Understanding Unreliable Narrators in Storytelling

[00:00:01]Every story with a narrator has a "narrative  voice." In simple terms, it's the tone and   approach taken by the narration - which is  basically all the parts of the story that   aren't dialogue or inner monologue. Scene-setting,  physical descriptions, the actions a character   takes - all these things make up the "narrative  voice" of a typical prose story. The simplest   thing to nail down about it is whether it's in  first, second or third person. A first-person   narrative voice talks in terms of "I" and "we" and  "me", taking the explicit role of a character in   the story retelling events that happened to them.  A second-person narrative voice is more rare,   talking in terms of "you" and "yours", taking the  role of an observer describing the actions of the   person they're dealing with. A third-person  narrative voice is probably the most common,   using "they" and "them" and "he" and "her" to  describe the actions of the characters from a   wholly externalized perspective. A second- or  third-person narration essentially implies the   existence of a narrator character, intangible  and unobservable, narrating the events of the   story and the thoughts and insights of its  characters from a separated standpoint. And   this is a very useful subdivision to draw,  but it is not where the conversation about   narrative voice ends. For one thing, these  implied spectator-mode narrator characters   have varying levels of omniscience - how much  they actually know about the story and can   communicate to the audience. "Limited" is probably  the narrowest field of view - the narrative voice   is limited to the thoughts and inner monologue  of one character at a time, but will limit its   narration of all other characters to the POV  character's own observations. For instance,   a third-person-limited narrator might volunteer  that the POV character is feeling tired or scared,   but when discussing a different character  the narrator will only volunteer that the POV   character thinks they're putting on a brave face.  "Objective" is the narrative equivalent of a wide   shot. This narrator doesn't have close-in access  to anyone's thoughts and motivations, and will   instead simply narrate their actions and dialogue,  leaving the audience to infer how they're feeling   and why they're doing the things they're doing.  And finally, "omniscient" is a narrative voice   that knows what everyone is thinking and  feeling, and can freely switch perspectives   between different characters to show what  anyone and everyone is feeling about the story.  This might seem like a lot of trouble for  a character that doesn't even really exist.   Narration is invisible, the glue that holds prose  storytelling together because otherwise it would   just be scripted dialogue and stage directions.  Narration is how the story describes itself.   And because there's no way for language to be  completely neutral and objective, the choices   a writer makes in how a narrator - even a  third-person omniscient narrator - describes   the story they're dealing with implicitly creates  a characterization for that narrator. There's a   difference between saying a character "said"  something and saying they hissed or murmured   or managed or yelled or shrieked. They all  feel different and mean different things,   and imply different things about the character  they're describing - and thus indirectly imply   things about the narrator holding that opinion.  A supposedly distant narrator might describe a   character as beautiful or repulsive or  terrifying or irresistibly attractive   with the implication that these opinions  must be universally held because they're   being stated by a character that's supposed  to be neutral, despite the fact that these   are subjective assessments that would vary from  character to character and audience to audience.  If a writer isn't careful about this, the  narrator can end up as an unintentional   and invisible author self-insert, giving the  writer's unfiltered perspective on the story   they're telling, perhaps revealing implicit biases  or blind spots on the part of the storyteller.  But the thing is, this is a feature and not a bug.  A completely neutral narrator, if such a thing   could even exist, would be distant and sterile  and make the story needlessly difficult to parse.   The fact is, the narrator is a character and that  character has a voice, and while some stories make   that explicit by having the narrator be one of the  named characters recounting the story from their   perspective, some story formats make this less  obvious, and the characterization of the narrator   is really only demonstrated in the specific ways  they choose to describe the plot and characters.   The narrative voice has opinions and biases and  finds certain things funny or worth pointing out   while others might be glossed over as irrelevant.  In a medium with a narrator, the narrator becomes   a lens through which the audience experiences the  thoughts and actions of every other character.  Which means, on the most basic and technical  level, every narrator is an unreliable   narrator. A theoretically perfectly reliable  narrator would explain absolutely everything   that's happening so the audience is fully  briefed on the intricacies of the story,   which is impossible because at bare minimum  language itself isn't that precise, and, broadly,   it'd just take way too long. There'll always be  room for confusion and interpretation. But aside   for a certain baseline background radiation of  unreliability, there are many ways for a narrator   to be notably unreliable. It's generally expected  that a narrator or POV character is being honest   and correct about what's happening, at least  to the extent of their ability to understand   it - but what happens when that's not true? Now the convenient thing about this line of   questioning is it actually takes us out of the  prose-only zone. See, most mediums of storytelling   don't have narrators, because most of them don't  need narrators. In prose writing, the narrator   is the only way for the story to set the stage,  paint the picture of the setting for the audience,   tell us how the characters look and act and  feel and broadly communicate information to the   audience that the characters themselves wouldn't  immediately want to volunteer. If you want to   get across to your audience that a character  is a redhead with green eyes and freckles,   unless that character feels like saying those  things out loud, it's gonna have to come through   from the narrator. This stops being true as soon  as a visual element is introduced to the story,   so comics, picturebooks, shows and movies don't  need narration - and in fact often suffer from   its inclusion, because audio narration can feel  like a clumsy and abstract way to communicate   plot beats in a visual medium. Without a  narrator, it might seem difficult to have   an unreliable narrator. But even in stories that  don't have third-person narrators, you can have   unreliable narration or an unreliable narrative.  Which is good, because otherwise it'd be really   hard for me to put visual examples in this video! The easiest way to justify having an unreliable   narrator in a non-prose medium is to have  a character in the story tell a story. This   character thus takes on the role of a first-person  narrator for the benefit of the other characters   in the story. This can be a framing sequence  for an entire movie or a full episode of a show,   where the story begins with a character starting  their story and ends shortly after they finish.   Maybe the characters are all sharing ghost  stories around a campfire, or someone is   being interrogated by the authorities about  what they did for a recent heist or crime,   or the characters are reading through another  character's diary, or a movie opens with narration   from an unseen much older version of one of the  characters we're about to meet, or maybe several   characters are all weighing in on their side of  a big story, Rashomon-style. In these scenarios,   the characters can all be expected to be biased by  their personal eccentricities, and that means the   writer can get a lot of mileage out of letting  every character tell the story unreliably in a   way that makes sense for them. We get to see  how the characters flanderize each other in   their own exaggerated ways. If two characters  dislike each other, the way they characterize   each other is liable to be cartoonishly evil or  cruel or otherwise hilariously out of character. A   prideful egotist might tell a story where they're  the heroic champion that singlehandedly saved the   day by fighting off a truly implausible number  of bad guys. A villain might cast themselves as   the victim. A parent might censor the swearwords  and steamy bits of a story they're telling their   kid. A child's version of a story is liable  to be a bit more unfocused than an adult's.   This is a way to reinforce characterization while  also giving the audience a fun mystery to puzzle   through - what parts of the story are true? What  parts are exaggerated? What really happened? In   cases where only one character is telling the  story so we can't cross-reference with anyone   else's version, we're often directed to take  them at their word for the bulk of the runtime,   only for doubt to be cast on them at the very  end for a fun twist reveal. Maybe the happy   ending they described was wishful thinking, or  the story is revealed to be an elaborate lie.  Broadly there are three ways narrators can  be unreliable. The first and most obvious is   if they are lying liars. These narrators will  flat-out deceive characters and audience alike   by either misrepresenting events or leaving out  key information, like for instance "and in those   intervening ten minutes I totally killed  that guy." The second type are ignorant or   inattentive. These narrators aren't intentionally  misrepresenting the story, but they don't really   understand what's going on, so their version  of events might skip past key details they've   missed and be broadly not very useful. This is  most common in story-within-a-story settings,   because it can easily be played for comedy if  the narrator chooses to focus on a seemingly   incidental detail while extremely significant  things are happening in the background. And the   third type is hallucinatory. These narrators  are also not intentionally misrepresenting   the story most of the time, but their actual  perception of reality is unreliable because   the way they perceive their surroundings is not  how they're really happening. This works quite   well in visual mediums, because the camera  can show what the character is perceiving   with subtle hints that it's not exactly  bound by physical laws or shot continuity.

[00:07:53]Unreliable narration can also be used for  dramatic irony, if the character narrates a   situation playing out one way while the audience  sees it play out differently - presumably how it   really happened, giving the audience information  that the characters hearing the story don't   have. A character might describe themselves as  totally in control of their life while the camera   treacherously reveals them to be a hot mess.  In contrast, a character might also describe   themself as lame or uncool or broadly disliked,  only for the objective POV to show them surrounded   by people who like and value them, and only their  insecurities are making them tell the story that   way. This is again used to highlight the character  and their unique traits; the way they tell a story   tells us something useful about this character -  how their own perceptions color reality, and how   honest they're willing to be with themselves or  the people around them. This is very useful stuff.  But the versions we've been discussing  have all been safely compartmentalized   in a story-within-a-story and used for the purpose  of highlighting a character's unique wackiness in   the way they specifically see the world and tell  a story. In these scenarios we typically trust   that the camera outside that story-within-a-story  is still going to be honest with us about what's   really going on. This is not always true, and  that's where things can start getting complicated.  When an unreliable narrator has control over  the entire story, things start getting dicey. In   general, an audience will start a story trusting  that the story is not lying to them. It likely   isn't telling them everything - even an omniscient  perspective typically keeps the camera pointed   away from the really juicy reveals - but the  audience generally trusts that what they're seeing   is what's really happening. They don't go into a  story assuming the story is going to lie to them.  However, there are ways a writer can clue  in the audience that they should start being   suspicious. Maybe characters are acting strangely  or inconsistently, or backgrounds are shifting in   ways that don't make sense, or continuity errors  are cropping up between shots. An audience might   begin to get suspicious, and will generally  have those suspicions confirmed fairly soon   after the evidence piles up. The suspicious scene  might be revealed to be a dream, or a simulation,   or a hallucination, or somebody's mind palace,  or a brief glimpse into an alternate timeline,   or any number of plot-based handwavey  justifications the writers can use to   briefly bait-and-switch the audience with an  impossible or deeply unlikely scenario. This   can be a lot of fun, but it has a detrimental  effect on audience investment in the long term.   If they've already done the nightmare sequence or  the holodeck training program trick enough times,   the audience might start feeling broadly unworried  when a situation looks particularly perilous,   because it's even odds it's not even really  happening. This means when it turns out to   actually be happening, the audience has to play  catch-up on their investment. And unfortunately,   if the audience has picked up on "clever clues"  that the scenario was fake - like the writing   making no sense or continuity errors in the shot  - that goes from being very cool and smart writing   to subtly hint that the events aren't really  happening to the unfortunate realization that it's   actually just bad writing and poor set management.  It's a risky trope to overuse because it hits with   less punch every time, and diminishes the punch  of the actual non-fake plot twists - and it can   train an audience to pick apart your writing  looking for clues, which is of course a good   and fun thing for an audience to do, but it can  be a bit disheartening for the writer regardless.  But of course, one easy way to avoid that issue  is to tell only one self-contained story - like   a standalone book or movie - and have that  turn out to be an unreliable narrative,   because that means there aren't later installments  for the audience to disbelieve. One relatively   common way this plays out is a reveal near  the end of the story that the POV character   has been hallucinating for large portions of it,  and whole scenes didn't happen, or didn't play out   the way they thought they did. Or a character's  memories have specifically been tinkered with,   and any flashbacks and recollections they had  aren't what they seemed to be. These reveals   usually hit late in the game and make a lot more  sense on a second viewing, where the audience   can be on the lookout for particularly  exciting hints at the break in reality.  Of course, once you establish the idea that the  POV character's perceptions can't be trusted - on   a certain level, that's kinda game over. For  the purposes of this story, the audience can no   longer be convinced that anything they're seeing  is real. It has a similarly disengaging effect   to the "all a dream" twist, since it confirms  that the story can and will pull the rug out   from under the audience and it's pretty much  impossible to take anything the story presents   at face value anymore. Most of the time this  is a feature and not a bug, since this reveal   usually hits when the story is already almost  over and it can be very cool for an audience to   go back through a story like that and catch all  the hints and double meanings they missed the   first time through. If the story doesn't have  to convince the audience that they're seeing   something that's really definitely happening, it  isn't a problem that they basically can't convince   them anymore. But this can become an issue in  serialized media or other longform storytelling,   because if the story gets too comfortable playing  with the audience's perceptions of reality,   then as mentioned the audience is going to  disinvest - and if the writing gets too shoddy,   the audience might literally refuse to believe  it's actually happening. The problem is only   worsened if the writer gets cute with it, as they  so often do, like having a character snap out of a   horrifying unreality into a scene where they're  being rescued - which in turn dissolves into a   horrifying unreality until they wake up to being  rescued again but for real this time. It's pretty   understandable that both the characters and the  audience might have a little trouble believing the   story is actually ending the way it says it is. And unfortunately, just like in "all a dream"   scenarios, this is technically an interpretation  that could be applied to literally any story   ever. The idea that a story is being told by  an unreliable narrator lying to the audience   is technically always true because the writer is  lying to the audience by telling them a story in   the first place. This is one of those foundational  elements of storytelling, like the fourth wall,   where if the writer starts poking at it too  hard it can destabilize the storytelling process   itself. It's literally an attack on "suspension  of disbelief." In order to fully enjoy a story,   an audience can't be too hung up on the  fact that the story doesn't necessarily   make 100% sense or isn't 100% convincing in  the effects department. The story could be   critiqued or nitpicked for errors or speculative  fantasy that doesn't necessarily make sense,   but doing so damages the audience's ability to  take the story at face value, so when an audience   suspends their disbelief, they're choosing not to  do that in order to remain immersed in the story.   This is why things can get tricky when the story  tells the audience that they should have been   disbelieving them the whole time because actually  the story was a lie. It can be an exciting,   unusual approach to storytelling because it  directly encourages the audience to critically   examine what they're seeing rather than taking  the characters at face value, but if they take   that far enough - or rather, if the writer  directs them to take that far enough - the   audience's ability to stay immersed can be deeply  compromised. In the end, the audience won't know   when to trust the story, and if the story does  something they dislike, they might choose to   believe that it was a lie too. If the author  makes too much of their story a canonical lie,   they lose the ability to control the  audience's belief in what they're telling them.

[00:13:49]So… yeah!