Reading, reviewing, analysis, theory

That title's intended to deter. This post is hardcore literary thinking.

It's clear that there are at least three different approaches to reading a book, specifically a work of fiction.

Reading for pleasure is what most of us do most of the time. Reading to review is what reviewers, including bloggers, do. Reading to theorise is the most specialised form of all. Why do these three kinds of reading seem to use such different language?

I've just read The Confession of Katherine Howard by Suzannah Dunn. I made some notes as I went through it, because it struck me really early that there was an obvious mechanism in use, by which the narrator hinted at what was to come later, and, in effect promised to tell us more later. I called these "portents". At the same time, there was another mechanism by which the narrator referred to past events in a knowing way, aware that we would share her sense of the dramatic irony. I called these "omens". Now, I'm not sure if that's a model that will stand upin the long run: can we characterise a novel's control of information that way? I've also little doubt that the mechanism's been discussed before, though maybe not using those terms. But I was reading the book as a theorist.

As I went on, these devices seemed less prominent, or less obvious. What happens is that early promises (portents) are delivered, and the need for mystery becomes less. The notion of portents and their fulfillment is close to Barthes's S/Z, but I think he sees them as structural, rather that instrumental. This may be because he under-examines the process of reading for pleasure, so I'll move to that.

In the early stages of a novel, you need some incentive to stick around. You haven't yet formed any attachment to the characters. You need the promise that something interesting is going to happen. Portents, then, perform this function, just as the questions Barthes identifies ("Who is Sarrasine?") do. He talks a little dismissively about the possibility of a naive reading, and in the case of Howard that's a valid point. Regular readers of historical fiction will have a pretty good idea how it's going to turn out for her, and even I was able to deduce - applying the "divorced beheaded died divorced beheaded survived" mnemonic - the ending.

But that makes it harder, and the next type of reading - for review - would have to concern itself with how well the book engages the interest. So a reviewer might say things like "Dunn retells a well-known story with a fresh look", or "Although the outcome is never in doubt, Dunn keeps us guessing as to exactly how it will come about".

Reviewers also, inevitably, have to talk about character - a bit of a dubious area for theory. Characters have to be assessed for their believability, and maybe less so for their likeability. (It's usually important to be able to like the narrator.) Long ago, on this blog, I wrote about The Last Man. One of my big objections was that the characters' personalities were implausibly inconsistent. Theory would have difficulty in explaining why that matters. Again, it may be because the process of reading for pleasure isn't adequately considered.

A theorist would need to consider what it means for a character to be consistent; a reviewer would need to point out instances where there had been unreasonable inconsistency, and a reader would perhaps just feel that the novel isn't convincing: "I don't believe she'd do that".

I'll leave this thinking aloud there for now, but I will come back to it. One of the questions I want to look at is the way a poorly written book can be enjoyable when read for theory or, come to that, for criticism.