CHAPTER SIXTEEN

FIRST-PERSON NARRATIVE

When you use a first-person narrator, you are almost required to tell the story in someone else’s voice — the voice of the character telling the tale. A careless writer will have all her first-person narrator s talk amazingly like herself, but if you take characterization seriously, the use of first person will lead you to discover a new voice for each story told by a different narrator.

One mistake many writers have made — particularly nineteenth-century humorists like Artemus Ward — is to make the first-person narrator’s voice so eccentric or heavily accented that the story becomes almost unreadable. In fairness, I should point out that in Ward’s own time, stories tended to be read aloud; the heavily accented writing did not slow down the pace, since reading aloud is already slow; and it also provided the reader with a guide to pronouncing the comical accents.

But by and large you should attempt to create the narrator’s voice through his attitude and implied past, letting the speech reflect his educational level and regional accent only in syntax and word choice, not in odd spellings or endless pronunciation guides. Nothing is more deadly than trying to read sentence after sentence written like this: “Ah niver did figgah out whah in hivven’s nam e a good of boah lahk ‘at wen’ crazy an’ stahted in killin’ folks.” Furthermore, the narrator doesn’t hear his own accent anyway, and so would never write it that way. The narrator would write: “I never did figure out why in heaven’s name a good old boy like that went crazy and started in killing folks.” That’ s what he thinks he said, and it’s only because you have a different accent that you think his words should be spelled another way.

Cheapest of all is when writers try to show someone is uneducated by using apostrophes willy-nilly: “I’m goin’ t’ th’ store, Nell. We’re runnin’ out o’ beer.” In the first place, most people, even educated people, drop letters in normal informal speech. Anyone who never does is a hopeless prig. In the second place, the dropped g in ing endings is actually older than the pronounced g, and therefore is arguably more correct; certainly it is a natural survival of the ancient spoken tongue, and doesn’t really denote an uneducated person — except to someone who is uneducated. It almost invariably comes across as a way for the author to show that he sneers at the person who speaks that way, and only rarely is this the conclusion you’ll want the audience to reach about your first-person narrator.

WHICH PERSON IS FIRST?

The main limitation on the first-person narrative is that your narrator has to be present at the key scenes. A first-person narrator who merely hears about the major events of the story is no good to you at all. So you have to work your narrator into the action so tightly that he is present whenever you need him to observe something.

The easiest way is to make the narrator the protagonist (or vice versa — make your protagonist the narrator). The trouble here is that the protagonist is the character with whom the audience sympathizes. She is likely to do interesting and important things during the course of the story, or suffer terrible loss or pain; how well will her voice serve to tell about these things?

For instance, if one of the key events is the death of the protagonist’s beloved child, how coherently is she going to be able to write about the events leading up to it? If she is too emotional, it will become melodramatic; if too graphic, it will become unbearably intense. Yet if you retreat, and her narration becomes clear and cool, you run the risk of having the audience regard her as cold and heartless. This is not to say it can’t be done — it just requires a careful balance.

The timid writer, of course, will decide not to show that key event at all, or will use telling rather than showing:

We finally got Johnny into a decent school, Bill’s job was settling clown, and I could forget about those people and their terrible phone calls for hours at a time. I thought everything was going to be all right, until I heard someone screaming and pounding on my door one day. I knew at once that those people hadn’t forgotten me, that they had done something terrible, just like they said. I opened the door. It was my neighbor, Rainie. “He just drove off!” she cried. “Matt’s calling the ambulance — ”

There’s no point in me telling you about the next few days. If you have a child of your own, you already know; if you don’t, you can’t possibly understand. They didn’t try to contact me until we got home from Johnny’s burial. Maybe it was a sense of decency — presumably they have children, too. More likely they were waiting until they thought I was calmed down enough to be rational. To listen.

I listened. I still had a husband and two other children.

The narrator isn’t out on the street watching when her child is killed. No horrifying moment as she realizes that the truck is going to jump the side-walk and hit her son. No descriptions of a crumpled body on the street. For a first-person narrator to describe such things at all might seem ghoulish. To adequately express the emotions might be impossible. To describe it all coldly would be too clinical. Yet to skip over the events, as in the example, is rather coy.

Choose one. Maybe coyness is in character; maybe the character is clinical. Maybe you’re a good enough writer to tell the immediate feelings of a mother who watched her child die — without getting maudlin or grotesque. These choices are all available.

But there are several other choices to keep in mind. You can decide to use a different narrator. Why not the neighbor woman, Rainie? Make her the protagonist’s close confidante, so she can be closely involved in this woman’s struggle. She can actually see the accident with the less passionate horror of a bystander, avoiding the much stronger emotions of a parent. She has enough distance to be a clear, direct narrator; enough closeness to witness everything.

Or you can use a third-person narrator — with all the drawbacks and benefits that entails. We’ll discuss those later.

Arthur Conan Doyle chose well in deciding not to have Sherlock Holmes narrate his own stories. Using Watson as narrator allowed Doyle to withhold information from the audience without being unfair. Holmes knew certain information, but Watson didn’t, so Watson could tell us all that he knew in the order he found it out, without spoiling the surprise. Since Watson never knows as much as Holmes, neither do we.

There is another benefit, though. Imagine if we had to listen to Holmes’s intellectual, arrogant tone through every word of the story. Instead of admiring his godlike mind from below, we would find him insufferably conceited. He might even be ridiculous. This is the choice Agatha Christie made with Hercule Poirot — but Poirot was never worshipped as audiences have worshipped Sherlock Holmes.

The narrator’ s voice is your greatest asset — and your greatest drawback. Your first-person narrator can’t be a bore, or your story will be boring. She can’t describe herself performing noble acts, or she will seem vain for having told the tale at all.

Yet you can tell us much about your narrator by showing him do a brave, heroic act without him giving us a sign that he realizes the act was heroic at all. Or he can do something terrible, all the while explaining exactly why his crime was not a crime at all, but a necessary act — while we listen in horror.

She wouldn’t be quiet, even when I tried to tell her how important it was for her not to say those things. There’s some things a man just doesn’t have to put up with from a woman. You listen to them blab on all the time about their girlfriends and going shopping and what the kids did, and you figure that’s just what women’s heads are full of. But when she starts getting down on a man for doing what men do, well, that’s over the line, that’s more than a man has to put up with. What she’s really doing, she’s just trying to get you to prove to her that you really are a man, maybe sometimes just because you’ve been too tired sometimes, or too nice about it in bed, so you hear her talking like that, you don’t put your hands in your pockets. You knock her around, you let her know that you still got the power in your arm, you still got the strength to be the man she needs you to be. It hurts her, of course, but it hurts her sweet, that’s what my dad always said, she gets a bloody lip but it tastes sweet to her because she knows she’s got a real man. Only this time she just wouldn’t quiet down, she just kept yelling at me and saying crap that I don’t have to put up with, and then she kept trying to go out onto the street and spread all our family business all over the neighborhood, and I couldn’t let her do that, could I? You wouldn’t either, man, and don’t tell me you never hit your woman a little bit harder than you meant to, what with her mouthing off.

We may not love this character, but we know him better from hearing his version of his actions than we ever would by hearing them described by someone else. This passage ostensibly defends the narrator’s mistreatment of his wife, but in fact it reveals very clearly his monstrous misconception of the way other people think and feel. That’s one of the best reasons to use first person — to let us live for a while in a strange or twisted world, to see the world as someone else sees it. Yet because the narrator is not the author, but rather a character, the readers know that the author doesn’t necessarily agree with the narrator. In fact, in this passage, if I handled the irony properly, it should be clear to a late-twentieth-century reader that the author is completely out of sympathy with the narrator.

NO FOURTH WALL

A third-person narrator flits like an invisible bird from place to place — readers don’t usually spend much time worrying about how she happens to know all this stuff, or why she’s writing it down. The narrator is a story-teller, plain and simple; we ignore her and listen to the tale.

But the first-person narrator is physically taking part in the story. Therefore, he must have some reason for telling the story. By implication, he must also have some idea of who his audience is. Even though you, the author, may be maintaining a fourth wall between your characters and your readers, he, the narrator, is not keeping that fourth wall between himself and the audience he thinks he’s telling the story to.

The most common way of dealing with this problem has always been the frame story. Several people gather, conversing; one thing leads to another, until one begins to entertain or inform all the others by telling the main story. No one expects anything significant to happen in the frame — it’s just an excuse for the first-person narrator to tell his story to an audience that is not the reader of the book. The frame is told in third person; only the tale-within-a-tale is told in first person.

You know of many examples, I’m sure. Rudyard Kipling used the device often; H.G. Wells’s The Time Machine has a frame story, as do countless tales-told-in-a-bar. One drawback is that such stories are oral, and so you deny yourself the use of formal written language. Another problem is that since the story opens with the frame, if the frame is dull the audience may never get to the story you really care about.

The frame is not the only way to deal with the fact that the character is narrating a story. Some first-person stories are told as epistolaries, letters from one person to another (The Color Purple, by Alice Walker, for instance). Some are cast as speeches, diary entries, essays, explanations to a judge, confessions to an analyst (remember the punch line at the end of Portnoy’s Complaint?). The narrator’s purpose in writing may be to tell a curious tale, to persuade the presumed audience to a course of action, to excuse the narrator for some crime. Or the narrator may be explaining why he admires his friend, who is the protagonist of the story — which is presumably the reason why Watson set down his tales of Sherlock Holmes and Archie Goodwin told us of the exploits of Nero Wolfe.

In choosing a first-person narrator you should have in mind what his reason is for telling the tale; tale-telling is part of his character. Whether you explain her purpose or not, knowing it yourself will help you shape and control the presentation of the story; it will help establish which events the character would tell and which she would leave out, which she would lie about and which she would tell straight.

UNRELIABLE NARRATORS

What? Your first-person narrator might lie? Of course. But if you mean him to be a liar, you must find ways to let your audience know he’s unreliable.

The easiest way is to have him get caught in one lie and admit it — the audience immediately begins to suspect him of lying about other things, too. Even then, your audience has a right to expect that you, the author, will let them know which of the narrator’s statements to believe and which are lies.

One way to clue in the audience is to establish another character, not the narrator, whose word we trust, and let her corroborate the key events that really happened. Usually this corroboration takes place in scenes within the story, but some storytellers go to the extreme of letting this more reliable narrator take over the first-person narration part of the way through.

Switching first-person narrators mid-story is usually ineffective and always difficult, because it violates the illusion that the character is “really” telling the tale. But if you find you must change narrators, it helps to give your readers some clue. For instance, if the first eight chapters are narrated by Nora, you might put in a division page that says, “Part I: Nora.” When Pete takes over as narrator, again you put in a page that is blank except for the words “Part II: Pete.” Or you could establish multiple narrators in a frame — both characters are present in the bar or the courtroom, and we expect both to tell their parts of the story.

More difficult than changing to a more-reliable narrator is the technique of letting us know the truth of the story by implication. The narrator is lying about the things that matter to him; you must therefore carefully let us know his motive for lying so that we’ll know which parts of his story would need to be faked to accomplish his purpose. Is he concealing the facts about his own crimes? Then you can lead us to doubt his words concerning his own alibi or his own reaction to the crime. Is the story a letter in which the narrator is trying to persuade another character of his faithful love for her? Then obviously we will doubt his story about what actually went on when he was alone in the room with her rival.

The use of an unreliable narrator can add a delicious element of uncertainty to a story, with occasional revisions of the readers’ understanding of all that went before. But used badly, or to excess, the unreliable narrator leaves the reader wondering why he’s bothering to read the story, or furious that the author never let him know what “really” happened. It’s a dangerous thing to attempt, and only occasionally worth doing.

One exception, just to show you what it’s like when it’s done well, is Thomas Gavin’s brilliant novel The Last Film of Emile Vico. The first-person narrator is a 1930s movie cameraman, writing the book as a memoir of his relationship with Vico, who has recently disappeared under mysterious circumstances. But we soon realize that the narrator, Griswold Farley, suspects himself of murdering Vico — or rather suspects the other personality that occasionally dominates him, a figure he calls Spyhawk. He suspects Spy-hawk because Spyhawk seems to know more about the events surrounding Vico’s disappearance than Farley himself knows. However, Farley can’t be sure — in the past, Spyhawk has caused Farley to feel guilt for wrongs that he didn’t commit. Thus Farley himself knows that his own memories are not reliable; nor can he trust the feelings Spyhawk gives him. The result is that the entire novel is structured as an idea story — Farley and the reader are trying to discover the truth about the mystery of Vico’s disappearance and Farley’s involvement in it.

Another reason to study Gavin’s book is that he handles the first-person point of view so expertly. The narrator constantly sees everything from a camera’s- eye view, as if he observes even his own life through a lens; indeed, his alternate personality is exactly the kind of “spy” that a cameraman represents. This motif shows up throughout the narrative, so that it’s part of the narrator’s self; after a short time, the reader is no longer consciously aware of it, yet continues to see the story as if framed in a camera’s shot.

At the beginning of chapter eight, as Farley launches into a flashback, he switches to present tense, because he is writing this part of his memoir as if it were a movie script. Instead of being an odd, inappropriate choice, present tense has purpose and meaning within the story; it is exactly appropriate for what narrator Farley and author Gavin are trying to accomplish.

DISTANCE IN TIME

One thing Gavin is wrestling with in The Last Film of Emile Vico is a problem that comes with all first-person narrators: the problem of time. The narrator, as a participant in the events, is telling about what happened in the past. He is looking backward. He is distant in time from the story itself.

Contrast this with the third-person narrator. Even though most third-person accounts are told in past tense, they feel quite immediate. There is not necessarily any sense of the narrator remembering the events. They are recounted as they are experienced. There is no distance in time.

However, with third person there is distance in space. That is, the narrator, though she can dip into one or more minds, is never a person who is actually there. She is always an invisible observer, always at some distance.

So first person is distant in time, third person in space. Consciously or not, storytellers struggle to break down both barriers and achieve immediacy. The use of present tense and stream of consciousness were attempts to bridge the first-person time barrier — with little success, I might add, since both techniques tend to drive away the vast majority of the potential audience. The use of deep penetration in the limited third person is an attempt to break down the barrier of space in that narrative voice, and it works very well; thus it has become the most widely used narrative approach. (I’ll explain “deep penetration” when I deal with third person in chapter 17.)

One way to minimize the distance in time is to have the first-person narrator tell the story in chunks, writing it as the story goes along. Gavin does this in Emile Vico. The memoir is begun in a hotel room, where the narrator is in hiding, afraid that a relentless-seeming police detective is going to find evidence linking him to Vico’s disappearance. At the time the early chapters are written, the narrator himself does not know how things will come out. He does not know the end from the beginning, because the first part, at least, is written before the story has ended — solving another problem with first-person narration that I’ll deal with more in a moment.

Another example of solving the time-distance problem is Gene Wolfe’s historical novel Soldier of the Mist. The narrator is a former soldier, apparently a survivor of the invading Persian army at the time of Thermopylae. A wound — or a curse from the gods — has stolen from him his ability to retain long-term memory. He wakes up each morning remembering nothing from the night before. So the novel is written as the journal he keeps to remind himself of his entire life — the book becomes his memory. Each day he begins by reading all of the book to date, until it becomes too long; his friends or fellow travelers even have to remind him to read the book, because he forgets that he has written it. Therefore there are gaps during the times when he forgot the book existed, and what happened during those lost sections can never be recovered except for the few scraps of information that others can give him about himself.

Certainly this book has defeated the first-person time-distance problem — but, alas, at a high price, since we are forced to put up with some of the repetitions and irrelevancies that such an artless character would include in a book he is writing, not to entertain, but to inform himself. In short, the very truthfulness of the characterization makes it harder to maintain the intensity of emotional involvement, since reading itself becomes hard work under that circumstance. In the art of storytelling, every good thing has its price. In the case of Wolfe’s and Gavin’s books, the price is well worth paying, in my opinion. Other readers, though, may not agree. Each author has treated his first-person narrator more realistically, which opens his book to one group of readers; but in the process, the book has also been closed to another potential audience. That’s what happens with every choice you make.

WITHHOLDING INFORMATION

One technical problem with most first-person stories, arising out of distance in time, is that the narrator knows the end of the story. There’s really nothing to stop him from announcing it from the start. Imagine a book that begins this way:

In the Case of the Vanishing Hitchhiker, we found out by the end that the hitchhiker was really the long-lost daughter of the man who picked her up, traveling in disguise, and he never knew it until long after he killed her and threw her body into the newly poured cement foundation of his new office building.

After that sentence, there isn’t much of the mystery left to wonder about through the rest of the book. Yet potentially the first-person narrator could give us such information at the beginning of most stories.

Telling the end at the beginning is a fatalerror only with idea stories; many a character or milieu or event story thrives on the dramatic irony that comes from knowing the end from the beginning. Still, the fact that the first-person narrator doesn’t tell us the ending is a constant, unconscious reminder of artifice. She is deliberately leaving us in suspense. So unless the narrator is supposed to be a mystery writer, leaving us in suspense is probably out of character for her.

This is not the problem it might be, because the contemporary community of writers and readers has developed a convention for dealing fairly with the reader in first-person stories. Readers allow the first-person narrator to withhold the ending, as long as he tells us at each stage in the story all that the character knew at that point in time.

If the narrator is a detective, he tells us everything that the barmaid told him after he gave her a sawbuck tip. He doesn’t say, “She told me more, too, but I didn’t realize how important it was till later,” and then hold back the information until the end of the story — if he does more than a few times, we start to get annoyed, and properly so. The author is diddling with us. She is creating more distance between us and the story by making the narrator an artificer, our enemy in the quest for information instead of our ally. The author who does this usually thinks she’s increasing the suspense. In fact, she’s weakening the suspense by decreasing the readers’ involvement with and trust in the narrator.

That was only a mild example. You’ve seen worse — and if you’ll think back to your response at the time, you’ll realize how annoying it was, how very ineffective and distancing. For instance, instead of saying “She told me more, too” at the time of the interview, some authors don’t even say that much — they just have the character remember at some key moment later on, “I thought back and remembered something else the barmaid said, something that didn’t seem important to me at the time.” Now, if what he remembers is something he told us she said at the time, that’s perfectly fair; but if this is new information to the reader, we have a right to feel that we’ve been improperly deceived.

The worst case is when the first-person narrator refuses to tell us something that he himself did. Some of the best writers have done this, of course, but that doesn’t make it any less faulty. Here’s a passage near the end of a mystery novel:

Everything was clear in my mind now. Only a few things remained to be done, to set things up properly. I called Jim and asked him to make a couple of phone calls, and I stopped at the Seven-Eleven to buy a simple household article. Then I drove to Maynard’s mansion and rang the doorbell. Everyone would be there that night, I knew.

Until now the narrator has been telling us everything he did at the time he did it. Now, though, he is deliberately withholding information about what he did — who it was that he asked Jim to call and what simple household article he bought. Now, if it has been established that the narrator is consciously writing a mystery story (as with Archie Goodwin in Rex Stout’s mysteries), the n the character is perfectly justified in violating the convention of telling what the narrator knew at the time he knew it. But if the narrator has not been established as a mystery writer, this technique will violate his character and introduce falseness into the tale.

The fact that the narrator is telling the story at all makes it obvious that whatever risks she went through in the course of the story, she lived through it, so putting the narrator in jeopardy of death won’t be terribly convincing. There are other kinds of jeopardy, though, that can still work fine. While the first-person narrator can’t die, that doesn’t mean that terrible, irrevocable things can’t happen to her. In Stephen King’s Misery, one of the horrors of the book is that the narrator, though he obviously survived, still lost limbs and other appendages to his mad captor’s blade. When his captor threatened to do awful things to him, we knew that those awful things could actually happen; the jeopardy was quite convincing.

LAPSES

All these drawbacks to first-person narrative are problems that arise when you handle first person well. Alack, I’m forced to tell you the sad truth that first person is very difficult. Though first person is usually the first choice of the novice storyteller, since it seems so simple and natural, it is considerably harder to handle well than third person, so that the novice usually betrays himself.

Where do the mistakes come? Most commonly, the novice writer inadvertently confesses in the first page that the first-person narrator is a fraud, that he is merely a mask behind which an incompetent writer is trying to hide:

I watched Nora from across the room, the way her hands danced in the air like mad ballerinas, graceful and yet far too busy. She was upset, worried about the upcoming deal. The people who tried to converse with her were all so boring, their talk so petty; yet she tried to act as if she were interested, even excited about the subject at hand, so that they would never guess how tense she was. She thought back to how things began, back in Rotterdam in the last years before she met Pete and her life dissolved in ruins …

I don’t need to go on, do I? There is no way in the world that the first-person narrator can possibly know what is worrying Nora, or her motives as she converses with other people. Still, he might be merely guessing at her thoughts or motives — until we get to the last sentence, where he gets inside her head for a flashback. This is simply impossible — it is a technique of third-person narrative, one which is completely unavailable to first-person narrators unless they happen to have supernatural powers. Yet you would be amazed how many young writers make this mistake.

The first-person flaws in this next example are more subtle:

I awoke with a brutal headache. My hand brushed a paper off my pillow as I reached across the bed. I opened my eyes and winced against the pain of the sunlight streaming through the window. I was overcome with a sense of terrible loss; grief streamed through me again. I got up and staggered to the bathroom, each step like a knife through my head. I took the aspirin bottle off the shelf, turned it upside down. I turned on the water and got into the shower. It beat down on my head, streamed down my face, rivulets pouring down my body, cleansing me. I toweled myself roughly, then dressed in the same clothes I had dropped on the bathroom floor. Grief was all I could think of, grief so deep I felt nauseated. There was nothing to eat in the kitchen except peanut butter, graham crackers, and baking soda. I put a teaspoon full of baking soda in a glass of water and drank it down.

The flaw here isn’t that the passage is cold and melodramatic by turns — though of course it is. The flaw is that the first-person narrator is watching himself as if from a distance, not seeing inside his own head at all. He sees what he does, but never why. We watch him as if through a camera — but since he is the narrator, he wouldn’t watch himself do these things, he would remember them from the inside.

He didn’t observe these actions when they were going on, he performed them. Yet we are given no clue about what any of his actions mean. He might be hung over, but he might also be sick. And why are we told so much about the shower? What does the shower mean? Why does it matter? It seems like any other shower. We all get wet in the shower. We all have the water beat on our heads and stream down our faces; the whole point of showers is for them to cleanse us. There is no reason for us to be shown this particular shower, because it is no different from any other shower, and the narrator has given us no reason to think it means more than usual.

In fact, if a friend of yours were telling you a story, and he got off onto a tangent about his shower — “the water felt so good beating down on my head, streaming down my body, cleansing me” — wouldn’t you tell him to forget about the stupid shower and get on with the story? Of course you would. So why should the reader, who is not your friend (and unlikely to become one, if you write like this), put up with such irrelevant nonsense?

The only seeming exceptions are the two melodramatic sentences about strong emotion: “I was overcome …” and “Grief was all I could think of …” Yet even here, we are not told what he is grieving about. So this barely qualifies as being inside the narrator’ s head. Instead we are given abstract labels for emotions, not the experience of those emotions or the reasons why the narrator feels them.

If there is any point to using a first-person narrator, it is in order to experience everything through his perceptions, colored by his attitudes, driven by his motives — yet we got nothing of that in this sample. This supposedly first-person account is as impersonal as a phone book. It is also exactly what a majority of novices do when writing first-person accounts.

Here is the same passage told more as a real person might tell of it:

I woke that morning with a brutal headache. I reached out for Nora, as usual, but the bed was empty. Just a piece of paper, which I brushed off my pillow, not caring to know what the note said. It wasn’t from her. She hadn’t been here for days. Months. When would I stop reaching for her? On my deathbed would I expect to find her there, and once again be disappointed? No, maybe on my deathbed she’d be there, watching me so she could enjoy the process, the bitch.

I opened my eyes and regretted it at once — sunlight streaming through the window is never kind to a man with a hangover like the one I had. I got up and staggered to the bathroom, each step like a knife through my head. The shower was too cold, then too hot, and they don’t make a brand of soap that could have made me feel clean. The aspirin bottle was empty, of course, but it didn’t matter — there weren’t enough aspirin in the world to deal with a headache like mine.

I toweled myself roughly, punishing myself for being the kind of jerk who has to wake up alone. Then I got dressed. I wasn’t completely uncivilized — I thought of putting on clean clothes. But it wasn’t worth the effort. I put on the same clothes I had dropped on the bathroom floor.

There was nothing to eat in the kitchen except peanut butter, graham crackers, and baking soda. The peanut butter and graham crackers made me want to puke. I put a teaspoon full of baking soda in a glass of water and drank it down. Turned out even worse than I expected. I went back into the bathroom and threw up. Oh what a beautiful morning.

This version, while it still doesn’t tell us why Nora left, at least gives us more reason to care about what’s going on. We aren’t seeing the narrator from the outside, we’re watching him from the inside — which is exactly what first-person narration is supposed to do.

Note that we get characterization this time, which was almost entirely missing from the first version of this passage. We know why his hand brushes the pillow; we know why he doesn’t pick up the note. We know how he feels about Nora — not just nebulous and melodramatic feelings of grief, but clear, specific attitudes and emotions. He doesn’t describe the shower, he responds to it — an attitude, not a photograph. We know why he decides to wear the dirty clothes from the day before.

Your first-person narrator might be the kind of person who doesn’t easily confess his motives or his feelings. Of course, in that case one wonders why he would write the story at all, or why the author would be so self-destructive as to attempt to write a first-person story told by a taciturn character. If for some reason you do want to write a story told by such a character, even he would not write the first version of this passage. If he is not in the mood for confession, he would not describe his morning. In particular, he would not confess to such things as putting on dirty clothes, nor would he describe something as private as a shower.

If the first-person narrator doesn’t want to confess anything personal, that is also an attitude and will show up in his writing:

Everything that happened to me this morning? All right, I woke up hung-over and there wasn’t any aspirin in the cabinet. I tried to settle my stomach with baking soda and ended up puking. I put on dirty clothes and went outside and spent the rest of the morning yelling obscenities at passing drivers and kicking dogs and little children. I ate lunch at Mc-Donald’s and didn’t throw away my trash or stack my tray. That took me up till noon. Is that what you wanted to know?

First-person narration must reveal the narrator’s character or it isn’t worth doing. The narrator must be the kind of person who would tell the tale, and her motives and attitudes must show up in the story. If you find that you can’t do this, then you have three choices: You can admit that first-person narrative isn’t going to work in this story, and switch to third person; invent your first-person character and create her voice by discovering her attitudes, motives, expectations, and past; or experiment with other first-person narrators until you find one whose character you can create.