reality. None of this is real, right? There is no boy Tom Sawyer, no Becky Thatcher he’s sweet on, no pal named Huck. They’re all built up out of words to trick the mind into believing, if just for a little while, that they exist. The mind, for its part, has to play along, not only believing (or at least not disbelieving) but actively taking the elements and adding to them in ways that render a more complete portrait than the one that actually exists on the page. But also, and this is key, no narrator. That creature who talks to us for all those pages is no more real than the rest of the fiction. In fact, it’s a general truth; there is something like the Law of Look Who’s Talking: The narrator of a fictional work is an imaginative and linguistic construct, every bit as much as the characters or events. The “omniscience” of omniscient narration is not that of God but of a godlike fictional construct, and the “person” of first- or third-person narration is not the author but a made-up entity into which the author sends a voice. (I have made a career of studying ephemera and shadows and things not really there; I could just as well be an economist.) So when we talk about what effects a certain technique produces, we’re really speaking of creating certain sorts of illusion. There, I’m glad I got that off my chest.
NOW THEN, THE PROBLEM with first-person narration: here are the disadvantages that make it thorny.
1. The narrator can’t know what other people think.
2. The narrator can’t go where other people go when he’s not around.
3. He’s frequently mistaken about other people.
4. He’s frequently mistaken about himself.
5. He can never get more than a partial grasp on objective truth.
6. He may be hiding something.
Given those limitations, why would a writer choose to employ such a shaky point of view? Consider the advantages.
1. The narrator can’t know what other people think.
2. The narrator can’t go where other people go when he’s not around.
3. He’s frequently mistaken about other people.
4. He’s frequently mistaken about himself.
5. He can never get more than a partial grasp on objective truth.
6. He may be hiding something.
So what are the effects and functions of the first person? First, immediacy. If you want readers to feel close to a character, let her speak for herself. It is good for those growing-up stories because they need us to identify with Pip or David or Huck. That sort of identification is also useful in the picaresque, the narrative version of The Rake’s Progress. The pícaro (the term comes from Spanish) is a rogue or rake who makes his way through the world adventure by adventure, never scrupling too much about common ideas of right and wrong or even about the law, except to keep it off his neck. The scoundrel can be rendered more sympathetic by letting him—or someone very close to him—tell his story, as in the case of Saul Bellow’s The Adventures of Augie March (1953) or Jack Kerouac’s On the Road (1957). Okay, pretty much any Kerouac. We like Augie because he’s charming and witty and direct, Sal Paradise because he’s brooding and wistful and joyous by turns, and both of them because they talk to us.
You can push that identification-equals-sympathy equation one more step and use first-person narration to write about really awful creatures. There are a significant number of novels in which the main character is a monster, usually figuratively but once in a while literally, but in any case some being who in any other sort of book would be the villain. Think about it: would you be friends with Alex, from A Clockwork Orange (1962)? Would you even want to know him? The thing that makes most people lose sleep at night is the thought that there might be an Alex in their county—amoral, murderous, delighting in the aesthetics of violence, waiting for opportunity. But the thing is, in print, in his own voice, Alex is a charmer. Intelligent. Handy with a phrase. Witty. Clever. Sympathetic. Now how do you get from “amoral” and “murderous” to “sympathetic”? Words. Alex—or rather, his creator, Anthony Burgess—spins a web of words to ensnare us in the more attractive elements of his personality while shielding us from a too-direct view of his more heinous actions. And “spinning a web of words” is exactly how the eponymous antihero of John Gardner’s Grendel describes his self-conscious movement through the world. Gardner’s task is even tougher than Burgess’s: the teenage thug is a fairly recent development, but Grendel has been the byword for evil for more than a millennium. Nor does the novelist want to rehabilitate him or make him cute and cuddly. He’s still a monster, still eats people, still revels in mayhem and blood. Not exactly best-buddy material. What Gardner wants to do is something trickier and more challenging, to keep him a monster and nevertheless make him sympathetic. So he makes him a good deal like us (except for the people-eating part): witty, observant, interested in language and what it can do for good or ill, interested in and critical of the shortcomings of the human society around him, alienated, aggrieved, and a little bit sorry for himself. As a result, readers are able to see the necessity of Beowulf’s ultimate triumph while regretting the loss of Grendel’s remarkable voice. And if you can do that for a twelve-hundred-year-old monster, you’ve really done something.
Now, the fact that there are only six options, and more like four for all intents and purposes, doesn’t really limit the way stories are told very much. Novelists can mix and match, adjust, vary, and just plain abuse the options they’re presented with. In his masterful Snow, Orhan Pamuk employs a kaleidoscopic point of view that seems for much of the novel to be nearly omniscient, then limited, and finally first person as a narrator from outside the story proper, called Orhan just to confuse the issue, reveals that he has been assembling this tale from documents left by the main character, a poet named Ka, as well as statements made about the now-deceased writer by friends and acquaintances. Both the apparent omniscience and objectivity are illusory; as a friend, Orhan is moved by feelings of loyalty and confusion in recounting the events of Ka’s final trip home to Turkey. And here’s the problem with categories: they’re often fairly crude devices. Snow is a remarkably subtle narrative performance. Is it “first-person secondary”? Well, yes, sort of. That is, the narrator turns out to be a named character who interacts with others in the story. Yet he’s also a chameleon who imitates the devices of various third-person points of view. The definition doesn’t really account for the changes a writer of Pamuk’s ability can bring.
And if that’s a problematic first-person novel, try John Fowles’s The French Lieutenant’s Woman (1969), in which the third-person narrator actually appears. Twice. The book takes the form of a Victorian novel, exploiting its conventions of omniscience (when it suits the narrative purpose) and reader-narrator intimacy. Two times, however, a person resembling Fowles himself (in different guises) turns up, once in a railway carriage where he observes the main character and once toward the end, when he sets his watch back fifteen minutes, thereby allowing for the book’s famous two endings. There is only one person who can reset time in a novel, and he’s not a character inside the story. Moreover, Fowles frequently uses “I” to refer to the narrator, although he never interacts with anyone or anything in the story. Even his ride on the train and walk outside the residence of Dante Gabriel Rossetti have a kind of stage artifice about them, suggesting he is metaphorically, rather than literally, present. So then, first person? Third? Third, pretty clearly, since the narrator isn’t really a character in the novel, but like Snow, Fowles’s book is far more subtle and sinuous in its use of point of view than any mere definition can ever hope to be.
So what? Why should we care about narrative POV? Does it really matter if it’s first or third—or that the definitions don’t always get the job done?
Well, yes and no. First, the no. You can get through most novels satisfactorily without paying much attention to who’s doing the telling. You’ll get the general drift of almost every novel, and everything there is to get with a lot of them. But also, yes. Who narrates matters in terms of trust. In general, third-person narration can be relied upon to be accurate. Narratives with “I”—as we shall see shortly—not so much. Real people lie. Misremember. Get confused. In one of history’s most famous writers’ tiffs, Mary McCarthy said of her former friend Lillian Hellman that “every word she says is a lie, and that includes ‘and’ and ‘the.’” If persons in the real world have a tenuous connection to veracity, why should we expect those in made-up stories to be more truthful?
But differences can matter in other ways, too. What’s our relationship to the story told? How far removed are we from it? How immersed in it? How are we being manipulated to see events one way or another? These things are often dictated by the choice of narrator. So we just might want to notice who that elusive creature is.