Searching for Literary Sasquatch, the Elusive Narrative Voice

Aaron Gilbreath


Narrative voice is a lot like the Tao: if you can define the Tao, it is not the Tao. Writers in and out of workshop, especially prose writers, talk frequently of the power of voice, the elusiveness and necessity of it. What you don't hear a lot of are definitions. I feel like all we writers have to work with is the literary version of Supreme Court Justice Potter Stewart's infamous conception of hardcore pornography: I can't define it, but I know it when I see it.

So, what the hell is a narrative voice? Is voice the way the word "hell" transforms an otherwise utilitarian sentence into a sentence with a stylistically discernible character? Is it having a character say the above line in a raspy, Alabama Waffle House waitress' drawl? No, that's more of an accent, and accents are an attribute of the spoken.

Voice seems a composite of elements. Yes, an accent may be part of voice, but it's a minor part, and too much of an accent can easily push the prose toward stereotype. At its fundamental level, voice is the opposite of stereotype, the antithesis of cliché, a palpable originality as recognizable to readers as a fingerprint might be to detectives.

When I was younger I found personality in my favorite music and books and, as an aspiring writer, wanted only to keep a supply of personality in my pantry, to be taken like a potion when I needed to enliven my bland writing. I wanted Iggy and the Stooges inside my ballpoint pen, the power of "Search and Destroy" in my prose. Whatever the literary equivalent of a pierced septum and a knuckle tattoo was, I wanted it It's easy to see personality as something godlike, but in your quest to possess it, you can just as easily lapse into a lazy pursuit: wearing band t-shirts, dying your hair electric blue, making loud public statements so that everyone will see and think you're something. But voice isn't so desperate or insincere in its calculation.


Narrative voice is personality and persona. Despite sharing a root word, there is a difference between the two terms. The word "persona" derives from the Latin word for "mask" or "character."

Personality both on the page and in certain social situations is to some degree consciously manufactured. As social creatures, people present themselves differently depending on the situation, playing certain parts for specific crowds. To elicit the response we want from these different audiences, we adjust our personality, or at least our speaking voice. Certain friends might find your quoting of Nietzsche a bit off-putting, so while eating with them, you repress the Nietzsche comments and instead chat about TV or football. Other friends might dig the Nietzsche stuff and loath football. Such adjusts require forecasting the audience's reaction to certain information: what do we, the writer, want to elicit in the reader?

Adjusting to our audience doesn't necessarily make the persona inauthentic. Sometimes it's faking, the way me speaking in an Alabama accent would be. Sometimes it's simply editing out certain parts of ourselves to highlight others. Tweaking what's already a part of our personality is very different than wholly conjuring new elements in order to mislead or conceal. But it gets tricky. A 'mask' is by definition a cover, and so relying excessively on a persona may become inauthentic. Yet when persona is simply the public face of a larger, underlying personality, then small adjustments are a matter of refining the authentic speaking voice that is most natural to us.

But comparing the narrative voice to our speaking voice is problematic: speech isn't the right analogy. In his introduction to The Best American Essays 2004, author Louis Menand says writing is the opposite of speech because writing is not spontaneous. "Some confusion about what it means to have a voice arises from the metaphor itself. Many readers, and many writers for that matter, think that effectiveness in writing has something to do with how close it is to speech." Menand explains that, one reason "that speech is a bad metaphor for writing is that writing, for 99 percent of people who do it, is the opposite of spontaneous. Some writers write many drafts of a piece, and some write one draft, at the pace of a snail. But chattiness, slanginess, in-your-face-ness, and any other feature of writing that is conventionally characterized as "like speech" are all usually the results of intense experimentation, revision, calibrating, walks around the block, unnecessary phone calls, and recalibrating. Writers are people in whom l'espirit de l'escalier is a recurrent experience: they are always thinking of the perfect riposte when the moment for saying it has already passed. So they wait a few years and put it in print. Writers are no mere copyists of language; they are polishers, embellishers, perfecters. They are people who spend hours getting the timing exactly right – so that it sounds absolutely unrehearsed." This explains why readers who meet their favorite writers in person often find the writer's speaking voice devoid of the traits they find endearing in print: humorists who aren't funny, verbose writers who speak little.

In his book On Writing Well, William Zinsser briefly explores this spoken-written voice distinction by quoting an introductory line from one of E.B. White's pieces:

I spent several days and nights in mid-September with an ailing pig and I feel driven to account for this stretch of time, more particularly since the pig died at last, and I lived, and things might easily have gone the other way around and none left to do the accounting.


Zinsser says, "Nothing about [this material] is accidental. It's a disciplined act of writing. The grammar is formal, the words are plain and precise, and the cadences are those of a poet." That writing is laborious work is hardly news, but what interests me about Zinsser's comment is that he describes the voice that resulted from White's "methodical act of composition" as one of disarming warmth, and in so doing Zinsser establishes the relationship between discipline and voice. Voice is not purely a spoken personality, nor is it the channeled form of one's inner person. Rather, voice is a highly refined projection. While not necessarily a fabrication in the sense that it's a forgery, voice is somewhat manufactured because it is achieved through revision, which involves conscious, deliberate decision-making and careful refinement. So contrary to what many young writers might hope or believe, voice isn't a mountain spring gurgling, it's a whisky still distilling. It does not arrive ready-made.

If certain things can be defined as the sum of their parts, then voice seems to me partly an amalgam of its parts. It's the collective effect of an author's style, phrasing, word choice, rhythm and cadence, and maybe a little accent and affect too, depending on the narrator. Beyond the texture of the words themselves, voice expresses some grander vision, the artist's worldview. If there could be gleaned some sort of essential recipe, it might be something like this: voice is word choice plus favored sentence structures plus rhythm plus ratio of sentence complexity to simplicity plus some personalized sonic fingerprint that impresses itself onto the sound of one's words in some unquantifiable way. Voice is essentially the slippery spirit of a narrator's personality filtered through a persona—and the more we attempt to grab hold of that personality for inspection, the more it slips from our grasp.

And that's as close to holding this elusive literary Sasquatch as I can get.

In lieu of a more concise definition, here are three works of fiction whose narrator's evidence strong, distinct voices: Stacey Richter's story "Christ, Their Lord;" David Gates' novel Jernigan; and Barry Hannah's novel Ray.


In Richter's story the narrator says:

I need a special sticker on my car now to get in and out of the neighborhood in the evening. The sticker designates me as an accursed resident of Yuletide Village as I honk and scowl and inch my way at five miles per hour through the smiling and entranced crowds touring the holiday lights, until I manage to terrify a lady in a three-wheeled Handy Cart out of the way and turn into my own driveway—a pool of darkness in a sea of light. For we have no decorations. No crèche, no baby Jesus, no rebellious Star of David. No animatronic carolers jerking to and fro like drunken frat dudes, no Charlie Brown in a snowsuit, nothing to delight children or adults or disabled adults or grandparents or sullen adolescents with creamy, sentimental centers who are not really so tough. What we offer the holiday crowd is this: a little nothing for no one.

I would characterize the narrator's voice as: witty, cynical, darkly comic, urban, youthful, and energetic. Using the above recipe as a guide, we can isolate some signature elements. Here we see: (a) word choice ("dude," "to and fro," "animatronic") + (b) favored sentence structures (specifically, short sentences mixed with lengthier ones (one is four-lines long)) + (c) rhythm + (d) ratio of sentence complexity versus simplicity + (e) some unquantifiable personalized sonic fingerprint (such as the unique phrase "creamy, sentimental centers," and Richter's engaging mix of high- and low-brow syntax). The voice is a noticeably bitter one: notice words such as "accursed," and the frequency of the word "no." It's also comic—characterizing a Star of David as rebellious; mentioning "disabled adults," which echoes an earlier joke the narrator made about busloads of handicapped adults wandering the neighborhood. Also, the voice carries a mournful quality, a sense of deep alienation. This is Richter's singular alchemy. But is it Richter's voice or simply the narrator's?


In first-person personal nonfiction, we can more easily speak of the narrator's voice as the author's voice. In fiction, the two voices can differ wildly. In the same way that the details of a fictional story don't necessarily mirror the facts of the author's life, neither is the narrator's voice necessarily that of the author. It's something novelists hear all the time, with readers asking, "How much of this is autobiographical?" and "Did that really happen to you?" Or worse: calling the fictional narrator you, instead of the narrator, when recounting events in the story. Novels aren't thinly disguised memoirs. And the writer's and their characters' voices do not always match. Yet, in certain cases and in varying degrees, authors' overall voices shapes and colors their narrators' voices, with many telltale traits carrying over from real life into the printed one.

A revealing comparison to make here would be Nabokov doing Humbert Humbert and Nabokov doing Pnin. Instead, with the above in mind, let us now turn to the namesake of David Gate's novel Jernigan, as he introduces himself to his son's band mates:  

"Peter Jernigan," I said, giving the man's handshake. God knows why Pete: I never went by Pete. Powerful Pete. Some compulsion to be hearty. By which I guess I mean not oppressively parental. "Gang all here?" I said.

Using my homebrewed formulation, here we see (a) word choice ("hearty" and "God knows why") + (b) favored sentence structures (short, incomplete ones) + (c) rhythm (taken alone, some of the sentence fragments have a jerky, stop start motion, but together, they create a highly rhythmic, melodic feel) + (d) ratio of sentence complexity versus simplicity + (e) some personalized sonic fingerprint. Here the latter comes in the form of unique phrases and word combinations such as "oppressively parental" and "the man's handshake," which suggest buried notions and views about the larger world, including ideas about bad parenting, gender stereotypes, and masculinity. These notions impress themselves onto the sound of Jernigan's words and the texture of his whole persona. Here, words are character. They reflect his very being. He's defeated. He says oppressively parental, but also, it's his spirit that's oppressed—by alcohol, negativity, a bad attitude and a self-destructive bent. He's hyperconscious, constantly studying his own words and behavior because he's perplexed by them (God knows why he acts as he does). Here he's confused and struggling to maintain appearances by giving the right first impression. And he's lying to misrepresent himself as Mister Cool, one of the kids.

This is the voice of a selfish, antagonistic, aggressively opinionated and intellectually arrogant alcoholic confounded by his own determination to ruin himself. The voice is dry, sardonic, cold. It's acidic and derisive, smart and funny as hell. It's also distant and self-aware. If I can isolate the essence of Jernigan's voice, though, it's his use of incomplete sentences. Fragments dominate Jernigan's inner dialogue and a substantial amount of the novel's prose, and this gives much of the prose its distinctive flavor, a flavor that is clearly the result of the author's meticulous labor.

Here's another gem from an early scene where Jernigan thinks:

God damned good and tired around here listening to all this shit about how Jernigan only lives in his own head.

Thought, too, about trying to find some paper and writing to my son, asking his forgiveness. Oh, not for anything all that specific. It would be the thought that counted. Except what were you going to say after asking forgiveness? Pledge to do better? Right, I can see you now, doing better.

The overall effect of these fragments on me was initially jarring, but it soon made the reading experience the incredible one that it was. The fragments alternately soothed, amused and confused me, coaxing me into an anaesthetized warp-zone that mirrored the narrator's languid, lost and irritable state. But his voice's essence is greater than all that. To use Jernigan's own words, his is the voice of a person who only lives in his own head. It's the sound of a person's unfiltered thinking mind, even when spoken aloud. The voice's effect is similar to that of Dennis Johnson's prose in Jesus' Son, which lets readers feel his characters' drugged up confusion.


Like Jernigan, Barry Hannah's novel Ray is one dominated by voice. It's written in third-person, and the narrator's language is wild, energetic, bizarre. For instance: Ray says, "So I drove back to the fancy rich thing of my house." The "thing of a house" part grabs me here; the description's so unnecessarily long. You could cut that and rearrange the words to make it concise—which Hannah does in many other places—but he chose this. The result is pure music, and it both sings and lingers in the mind.

So too when Ray flies to Atlanta to see his lover and says, "Because of the photograph, I was cooled down toward much gesture of lust for her, though."

Later the narrator says, "And although you try to get shut of those gorgeous moments when we had nothing but good neighbors, the pines, and the sky to look at, it's true, we had a sublimity." I love this line even though I'm only partially sure what it's saying. Again, there's music in it. It mixes specifics with generalities, and the peaceful domestic images that are familiar (the pines, the sky). Yet the line is redolent of transcendence, partly because it's also so suggestive and ethereal, partly because it's not technically a complete sentence. You might say the same for much of Hannah's narrative voice.

Odd diction, bizarre inversions and unusual, unexpected phrasings define Hannah's prose. It's often reaching for new lyric frontiers, like a John Coltrane solo—striving for new phrases capable of expressing what the speaker has churning inside. It's what a recent reviewer on Slate called "Hannah's Martian contortions of language…" If the language in Ray is sometimes mystifying, there's at least always magic in it. At their tamest, his sentences are interesting. At their best, they're almost a whole new way of seeing the world. Like James Brown's dance moves, Hannah's voice sometimes feels simultaneously out of control and finely honed. The energy of its movement is one of abandon, yet it never feels unmanaged. That wildness feels constructed—which is what Zinsser was getting at with EB White.

The narrators in all three of these stories have very discernible personalities, and the voices—rather than the plots or the likability of the characters—play an overwhelming role in what makes each book so enthralling. I've heard writers, writing students and magazine editors characterize voice-driven writing as the most compelling type of literary writing, the type that readers and editors seek. Personally, a strong voice can carry me through a relatively uneventful narrative, so that how the story is told becomes more important that what was told.

Although he offers no definition, Harper's previous editor Lewis H. Lapham discusses the power and necessity of voice in literary works in his piece "On Reading." "On first opening a book," Lapham says, "I listen for the sound of the human voice. By this device I am absolved from reading much of what is published in a given year… I listen, instead, for a voice in which I can hear the music of the human improvisation as performed through 5,000 years on the stage of recorded time… As a student, and later as an editor and occasional writer of reviews, I used to feel obligated to finish every book I began to read. This I no longer do. If within the first few pages I cannot hear the author's voice… I abandon him at the first convenient opportunity."

I treat books the same way Lapham does, and the technique (or impatience) touches upon something that most people experience in real life: we meet someone for the first time, listen to them talk—both what they say and how they say it—and we either like this person or we don't based on that first impression. Voice is part of our literary first impression.

But let me back up: if personality is often a deliberate projection, saying that a narrative personality is partly constructed isn't to imply that it's a complete façade or disingenuous. Much of our narrative personality seems to come from the character of our natural speaking and inner voices, in that, as speakers and thinkers, we all, over the years, develop certain favored sentence structures, idioms and phrases. There are constructs we gravitate to without thinking, almost a default setting, which form the basis of all revising we do to that base material. For instance, if a woman I was dating said I must not care much for her because I'd never asked what she did for a living, I wouldn't, as Peter Jernigan did when faced with this situation, think, "Huh. Yeah. Well. Certainly had me nailed on that one." I would likely think, "Uh oh, she has me there. How have I been so self-involved not to ask?" I would phrase this not as a series of Jernigan fragments but as complete sentences or some kind of countrified phrases that my dad got me saying as a kid. There appears to be a subconscious element to voice, in that, no matter how many times a writer rewrites and recalibrates a sentence, there was already a preexisting framework—a limited number of phrasing options—from within which we were, without knowing it, working. This seems to be the "natural phrasing" that Menand references when comparing voice to signing.

When I think about "phrasing," I think about drumming. There's the straight 4/4 beat of rock and roll. There's the "back beat" in hard bop jazz. There's the 1-2, 1-2 beat in surf instrumental music. And then there's someone like jazz drummer Philly Jo Jones who's so fluid and spontaneous that his drumming conforms to no standard beat, only to the shifting, second-to-second demands of the improvised moment. Philly Jo's playing is intuitive, which makes his phrasing unpredictable, but I recognize it every time I hear it as his own. 4/4 just wasn't his voice. His voice is defined as much by what he doesn't play as by what he does.

So, even in the absence of a definition, I believe what Zinsser says when he says that "My commodity as a writer, whatever I'm writing about, is me. And your commodity is you. Don't alter your voice to fit your subject. Develop one voice that readers will recognize when they hear it on the page, a voice that's enjoyable not only in its musical line but in its avoidance of sounds that would cheapen its tone: breeziness and condescension and clichés." As if to put the fear of God in writers, Menand says that "There are probably all kinds of literary sins that prevent a piece of writing from having a voice, but there seems to be no guaranteed technique for creating one." Which again leaves us with the idea of voice being a lot like Supreme Court Justice Potter Stewart's conception of hardcore pornography: I can't define it, but I know it when I see it. That or I just prefer a raunchy metaphor. I could have said "ribald," but that wouldn't be my voice.