I’m about to spend the next nine months or so studying for my specialist exam, an oral exam conducted by a committee of three of my professors. I have to read around 30 articles or chapters, and be able to answer detailed questions about them all. But first, I have to figure out what I should be reading. This is tricky.
There are easily thousands of articles I should read. Way, way more than I’ll ever be able to actually get to. I can read a lot of abstracts, but to really digest an interesting reading takes not only time, but mental space. In order to deeply process a new idea, I have to also not read a lot of other things. So there is a hard upper-limit to the amount of high quality work I can do between now and the spring. A lot of naps will be involved.
I think of it of a particularly large and vexing instance of the frame problem. To quote the article I just linked, the most succinct formulation of the problem I know is as such:
How could an inference process tractably be confined to just what is relevant, given that relevance is holistic, open-ended, and context-sensitive?
Another way to put this problem which comes from Vervaeke (here) is to ask how it is we can intelligently ignore almost all the information available to us. And you can’t ignore it by checking to make sure that you don’t need to ignore it – because then you haven’t ignored it. Vervaeke et al. think we solve it by being self-organized, by being an economy of cognitive strategies that compete. That sounds plausible to me, but it doesn’t provide much guidance as to how I should determine my reading list.
Intersubjectivity will help, of course. I’ll produce a list which is too long, and people who know more than me (my committee) will help me pare it down, and add things I missed. But if I’m really making something novel, that will only take me so far.
I know, I know… I’m wallowing in the difficulty. People solve this kind of problem all day long without even noticing, despite its apparent impossibility. We really are remarkable animals.