Today’s starting point? “Trust Me: Three Books With (In)credible Narrators”
Someone — meaning most of the commenters — does not understand the concept of ‘unreliable narrator.’
First of all the main post itself is at least mildly interesting, in particular insofar as it provides a concrete reference as to the popularization of the term (given here as [a] the 1960s by [b] Wayne C. Booth). Furthermore the definition given is both concise and relatively flexible:
An unreliable narrator is one who tells a tale with compromised credibility, whether the narrator herself understands that or not. The reader usually finds this out only slowly, as cracks in the narrator’s version of events begin to appear.
This establishes a three-part relationship between reader, narrator, and the story. We might say narratee, narrator, narrative/narrated. We’re not, at this point, contemplating the relationship between the historical, ‘real,’ author and the narrator. And more broadly speaking narrator reliability is a matter of how much [a] the reader can ‘trust’ the narrator and [b] how well what the narrator narrates corresponds to the underlying events of the narrative.
It’s not just that this takes place in fiction; it is a fiction. The reader cannot trust the narrator for the simple reason that the narrator is not a person. I likewise do not trust a storm or a stone, let alone a storyteller. Secondly we cannot assume anything beyond the surface of the narrative; the idea of there being a “what really happened” is a fiction, and all we have is the text.
This, too, is a fiction. Or at least a convenience and contrivance. We always approach tales with baggage. What we’ve read from or heard about the author, what we expect from the genre, that we consider the story a typical work of the author or part of a genre, and so on. We take certain narrative conventions — narrative voice, time and place, some level of consistency if not exactly sense-making — for granted.
And so to be accurate or at least as accurate as possible, all narrators are unreliable. At best they are untrustworthy to the extent that we take them as ‘people’ or at least ‘personalities,’ and everyone is capable of deceiving, intentionally or not; but more properly they are all untrustworthy by not belonging to a category that can be trusted. As there is within the text no literal, only figurative ‘within,’ no ‘depth,’ no ‘layers,’ only the interactions of words; there cannot be surface and depth, style and content, what we are told and what is real. There are no unreliable narrators; only unreliable readers. We trust the untrustworthy, we see depth where there is only surface, we treat context as text.
But by this standard it’s not even reasonable to talk about reliable or unreliable narrators. We’ve been perhaps a bit too strict in our reasoning, or at least a bit too heavy-handed and direct. It did at least reveal something, that we have to remember that all narratives are constructs, fictions not only at the level of story, but at all levels of construction: author, narrator, narrative … reader. Some are more fixed, some more fluid than others. In that sense all fiction is postmodern and all fiction is metafiction. But this does not address degrees; readers can — and ought to! — feel more certain or settled with some narrators and narratives than others. History and experience are reliable up to a point. Returning then to Wayne C. Booth and Amy Wilson’s brief post, the salient point will be the dynamic of “compromised credibility” as credibility compromised during our reading of the narrative progresses. The narrator we know not to trust from the beginning is trivially unreliable; the narrator whose psyche and narrative that show no cracks over the course of the story engenders no distrust, technical unreliability aside. In film we might take both Verbal Kint ‘(The Usual Suspects’) and the unnamed narrator of ‘Fight Club’ as illustration. Amy Wilson provides three useful examples: Pale Fire (Nabokov), We Need to Talk About Kevin (Shriver), and London Fields (Amis).
In my literature in translation course I’ve provided students with a couple narrators who are subtly unreliable for slightly different reasons. The first person voice in Schnitzler’s Fräulein Else (1926) is vaguely ‘hysterical,’ although the term is problematic. She has a wild imagination, but she is also very intelligent. Her observations seem sound but perhaps hyper-reasoned; her narrative voice also transitions between past, present, and a subjunctive. The reader who expects certainty will be let down. By virtue of being an early stream-of-consciousness voice Else is obviously unreliable whenever she speaks about that beyond her experience. We also dealt with Gustav von Aschenbach in Thomas Mann’s Death in Venice (1912). The narrative thread of the novella is tied to Aschenbach, but there are two voices, one of which is mainly the protagonists, the other which is slightly more ironic and distanced, one that can see the gap between Aschenbach’s decisions and ‘reality.’ The reader is not mislead (as long as we read closely), but Aschenbach misleads himself. And while not the only other such voice, the last we usually deal with is that of Walter Faber in Max Frisch’s Homo Faber (1959). The reader comes to realize that the entirety of what we have read is an after-the-fact text composed by Faber. He has presented it as an objective, factual report, but he is intimately and subjectively involved. What the reader experiences as tension and revelation — and what is presented as novel for the narrator — is after-the-fact for Faber. Part rationalization, part justification, perhaps.
I would add these three to the trio provided by Wilson.
When I read the NPR comments I am, of course, dismayed, but, as the response must be, “Oh no! Someone is wrong on the internet!” Oscar Obel’s ill-supported attempt at humor in suggesting the Bible and other religious texts conflates and flattens the issue, and enlightens not one bit. The suggestion of Lolita is off and suggests that any narrative voice that is not ‘normal’ (disturbed, difficult with which to empathize, etc.) is worthy of the title unreliable, but ignores the definition provided by Wilson and Booth. Others, such as Leah Porter, Thomas Hilton, and Deborah Thompson at least provide more fascinating reading material: James Dickey’s To the White Sea, Gene Wolfe’s The Book of the New Sun (which I’ve been meaning to read), and Iain Pears’ An Instance of the Fingerpost respectively.
