[Aiwendil]

Originally Posted by Bethberry

I hope this makes sense!

It certainly does, and I fundamentally agree with you. Indeed, I think the gist of this thread is everyone vehemently agreeing that Tolkien was not a postmodernist. But let me just quibble and clarify what I was saying slightly.

The revision or editing of the "Riddles in the Dark" does not represent unreliable narrative. It more closely resembles the kind of niggling and endless changes that Tolkien submitted his Legendarium to, in his attempts to create a reliably consistent narrative. The fact that the chapter was changed after initial publication does not demonstrate unreliability but in fact a determined effort to provide consistency. Even more significantly, there is nothing in the revised chapter which points to any changes, which identifies any unreliablity. Contemporary readers only know about the change by reading material extraneous to the actual story of TH, epitextual devices that can mediate for the reader, to use Gerard Genette's term.

To be clear, I'm not saying that revising the chapter has anything to do with an unreliable narrator. What I'm saying is that Tolkien's intention at one time was to let the first edition of TH stand, and to let LotR reveal that TH has an unreliable narrator (by having Gandalf offer a correction to Bilbo's account). He came up with this solution because he did not think the publisher would allow him to substitute a whole new chapter in a revised edition of TH, but as it turned out, the publisher did allow the substitution. So, yes, it's an unreliable narrator that he only came up with out of (imagined) practical necessity, it's one that is not shown to be unreliable within its own work, but only by a sequel, and it's one that only exists in the first edition of the book and thus doesn't exist at all for today's readers. But I nonetheless think that it does technically fit within the term "unreliable narrator".

Nor does the suggestion that the

really fit postmodern ideas about the inability of language at all to reflect reality or truth. Differences of interpretation have long been part of literary texts and characters galore are known for their lies, their loss of memory, their misrepresentation and fairy otherworlds are part of European literature for centuries.

Yes, but the idea Tolkien was toying with here was quite different from a character lying or misrepresenting something. It's most acutely seen with "The Drowning of Anadune". If, as Tolkien apparently contemplated at one time, he had published some form of the Silmarillion in which "The Drowning of Anadune" was included, then we would have had a pretty clear case of unreliable narration, since one section of the work would have given us one account of the creation and history of the world and the nature of the Valar and the Elves, and another section would have given us quite a contradictory account - with neither of these accounts being framed as an interpretation or lie or misunderstanding on the part of any character within the work. So again, even though the motivations here are completely different from those that lead to interest in unreliable narrators by postmodernists, I'd still argue that Tolkien at least contemplated using a device that technically fits under the definition of unreliable narration.

Again, though, this is quibbling stuff about definitions, and I mainly just wanted to clarify that no, I'm not claiming that revising one's book after it's been published, or writing about fairy otherworlds, somehow constitutes unreliable narration!

Incidentally, and probably somewhat beside the point, I'm not sure I agree that postmodernism necessarily has as a central tenet that

inability of language at all to reflect reality or truth.

Certainly, this is an idea underlying some works of postmodernism, but I'd argue that more generally, postmodernism is interested in questioning whether, and how, language is able to reflect reality or truth, without necessarily positing a particularly firm answer to that question.