Let’s go back to that first example: reading a letter. I care what the person wanted to say, so literary interpretation isn’t going to cut it. I could do that work, of course, but it isn’t going to get me the result that I want. Trying to discern authorial intent is a somewhat harder task: instead of working out the meaning of the text in itself, I try to anticipate what meaning a person would want to impart when they chose those words. It is a much more speculative activity than literary interpretation. The result is far less certain when trying to discover intent than when trying to discover meaning; ambiguities must be resolved, not acknowledged and incorporated into the reading. Prior knowledge of the person, however, counts as evidence here, which means that you do have more data to work with—unless you don’t know the person very well, in which case reliance on the person’s personality becomes a liability.
And, I think, this goes back to the questions in the second half of my post on John Green, Twilight, and Paper Towns. If we’re holding people accountable for what they wrote, we luckily have all of the evidence we need in the text itself. If we’re holding people accountable for what they intended to write, our project is in trouble from the outset. If we’re holding people accountable for which misinterpretations they could anticipate…that seems difficult, indeed. But, whatever we do, our understanding of the text must be an understanding of the text, and not anything else. That’s why I’m making these distinctions.
(For more on literary theory, see this index.)