Perspective is one of the most important choices a writer can make in fiction. It’s also a complex concept with a lot of parts. In Part 1 of this discussion, we broke down the basic grammatical foundation for perspective, which accounts for the common 1st and 3rd person terms as well as past and present. In Part 2, we worked on the limitations we might impose on ourselves with the concepts of limited, objective, and omniscient perspective. Today we’re talking about various structures that you might employ to tell your story. These are going to be, again, ways that you position your narrator relative to your imagined audience—not, necessarily, your real audience, although the imagined audience may include that as well.

The first question here is: Whose story is it, anyway?

Character Choices

Who will tell your story?

Part of answering this question is purely pragmatic. Who experiences the action? Who has the necessary knowledge to tell the story with sufficient information?

Generally, this is going to be the main character, or a small set of main characters that you can bounce between (whether limited or omniscient). But that’s not always the case. Sometimes a story is more interesting when seen through the eyes of someone related but not the main character: consider the use of the Genie in both Disney versions of Aladdin as a narrator, someone who has all the knowledge and power but is, fundamentally, an object or catalyst to the story rather than the main character.

There’s also a question of whose voice matters most when considering whose story this is, and consequently who you will focus on in your narration, whether as the actual narrator or as the narrator’s focus. This is particularly important when you are dealing with marginalized characters.

Consider, for instance, the problematic example of The Help (disclaimer: I only saw the movie, ). The entire premise of the story is that, for some reason, a white woman is more capable of telling the stories of the Black women who work as domestic servants, and thus the narration is focused largely on the white woman’s actions, feelings, and experiences. This is her story of collecting the Black women’s stories, not actually the Black women’s stories. In some ways this made The Help more palatable to white audiences, but in others it also diminished the voices of the Black women whose stories are being collected. This perspective-taking is a choice. The author decided that instead of simply telling the Black women’s stories in a compilation, or having a Black woman collect and narrate them, we would see them through the lens of our white journalist character. In part this identifies the film as targeted toward a white audience, rather than a Black audience.

In the same way, your choice of perspective character may mark your own story for a particular audience. This is not to say that audiences who don’t directly identify with your narrator or perspective character will not read your story. I am a cisgender, heterosexual woman and I very much enjoyed reading Reforged, which takes the perspective of a gay man and explores his internal struggles with duty, love, and desire. But at the same time, I can recognize that I’m probably not the intended audience of Reforged precisely because the main character doesn’t resemble me in most ways.

So one consideration when thinking about whose story it is to tell is whose story it is to hear. Who do you want to tell the story relates directly to who you want to hear the story.

Take another film example: the animated series Avatar: The Last Airbender. Whose story is it? It’s actually Katara’s story, not Aang’s, at least as far as who is telling it. Although of course there are scenes that are not focused on Katara or where Katara is absent, the opening voiceover makes it very clear that Katara is our narrator, and we are seeing the avatar through her understanding. We are introduced to characters and parts of the world in the order that Katara knows them, not Aang. Why does Katara get to tell this story? It’s a choice, but I think part of it lies with her being the most reliable and relatable character to narrate. But also because her journey mirrors the audience’s journey: the audience doesn’t know much about the avatar cycle, but Katara is able to explain it to the outsider audience because of her own skills and cultural stories and because of her access to the main players in the story (including herself).

So another consideration when thinking about which character to use as a narrator or a perspective character is who will be able to explain the story to your audience. This is the reason why isekai stories work so well in anime: the main character is, like the audience, dropped into an unfamiliar world and must learn its rules. Portal fantasy (like the Narnia books) works the same way. Academy stories tend to work the same way, because they are stories explicitly about learning the rules of a world and one’s place in those rules. In those genres, the perspective character is pretty obvious: it’s the new guy. That’s how you will be able to teach your world to your reader, who will learn alongside the new guy.

Ultimately, when choosing whose head to be in or even just who to follow (in objective perspectives), you’re looking for the person who can relate to your imagined audience in some way, who has access to the necessary information and scenes, and whose experience will show the reader the things you want them to see.

But there are some other fun things you can do with perspective too.

Frames and Letters

In the 20th and 21st centuries, readers have generally come to accept the admittedly strange fiction of some kind of invisible narrator just telling a story, or even the first-person narrator just narrating to… whom? Think about it. Pick up any three books published in the last, say, 50 years. I’ll wait. At least one of them, I’m sure, doesn’t really explain how the story is getting into your hands as a reader, or how it’s getting from the head of the narrator to the page, no matter what kind of narrator you have. Who is that objective 3rd person past tense narrator? Some random camera that just happens to observe the action?

Of course we can get into some deep philosophy about living in a world where we expect our actions to be recorded by a panoply of cameras, many operating entirely without human input, so that such narration makes sense to us, but that’s a post for another day.

Fundamentally, we have just come to accept that the price of a story, part of the all-important suspension of disbelief, is accepting that narration happens. Ours is not to question how.

But that wasn’t always the reader’s assumption. Take a look instead at early novels in the 18th century and going well into the 19th century. The vast majority of these will have some kind of fictive device in them that makes it possible for the reader to read the story. This is either structural, as in the case of diaries and epistolary novels, or it is a frame narrative. It can be both!

Consider Frankenstein, which is possibly the most brilliant uses of a frame narrative to come out of the 18th century. There are, structurally, three narrators, but they are all filtered through one. We have the sailor writing to his sister—presumably, then, we get to take the role of the sister as readers, or else the sister has shared this letter with us. This is the perspective through which the whole novel is filtered, a sort of unifying voice as well as a contrivance to explain how and why this story is being told.

The second perspective is that of Frankenstein, telling his saga to the sailor in exchange for refuge on the ship among the ice floes, an old device that is likewise used in The Odyssey when the story starts with Odysseus thanking his host for shelter.

The third, then, is the creature’s own voice, still filtered, though, through the sailor’s voice in his letter to his sister. The tone shifts considerably, and we are given a very different impression of this eloquent creature than Frankenstein had related.

So whose story is Frankenstein? Is it the creature’s? The eponymous undergraduate? The sailor?

Simply, yes. But we are invited, along with the sailor’s sister and the sailor himself, to consider the various perspectives offered. And the whole thing is presented in a frame narrative that explains how we got the story.

I love a good frame narrative. They’re brilliant devices when played well, but of course you don’t need one. Still, think about the frames that do work well.

One of the most brilliant frames in 20th century fiction is that of The Princess Bride. If you have only seen the movie, then you do have a sense of the skillful frame narrative in how we are presented with a grandfather reading a story to his sick grandson, complete with interruptions and discussion of the actions in the story. But if you’ve read the book, then you know the frame narrative in the book is so much more complex. In fact, if you didn’t know the conceit, as many readers didn’t, and struggled with “Why can’t I find an unabridged version?”, you’ve run into the complexity of the frame narrative in The Princess Bride. The frame narrative there is not just a frame in the way that Frankenstein‘s letter writing is: it’s a complete apparatus that affects the way the text even appears on the page. We are presented with the story of a man editing a book he was read, with considerable omissions and revisions, as a child, and attempting to amend the text accordingly to reconstruct the version he remembered rather than the version that, presumably, was on the page. The frame here is literally a frame, in that it is presented as footnotes, editorial interruptions, and other errata on the page itself. It frames the story in the same way that a scholarly edition of a classic is framed by scholarly interpretations.

But, returning to the historical use of frames, there is a common structure that you might consider, which is the epistolary novel.

True epistolary novels are rare now, as they’re seen as outmoded and contrived, but they do exist and are still very much an option, and of course some of the developments that come from the epistolary novel are very much still options.

The epistolary novel is a story told through letters. This gives us obvious perspective characters and a reason why the story is being told, but it also positions the reader as something of a voyeur. This is perhaps the oldest form of the true novel in English and was very popular in the 18th century. Technically, we can consider Frankenstein to be epistolary, but it’s really just one long letter rather than exchanged letters in the ways that the true epistolary novel was. You see, the trope fell apart pretty quickly, because there are some stories that simply don’t’ get to be told in letters only. My interest here is in some of the ways the trope fell apart, because those are great sources for inspiration for modern frame narratives that still work today.

One great example of the trope falling apart is an early American novel, Charlotte Temple. Charlotte Temple is narrated in large part through the perspective of a busybody neighbor who heard the whole thing. The titular character, Charlotte, is a pretty standard heroine by 18th century standards and does the things these heroines are supposed to do, including writing letters. But her letters are being intercepted by the man who leads to her “downfall,” explaining why it’s not a typical epistolary novel—and also an important part of the story, because his isolating her from family and friends is part of how he preys on her. Using the neighbor as a narrator is a brilliant device here; it gives us a plausible reason for the story to be told, allows a tearful death scene for our titular heroine (a highly prized thing for readers at the time), and gives us some room for errors in the story or uncertainty.

Another great example of the trope falling apart is Dracula. In fact, it’s my favorite novel precisely for this reason. Dracula includes letters, but also diaries, newspaper clippings, and any number of other bits and details that can be transcribed into a novel. But how did they get there? Mina! Mina is presented as trying to learn how to be a perfect reporter, to be able to transcribe conversations perfectly by shorthand and to type well, and finally as compiling the whole story as an archivist of the whole episode. We don’t get this explanation until the very end of the novel, but the pieces necessary to make it make sense are laid down from the first chapters.

It is precisely these frame narratives that are impossible to translate to the screen from a novel. The Princess Bride is the most successful I have ever seen, but it does so by completely changing the narrative. The film version shows us the experience of the child being told the story orally, while the novel version shows us the experience of the adult piecing together his childhood experience while being confronted with a text that doesn’t quite match that experience.

So what are some other ways you might use a frame or a quasi-epistolary style to tell your story?

You can, of course, use a narrator who is telling the story to a specific person or group. You can also use the “found narrative” trope, in which you present your story as someone’s diary or memoir that was found by someone else, perhaps hundreds of years later and then translated, or perhaps by a family member. You can use a reporter (returning, for instance, to The Help), researcher, or archivist.

The key is to think about not only who is telling the story, but why and to whom. Is this a report to a supervisor explaining why things went weird? Is this a letter to a dear friend? Is this a last will and testament turned confession? Is this a court case? Think about all the ways and times we use stories and draw inspiration from those. A breezy romantic comedy told as a wedding toast to the couple, for instance, or a horror story told through a doctor’s notes about a strange patient. These considerations can be creative and generative as much as they honor the narrative traditions of storytellers past.

Am I saying you need to have a frame narrative or try an epistolary style? That you need to have some premise for why and how this story is being told, and by who to whom?

Not at all. You are perfectly free to rely on your reader’s assumption that stories can happen in a void without explanation of those sorts. Readers are happy to accept it, and that’s the dominant storytelling mode these days. But if you’re struggling with perspective, you might try adding a frame.

Frames can add depth. They can add relationships (for instance, a father telling stories to his child), urgency (a deathbed confession), closure (a court case), and any number of other useful dimensions to your text. They can also resolve the earlier questions of grammar and perspective, because now you have a reason to choose what you choose. You’ll have guidance for head-hopping if it needs to happen.

But they aren’t strictly necessary. Like so many things in writing, they’re a tool you can use to present your story.

Just like it’s popular these days to hang art on the wall as a bare canvas or photos on a string with clothespins, you can tell your story without embellishment, framing, or additional context. But in the same way that a frame can draw attention to a painting or help it complement a room better, a frame can help orient your reader and situate your story in a context.

How you hang art on your wall is a purely stylistic choice, based on the kind of aesthetic you want in the space and the art you are displaying. It’s the same way with perspective in your writing: what story are you telling and how do you want to situate it for your reader?

Last week, we talked about the grammatical dimensions of picking a perspective (also called point of view or POV) for your story.

For a quick recap, you basically get two viable options for “person” and two for “tense”: that is, you can write in 1st or 3rd person (I or they), and you can write in past or present. This gives you a functional grid:

This is really the first consideration that an author has to make when writing, because it will determine the actual shape and grammar of every single sentence in the story. Not because the subject of every sentence has to be the narrator in a 1st person narrative, but because we have to consider the subject of every sentence relative to the narrator, whether that’s 1st, 2nd, or 3rd person.

The second consideration, though, is how broad that perspective will be. We have to add another axis to our chart: limited to omniscient. We may also include objective, perhaps as a true neutral, but that’s where our spectrum starts to fall apart (and I always forget that objective exists because in my mind it’s sort of a mode of omniscient–more on that later)

Generally these are only discussed in terms of 3rd person, but they are technically relevant in every grammatical position. These are not grammatical choices, but rather conceptual choices that will determine the scope of what you can and cannot say. They’re self-imposed limitations on your storytelling, but self-imposed limitations matter a lot.

Last time we talked about how comparisons to film and games, which are very common in discussing the craft of writing, don’t really hold up when discussing something that is fundamentally grammatical. The same thing holds true here to some degree because every medium and mode have their own strengths and weaknesses (or affordances and constraints if you want to be very technical), although here the metaphors are a little more useful. Consider, for instance, how it might break your immersion if a film director made most of the scenes in a film dark and moody with a lot of blue tones, but suddenly one scene is filled with as much color as a Kyary Pamyu Pamyu music video. That effect might be used on purpose, but that director had better have a darn good reason for the sudden splash! Or consider how confused a game player might be if a powerup suddenly looked different on the next level. As writers, we also strive for some internal consistency (except, of course, when we intentionally flout it) and our readers are trained by their own genre and reading knowledge to expect it from us.

So let’s dig back into our own reading training and review what limited, omniscient, and objective mean. In general these three terms will be used to describe 3rd person narrators, but they can technically be used for any combination.

The textbook definition is that limited perspective means that the reader only gets to see what one character sees. There’s a caveat here that it’s what one character sees at a time, because many very popular books of course have more than one perspective character, whether in 3rd or 1st person (for instance, Children of Blood and Bone). In general, first person narration sort of has to be limited, unless you are actually narrating from an omniscient character’s perspective (like 2nd person, rare and difficult to make work well, but technically possible). So for most people, thinking of limited perspective as “first person narration, but in 3rd person grammar” is going to work pretty well. It’s sort of like over-the-shoulder view in a video game. You generally see what the character sees, experience what the character experiences, but at some remove so you can watch them do it.

Omniscient perspectives are basically the opposite of limited perspectives. Omniscient literally means “all-knowing” (omni for all, scient as the same root word as science). Where, in a limited perspective, we only get to see and feel what a certain character sees and feels, omniscient allows us to see and feel what any character is seeing and feeling as relevant. This doesn’t mean you as the writer get to “head hop” without regard to narrative structure, but it does mean that you get to comment on the various ways that various characters might perceive the same thing. Omniscient narrators cover a huge range of tones and styles, ranging from having dialogue with characters’s innermost thoughts and even addressing the reader directly to being nearly invisible. In general, if you want an omniscient view, you need to use 3rd person, but that’s not an absolute rule (go ahead, write that experimental novel in 2nd person omniscient POV. Do it).

Objective perspectives are entirely uninterested in characters’ innermost thoughts and only report to the reader what a camera might capture. These are, likewise, almost universally 3rd person, although it could be done to have a Forrest Gump-like character who is not directly involved in the events you’re relating but somehow a perfect observer without commentary. In a sense, though, that’s the fiction we use with all 3rd person narrators anyway, isn’t it? That the narrator is somehow not part of any of the action but still able to relate it perfectly.

Let’s see if we can make an example. I’m going to narrate the exact same sequence of actions from each of these. Let’s tell a simple scene of a young adult celebrating a birthday alone. We’ll keep it 3rd person, past tense and just manipulate the limited/omniscient/objective variable.

Objective POV

Megan’s keys jingled as she shoved her apartment door open with her shoulder. She wore a white button-down blouse and a black pencil skirt. Her hair was brown and straight.

She closed the door behind her before she flipped on the light. The overhead light flickered. She tossed the keys on an Ikea stand by the door that had unopened mail stacked on top. The top letter was clearly marked as a final notice for a medical bill by the red stripe in the envelope window.

Megan opened the freezer. It only contained three things: two Marie Callender’s meals and a small boxed Carvel ice cream cake. She took the cake. She slid it out of its box and onto the counter and threw the box in an overflowing trash can. Then she took a spoon out of the sink full of dirty dishes, rinsed it off, and shoved it into the cake. “Happy birthday to me,” she said.

Notice how we get absolutely no commentary, but we still get lots of characterization. We can guess that Megan works in a customer-facing job, perhaps as a hotel clerk or a bank teller, based on her appearance. We can guess that her apartment isn’t very well maintained, so either she doesn’t care or (more likely) her landlord sucks. The medical bill tells us that she’s struggling with something financially and probably medically, too. There’s a lot here! But none of it is told directly by the narrator. The mode here feels almost photographic. This is how most films would introduce Megan. But let’s see what we can do with perspective to change the tone.

Limited POV

Megan struggled with her keys, but she struggled more with the door. It was always getting stuck. She shoved it with her shoulder. It figured that her birthday would suck, that she would have to work overtime and that the Karen in room 304 would be such a bitch.

She didn’t bother to change out of her work clothes. What was the point? She’d just be back in them in ten hours anyway. She closed the door and tossed her keys onto a stack of unopened bills from the hospital that was sitting on a cheap table. She turned on the light. It flickered. One more thing that didn’t work in this damn apartment. Fuck it. And fuck those bills. She wasn’t going to pay them.

The nearly-empty freezer was as depressing as the rest of the day, but she had been looking forward to that ice cream cake all day. It was small—all she could afford, really. That was fine. Perfect for just her to eat.

Megan glanced over the sink. There were dirty dishes there. Nearly every dish she owned. Oh well. She just rinsed off a spoon. That was all she needed.

“Happy birthday to me.”

We’ve got a very different tone here, because now we can hear more of Megan’s voice. She cusses a lot. She’s angry and tired and hungry. Our objective tone would have conveyed that eventually, of course, but this is just a clip.

The important part is we’re inside Megan’s head. We know not only that she has mounting medical bills, but we know how she feels about them and why she hasn’t opened them. We’re not inferring as much about her feelings.

But notice also we lost some of the visual aspect of the scene. This is more “telling” than the objective POV was “showing.” That’s not a bad thing. The question is: is it more important to convey the frustration that Megan feels about this moment in a close, intimate way, or is it more important to show how she feels alone by making the narration isolate her even from her own thoughts?

And we have a third option:

Omniscient POV

Megan struggled to open her door, the way that the five tenants before her had done. None of them had stayed long, and the landlord had done nothing but paint after each one left. Megan shoved the door open with her shoulder, an action so familiar now she didn’t even notice how it smarted. It definitely didn’t hurt as much as the guest in room 304’s insults today.

But even that was normal to Megan. She didn’t even bother changing out of her work clothes. Megan generally didn’t until it was time to hop in her unmade bed. She just tossed her keys on a cheap Ikea stand by the door and ignored the medical bills that were piled up there. She’d gotten so used to stacking them there unopened that she had not noticed the card from her aunt that arrived with them yesterday, which also lay unopened on the table.

Megan opened the freezer, which was nearly empty except for two Marie Callender’s microwave meals and a boxed Carvel ice cream cake. She pulled out the cake. She’d been looking forward to this all day. As she unboxed it on the counter, she realized that she didn’t have any clean dishes to eat it with. In her usual way, she just rinsed off the spoon she needed and stabbed it into the center of the cake.

“Happy birthday to me,” she said bitterly as her phone, left on silent, vainly tried to notify her that her aunt wanted to Facetime.

Oh, now we have a very different story! Now this is the story of a young person who feels alone, but who has at least one person who is trying to reach out to her, trying to remember her. The narrator tells the reader things that Megan doesn’t know and that a camera couldn’t show. That’s what makes this narrator omniscient.

Additional Considerations

One thing you probably noticed was that each iteration got longer. That’s not always going to be the case, but it is true that our omniscient narrator had a lot more details available to convey than our objective or limited narrators, at least in this story. And it is true that the objective narrator simply doesn’t get to make commentary, whereas the other two kinds of narrator do.

Personally, I love narrative commentary. For me, that’s a large part of the delight of a story, but that doesn’t mean I dislike objective narrators. Nor do I think any of these were easier to write.

The challenge of an objective narrator is in conveying the emotions and importance through describing actions and objects only. That requires a lot of control over the narration. You don’t get to say why the letters are unopened. You don’t get to say what Megan thinks about her landlord’s negligence, or even that the landlord is negligent. It’s the ultimate “show don’t tell” approach. And it’s hard.

The challenge of a limited narrator is balance. It’s tempting to do stream of consciousness in this style, and sometimes that’s what you want. But most of the time, it’s not. You have to balance external and internal description in limited narration. For me, switching between an internal and external space in a narration is a particular challenge—and a limited narrator is constantly juggling the two.

The challenge of an omniscient narrator is focus. Focus is built into the other two approaches, but an omniscient narrator has so much freedom that the author has to carefully weight what information out of literally all the available information is most important. Why does my omniscient narrator keep mentioning what Megan isn’t seeing? Obviously her aunt is going to play into the story, and less obviously I’m characterizing Megan as being basically blinded by her very reasonable frustration in this moment to the point that she’s ignoring the very thing she wants most, which is someone to show kindness to her and remember her. But too much head-hopping will confuse and frustrate your reader, who expects some sense of unity and purpose to everything you do as a writer. Many omniscient narrators, for this reason, wind up simply being multi-POV limited narrators (Terry Pratchett tended toward this mode).

So how are you going to pick which one to use?

Well, unlike the grammatical choice, this one doesn’t have to be made immediately upon writing your first sentence. The scope of your narrator’s knowledge, and consequently what you choose to show the reader when, can unfold as you work through your story and discover themes, motifs, character details, and other aspects of your story that will guide the choice. If you notice yourself tending toward objective, revise the rest to match; that’s probably the way you want to tell that story. If you notice yourself head-hopping, maybe you either want to limit to just one or two characters and need to rein in the hopping, or maybe that’s your narrative telling you it wants an omniscient narrator and you can lean into it, being deliberate about the hopping.

You’ll probably tend toward certain perspectives as a writer. That’s ok! While experimentation is vital to discovering your strengths and strengthening your weaknesses as a writer, you can also have a comfortable mode that you discover through experimentation. Perspective is likely to be part of that comfortable mode.

In the same way, some stories simply tell you what perspective they need. Next time we’ll talk about thematic elements and specific narrative structures such as epistolary and frame narratives, adding even more options and, hopefully, helpful heuristics to your POV choices.