Who Gets to Know What? Narrative Point of View and the Control of Information

Editorial image generated by author.

First person, second person and third person can look like surprisingly small grammatical choices. I opened the door. You opened the door. She opened the door. The action has not changed, and neither has the door. Yet change the narrative point of view and almost everything surrounding that action may change with it: what the reader knows, whose interpretation accompanies what they see, which information can plausibly be withheld and even what kind of surprise the story can produce.

Narrative perspective is less like choosing a camera angle than designing an information system. A mystery changes if the narrator knows the solution. A tragedy changes if the reader knows something the protagonist does not. A character study changes when we hear someone's rationalisations from inside their own head. A plot twist may depend on the narrator being unable to tell us something, choosing not to tell us, or telling us everything necessary while allowing us to draw the wrong conclusion.

The central question is therefore not only who is telling the story? It is who gets to know what?

Spoiler note: This essay contains a clearly marked major spoiler for Agatha Christie's The Murder of Roger Ackroyd. That section can be skipped without losing the wider argument.

Knowledge, Disclosure and Inference

Three separate mechanisms are often bundled together when we talk about narrative point of view. The first is knowledge: what can the narrator or viewpoint character know? A detective cannot normally report a conversation taking place elsewhere, while a child may witness something without understanding its significance. The second is disclosure: of everything the narrator does know, what does the narration actually tell us? Knowledge does not guarantee revelation. Information can be omitted, postponed, compressed or distorted.

The third is inference: what does the reader construct from what has been disclosed? Fiction could barely function without this step. Readers continuously supply chronology, causation, relationships, motives and physical continuity that the prose does not need to state explicitly. Most of the time authors rely on that cooperation to keep a story moving. Occasionally they use the same machinery against us.

Different narrative techniques place the gaps in different locations. First-person narration often restricts knowledge. An unreliable narrator can complicate disclosure or interpretation. Some of the cleanest misdirection leaves both knowledge and disclosure intact and lets the reader make the mistake. First, second and third person are therefore not simply alternative voices; they help determine where ignorance can plausibly exist.

First Person and the Unreliable Narrator

First person comes with an obvious restriction: the narrator cannot normally tell us what they do not know. That limitation is one of its greatest strengths. A detective entering a room does not know who committed the murder, and neither do we. A character encountering something inexplicable cannot immediately provide the answer unless they already possess it. Discovery belongs to protagonist and reader at roughly the same time.

Jim Butcher's Dresden Files, for example, gives us its supernatural world almost entirely through Harry Dresden. We know a great deal about what Harry understands and much less about what lies beyond his experience. When he learns something new about the supernatural world, ours expands with his. When he misjudges someone, we are liable to follow his interpretation until contrary evidence accumulates.

The series also creates an information asymmetry between Harry, the reader and everyone who has to judge him from the outside. In my discussion of how the Wardens' view of Harry changes in Dead Beat, this produces an almost inverted effect. His standing among them improves just as the reader acquires additional reasons to understand why their earlier suspicions were not entirely irrational. They see more of Harry's actions; we know more about the compromises and temptations behind them.

First person also makes observation difficult to separate from interpretation. Suppose a narrator tells us that Anna looked furious. Perhaps she did. Perhaps she was exhausted. Perhaps the narrator feels guilty and reads accusation into an otherwise neutral expression. A sentence that appears to describe Anna may tell us more about the person looking at her.

This is where unreliable narration becomes broader than deliberate lying. People misunderstand their own lives without consciously falsifying them. We invent flattering explanations for our behaviour, remember arguments differently from the people we argued with, misread motives and avoid conclusions we would rather not reach. A narrator can reproduce those habits perfectly sincerely.

The reader may then learn to perform two readings at once: the world as the narrator understands it and the world that seems to exist around them. A narrator who repeatedly assures us that everybody respects him may gradually communicate the opposite. A character who insists that a past event no longer matters can reveal its continuing importance through the effort spent dismissing it. Nothing has to be factually false. The narrator may simply be the last person in the book to understand their own story.

Third Person: Limited and Omniscient

Third person can create a superficial impression of greater objectivity because the narrator no longer says I. In close third-person narration, however, the informational boundaries may be almost identical to first person. We still see what one character notices, know what they know and encounter other people through their assumptions. Replacing I with she does not automatically move us outside the character's mind.

What limited third person makes particularly easy is movement between several such minds. Behaviour that seemed arrogant from one viewpoint may appear defensive from another. An inexplicable decision can become entirely rational once we see the information on which it was based. Neither character needs to be unusually unreliable. Their perspectives are incomplete because human perspectives generally are.

This makes multiple limited viewpoints useful for stories built around conflict. Two people may understand the same situation reasonably well and still reach incompatible conclusions because they possess different information, histories and incentives. The reader can then gain a privilege denied to either of them: an understanding of both positions without requiring the story to declare one consciousness objectively correct.

Omniscient narration removes many of these natural boundaries. The narration can enter several minds, describe events no protagonist witnesses and supply history or context unavailable to anyone currently involved. More information does not necessarily reduce suspense; it can change its form. If we do not know what waits behind a door, the question is what is there? If the narrator has already shown us the assassin waiting on the other side, the question becomes will the protagonist realise before opening it? Mystery has become dramatic irony, but uncertainty remains.

Omniscience can also expose situations in which everybody inside the story is wrong. A negotiation can be shown from above, with each participant misunderstanding what the others want. The reader sees a system that none of its participants can see whole.

What changes is the meaning of silence. A first-person narrator has a straightforward reason not to tell us what the antagonist is planning: they do not know. Once a narrative has established that it can move freely between minds, ignorance no longer supplies the same explanation. An omniscient narrator need not reveal everything, but selective silence becomes a structural choice. If a novel enters everyone's thoughts except those of the one person whose thoughts would solve the mystery, readers may begin to notice the shape of the absence.

Second-Person Narration and the Grammar of Address

Second person is rare enough in novels that encountering it can initially feel almost grammatically wrong. Charles Stross uses it for all three main perspectives of Halting State, in a style contemporary reviewers compared to the authoritative second-person language of video-game instructions.

First person gives us I did this. Third person gives us she did this. Second person gives us you did this. That does not mean the novel necessarily pretends that the literal reader is the protagonist. In Halting State, “you” refers to identifiable fictional characters. But we are still asked to process those characters through the grammar of address: the pronoun occupies a position normally reserved for someone being spoken to.

After a few pages the oddity can recede, only to return when “you” makes a decision the actual reader would never make. The distance between reader and protagonist has not vanished; it has been arranged differently.

Second person can become stranger still because the identity of “you” need not be the protagonist. A narrator might address another character, a dead relative, a younger version of themselves or some unidentified listener. The useful question is therefore not merely why an author has chosen second person, but whom the narrative imagines itself addressing. In some stories, the answer becomes part of the structure.

When the Narrator Knows More Than They Tell Us

Restricted knowledge is easy to accept. A character cannot report what they never witnessed. More dangerous territory appears when the narrator knows something, the reader would consider it extremely relevant, and the narration declines to mention it. Here the viewpoint itself becomes the hiding place.

Major spoiler warning: The Murder of Roger Ackroyd. The following paragraphs reveal the central solution to Agatha Christie's novel. If you intend to read it without knowing the answer, skip to The Reader Supplies the Missing Premise.

Christie's famous trick works because Dr James Sheppard is not merely the first-person narrator of the murder investigation. He is the murderer.

That creates a technical problem unlike the ordinary limitations of first person. Sheppard knows exactly what happened because he was there. Ignorance cannot protect the secret, so Christie instead controls disclosure. His account relies on omission, compression and carefully bounded description. He does not spend the novel falsely insisting that somebody else killed Roger Ackroyd; he provides an account that can be read as complete while excluding precisely the actions that would change its meaning.

The trick exploits an assumption readers grant a first-person narrator almost automatically: if the narrator knows something this important about their own actions, surely the narrative will mention it. The official Agatha Christie site's discussion of the novel's fair-play problem makes much the same distinction. Sheppard tells the case in the first person but leaves out his criminal actions; whether that constitutes cheating or legitimate misdirection depends on what obligations we think narration carries.

The ingenuity lies not only in concealing the murderer's identity. Christie turns the reader's assumptions about narration into part of the concealment. We interpret not just the sentences on the page but the absence of sentences we assume would have been there if something important had happened.

The Reader Supplies the Missing Premise

There is an even cleaner form of misdirection in which the narrator neither lies nor conspicuously conceals anything. The reader supplies the error.

A scene might tell us that someone enters a dark room, speaks and later leaves. Nothing necessarily establishes that another person answered, yet the construction may encourage us to remember a conversation having taken place. Two adjacent scenes may likewise invite us to assume they occur simultaneously even though the text never establishes their chronology.

These tricks work because reading always involves reconstruction. A novel cannot specify every causal connection, spatial relationship or identity without becoming unreadable. We infer them automatically, usually correctly. That efficiency also gives the writer somewhere to hide information.

The prose states X. We silently add Y. Later the story reveals that X was perfectly accurate and Y was something we contributed ourselves. Some twists feel especially satisfying for precisely this reason: the revelation does not replace a fact we were previously given but an assumption we had inserted between the facts.

The knowledge–disclosure–inference framework is clearest here. The narrator may know the truth and disclose every necessary fact. The misdirection occurs entirely in the reader's reconstruction.

When Structure Becomes Evidence

Once a novel uses several viewpoint characters, the distribution of those viewpoints can communicate information of its own. Imagine a mystery with six central characters. Five receive viewpoint chapters and one never does. There may be an innocent structural reason, but an experienced reader may start wondering why entering the sixth character's mind would be inconvenient.

Writers can exploit that expectation in either direction. A suspicious absence may be deliberate misdirection, while a character might receive a carefully selected viewpoint scene precisely because allowing us into their head appears to prove that they have nothing to hide. The author is no longer managing only the contents of scenes but the reader's awareness that those scenes have been selected.

Sequencing creates another layer. Character A can discover something in chapter ten that Character B will not learn until chapter twenty, leaving the reader with dangerous knowledge for ten chapters. Or the author can switch viewpoints immediately after Character A learns the truth but before revealing it to us. The chapter break itself then controls disclosure. What looks like novelistic architecture has become part of the information system.

Time can do something similar. Many first-person past-tense narratives imply a later version of the character looking back on events, even when the novel asks us not to dwell on the consequences of that arrangement. Within the remembered scene, the protagonist does not know what will happen next; the implied narrating self may already know how everything ends.

Most fiction handles this quietly. We accept that a retrospective narrator can reproduce earlier uncertainty without constantly announcing future consequences. But a writer can choose to exploit the gap between experiencing self and narrating self. A later narrator can mock earlier assumptions, foreshadow disasters, question their own memory or deliberately preserve the interpretation they held at the time. Even the existence of a retrospective narrator may leak information: someone calmly recounting apparently fatal events decades later has apparently survived them, unless the story eventually gives us reason to reconsider who has actually been speaking.

The relevant point is not that every past-tense first-person novel contains a literal future narrator sitting down to write a memoir. It is that the consciousness through which an event is experienced and the voice through which it reaches us need not possess exactly the same information.

Point of View Is Part of the Plot

Choosing a narrative point of view is difficult to separate from deciding what sort of story can be told through it. A story built around discovery benefits from having a natural home for ignorance. Dramatic irony requires giving the reader information someone inside the story lacks. A story about self-deception can place us inside a character's explanations long enough for the gap between those explanations and reality to become visible. Conflicting viewpoints can allow several incompatible interpretations to remain reasonable at once.

The constraint is often more useful than freedom. A political epic restricted to someone who never leaves a provincial village poses one set of problems; a murder mystery narrated from inside the murderer's consciousness poses another. Remove the restriction and some narrative difficulties disappear, but so do some of the possibilities created by having to work around them.

The choice of narrator determines three separate things: what can be known, what will be disclosed and what the reader must reconstruct. Mysteries can hide information in any of those gaps, but so can tragedies, romances, fantasies and political novels. Suspense, misunderstanding and revelation all depend to some degree on deciding who possesses which part of the picture.

Point of view does not simply determine the voice in which a story is told. It determines where ignorance is allowed to live.