The Too Unreliable Narrator (was: What really happened on the tower)

[justcarol67]

No: HPFGUIDX 156118


Carol earlier:
> > First, a technicality: a third-person narrator (unlike a first-
person narrator) is not a character, only the voice or persona that
tells the story, in this case usually but not always from Harry's pov. 
> > 
> > Second, and more important, the narrator does not deceive the
hero. The hero has no clue that he's in a story and has no way of
knowing what the narrator says about him. The narrator doesn't impose
limitations upon himself or herself. The limitations are imposed by
the author, who decides whether to use a third-person limited narrator
(limited to a particular character's pov), a third-person dramatic
narrator (who sees the characters from the outside), or what amounts
to a first-person narrator (detailed exposition delivered by one
character to another, notably Crouch!Moody in GoF or Dumbledore's
expository near-monologues in most of the other books).
> 
> Neri responded:
> I acknowledge the distinction between the author and the narrator. I 
> was aware of it all the time but didn't want to complicate this 
> discussion even further.
> 
Carol:
Good. I was chiefly concerned with the idea that Harry could be aware
of the narrator. He's not aware that he's a fictional creation,
either. And I was attempting to show, perhaps not very clearly, that
JKR varies her narrative technique depending on the needs of the plot
(more on that below) and that a third-person limited narrator has the
same limitations as a first-person narrator--he can't know anything
outside the pov character's frame of reference and, unlike the
dramatic narrator of some scenes, he can't be objective because he
reports the scene as the character sees it, including the character's
feelings and misconceptions.
> 
> > Carol earlier: 
> > <snip>
> > In a few instances, we have the narrator rather sneakily hiding
from the reader things that Harry knows perfectly well, for example
that Harry isn't giving Ron Felix Felicis.
> 
> Neri:
> No, we don't have *few* instances. That was my whole point in this 
> thread. We have exactly *one* known case throughout the whole series 
> where the author fools the reader by hiding from him things that 
> Harry knows perfectly well (BTW, weren't you mixing author and 
> narrator yourself here?). We also have exactly *one* case throughout 
> the whole series in which it is Harry himself who fools the reader. 
> Very interestingly these cases turn out to be the same – the Felix 
> Felicis incident. I don't think this is any coincidence.  

Carol:
You could be right here, but that doesn't undo the basic argument that
JKR uses the narrator to manipulate the reader in a variety of ways,
this being only one means of doing so. (But, yes, it was JKR who was
being sneaky by having the narrator conceal rather than report
information.) However, I think if we looked, we could find other
instances of concealed information, the caster of the Petrificus
Totalus spell possibly being one of them. Also, just being inside
Harry's head keeps us from knowing that Tonks is moping about Lupin,
to take one example. An omniscient narrator could get inside Tonks's
head and tell us what was wrong with her. Of course, he could get
inside Snape's head, too, and that would spoil half our fun. ;-)
> 
> 
> > Carol:
> > But in *no instance* does the narrator withhold information from
> > *Harry.*  <snip>
> > The narrator sometimes knows things that Harry doesn't know
> 
> Neri:
> I'm afraid that after the above two statements I'm more mixed than 
> ever <g>.
> 
Carol responds:
Sorry about that. I agree that my explanation wasn't quite clear. Let
me repeat that Harry has no idea of the existence of the narrator and
so the narrator can't withhold information from Harry, only from the
readers. Harry isn't reading the books. He's *in* them.

And the narrator *usually* knows only what Harry knows because he's
*usually* inside Harry's head. But JKR's narrative strategy isn't
perfectly consistent because she sometimes needs to reveal information
that Harry isn't--and can't be--aware of. The only way to reveal this
information *to the reader* but not to Harry is to get into someone
else's head or choose a different type of narrator altogether.
(Occasionally, another character, say Crouch!Moody or Dumbledore, can
serve as a first-person narrator within the third-person narrative to
provide exposition disguised as dialogue to Harry and the reader at
the same time, but that's not what I'm talking about here.)

Most of the time, as I said before, JKR uses a third-person limited
omniscient narrator who sees from Harry's pov. At times, however, this
narrative strategy won't work. In the first chapter of CoS, for
example, Harry is fifteen months old, so his pov wouldn't work even if
he were present for the events she wants to depict (Mr. Dursley's day
at the office, McGonagall waiting for Dumbledore, etc.). Her
third-person narrator first uses Vernon Dursley's (highly limited and
prejudiced) pov, then becomes a third-person dramatic narrator seeing
events and characters (McGonagall, Dumbledore, Hagrid) from the
outside. In the next chapter, we see ten-year-old Harry asleep (and
the house as it appears while he's sleeping--note that this strategy
is repeated in HBP, chapter 3) with a shift to Harry's pov as he wakes
up. With a few exceptions, most notably Hermione setting fire to
Snape's robes, the pov in SS/PS stays with Harry, and the third-person
narrator's "omniscience" is limited to what Harry knows or thinks he
knows (though, of course, the narrator doesn't report everything that
Harry says, does, thinks, sees, hears, or feels, or the books would be
impossible to read--or write). The narrator must be *selective*--or
rather, JKR must select what the narrator reports--and, of course,
control the *way* he reports it, including the completeness or
incompleteness of the information and the degree of objectivity and
accuracy with which it's reported. 

I can't remember any significant deviations from the usual
third-person limited Harrycentric pov in Cos or PoA, but in the first
chapter of GoF we get Frank Bryce's pov followed by a third-person
dramatic (objective) flashback of the events surrounding the Riddle
murders. The dreams in Go f and OoP provide a narrative device that
allows us to see from LV's pov (or Nagini's) with an overlay of
Harry's. HBP starts out with the third-person limited narrator in the
Muggle Prime Minister's mind, then a third-person dramatic narrator in
"Spinner's End" (and the beginning of "Will and Won't," while Harry is
asleep), and then a return to the usual Harrycentric pov, compounded
by an increasing hatred of Snape.

In the (very few) chapters where we see the characters from the
outside, we have nothing but our own preconceptions to color our view
of the characters' words and actions (but we can still sense that
information is being concealed from us because the characters conceal
it from each other). In the third-person limited chapters, we also
have the pov character's preconceptions and limitations to influence
our interpretation. When the narrator is inside Harry's mind, he (the
narrator) can't know anything that Harry doesn't know (the events in
"Spinner's End," for example, even though those events were reported
earlier in the book) because Harry doesn't. (If that's confusing, you
can think of those chapters as being told by a different narrator than
the one who sees through Harry's eyes, one who can see what's
happening to Snape and Narcissa and Bellatrix, but can't get into
their minds. Neither, of course, can we.) The reporting of events in
"Spinner's End" is straightforward, though of course we're still
confused by the withheld information and don't know how much of what
snape says is true or what his motives are; the reporting of many
other events and conversations includes paraphrases of Harry's
thoughts and his emotional reactions. His fear becomes our fear, his
mistrust our mistrust--the ubiquitous "Harry filter."

But JKR doesn't *just* control the pov (which may or may not be
reliable, depending on what's being reported or described). She also
makes sure that conversations are interrupted at key points, that
certain events occur almost simultaneously so that one seems to cause
the other (Harry's scar hurting "because" Snape looks at him) or one
is mistaken for the other (Madam Pince appears from behind the
shelves, disguising the fact that Draco is hiding there and
eavesdropping). The argument in the forest is reported by Hagrid, not
even overheard directly by Harry, distancing the reader still farther
from the action and increasing the likelihood of unreliable reporting.
And, of course, the characters themselves frequently provide seemingly
reasonable explanations for events and other characters' actions that
turn out to be incorrect (case in point, Hermione's idea that Tonks is
suffering from survivor's guilt). Sometimes they're right; sometimes
they're wrong. (If they were always wrong, there would be no point in
listening to anything they say. But even Hagrid can be right; even
Hermione can be wrong.)

JKR is manipulating us in various ways, just as she is in her
interviews. She doesn't want us to figure out exactly where the story
is going, or exactly where Snape's loyalties lie. Just as some clues
are real clues and some are red herrings, the narrator (and the
characters' version of events) is sometimes reliable and sometimes
not. We just have to know what to look for and interpret it for
ourselves--at least until she gives us the answers in Book 7. And even
then, we (the group members, not just you and I) still won't agree on
Snape or the Weasley Twins or Dumbledore or Merope.

Neri:
> OK, regardless of any technical issues, I think you and Pippin still 
> manage to avoid the main point of my thread. The Author obviously 
> can't allow the narrator to be *too* unreliable, because in that
case  the reader simply would never believe a word the narrator is
saying. 

Carol:
True. Rather like Snape in "Spinner's End," the misdirection has to
appear plausible and must be mixed with truth. The question is which
is which.

Neri:
> My question is: What is JKR's personal limit in using the 
> unreliability of her narrator to fool the readers? 

Carol:
I'm not sure that even JKR herself can answer that. But I'm inclined
to agree with those posters who think that she puts the needs of the
plot above consistency in other matters, whether it's narrative
technique or the number of years it's been since Gryffindor won the
Quidditch Cup.

Neri:
I'm interested in 
> definitions of this limit backed up by examples in the book
(examples in which we already know whether we were fooled or not, not 
> hypotheticals like the PT case).

Carol:
I've given numerous examples. Please see upthread. Look at the whole
presentation of Tonks in HBP if you want examples of misdirecting the
reader regarding a character's motives or feelings.

Neri:
 Pippin came up with a list of rules, 
> but I think that they basically boil down to "wherever the author 
> raises some doubt – anything goes", and since JKR almost always 
> raises *some* doubt, in practice it's "anything goes". So where is 
> the limit between tricking and conning? What *wouldn't* JKR do to 
> fool the reader?

Carol:
I'm not about to give a list of rules because you'd immediately find
an exception--and so would I. I think, however, that JKR wants to keep
us in doubt on certain key points, chiefly (IMO) whether Dumbledore
was right about Snape and where Snape's loyalties lie (possibly also
whether DD is really dead, but I'm not convinced about that one, or
about ESE!Lupin).

Carol, who can't believe she's still writing on this topic and hopes
she isn't boring anyone