A reader asks:
"Do you have any practical advice for how to develop into a more sophisticated reader? You mention practice and I was wondering if you meant a high volume of reading or something more?"
The end idea of reading is to notice things systematically as part of an organic and comprehensive experience of the act of reading. In my old days of teaching I would suggest to my students that liking a piece of writing or not liking it--or agreeing with something or disagreeing with it--is not much of a reaction; a good opinion needs to be concrete and specific, or at least be capable of becoming concrete or specific.
How does one reach this point? It requires both a plan and at least one guide. The nature of the plan and the identity of the guide hardly matter, because good readers revise their plans and end up finding multiple, partial guides. My first plan was a list of 'classics' prepared by my local library, and my first guides were the large volumes of literary criticism that were, for inexplicable reasons, in my high school library. I poked around the list to find the books that seemed most initially interesting to me, read them, read some reactions to them, and picked subsequent books off the list that seemed to match what I liked in the first ones. I was fortunate in that the list and the books of criticism were old: there's nothing that matches the old view of Great Books whose value can be known and measured. Not all of them are great, not all of them are still great, but it gives a useful outside measure for one's own reactions: "this book has been known and appreciated for 300 years, but I seem to think it's boring. Perhaps I'm missing something." And from these readings one goes looking for influences and disciples. It helps if the guides one chooses are willing to name-drop those authors they find especially noteworthy. Fortunately, almost all of them do. This should provide a framework in which individual acts of reading happen according to mood, interest, etc.
Once I've finished a book, I have a set of tentative reactions to it. At this point, I find it helpful to go and read criticism of the book, both positive and negative. Good criticism will give you a sense of what is important to notice and how the book works as a unified whole. It can cause you to correct your opinions by pointing out things you did not notice on your initial reading, and can confirm when your instincts are pointing in the right direction. I read a large number of Graham Greene novels, and enjoyed them, until I read a Christopher Hitchens essay that pointed out Greene's serious novels tended to repeat the same plot elements, themes, and structure; what I was responding to was less the quality of his writing than the fact he was producing certain outcomes and conclusions I found agreeable. I still think of Greene as a good writer, but only a good one, and one who writes novels that beg the question on his own most cherished beliefs. Alternatively, the criticism of Roberto Bolaño's 2666 confirmed for me that a number of its technical and structural features, which I had found essential to appreciating the book as a whole, were indeed functioning as I had suspected. Given enough time, this will trickle down into opinions about various people who write in a critical capacity: Zadie Smith, for example, is a very wise writer talking about the technical act of writing, but her literary recommendations are strongly colored by what she finds useful as a writer, and are not generally what I like to read.
The trick in working through a book is to think of writing as those critics do, as something other than the neutral content through which a story is transmitted. Structure matters, style matters; any act of writing is a collection of individual decisions about what to do--it is almost always phrased this way instead of that way for a reason. Write down individual passages you like, and figure out what you like about them, paying attention to the really boring stuff you hopefully learned in Freshman Comp in college: how is the paragraph structured? Are the individual sentences the same, or do they vary, and how do they vary? What's the cadence of each line? Where do the important words in the sentence end up? What unusual things does the writer do? It's often simply a matter of noticing what works and what doesn't: everything in Zola's Therese Raquin is good except for the passages in which he allows himself to present his psychological theories, which are dull, not insightful, and make the story lag; this plays out in the rest of his books: it's hard to think of a more insightful cataloguer of the range of human emotion who so poorly understands the nature of his own gift.
When you find yourself reacting strongly in a positive or negative manner, the important thing to sort out is why. Reading, reading criticism, and a reasonable attentiveness to your own reactions go quite far. I sometimes make the cooking analogy: if you never cook, cooking will be quite difficult; if you cook a bunch of disconnected things, you'll have individual spots of knowledge; if you supplement by reading the non-recipe portions of cookbooks, and working through a bunch of different options within one cookbook, you'll see the way the ideas and techniques repeat themselves.
It's also important to read mediocre books with some regularity. Occasionally finding novels one dislikes is a sign of being adventurous enough, either guessing wrong or finding an unanticipated weak spot in an author's oeuvre. Books less than classics also make the method of their composition much plainer; the various imperfections are easier to see, and this is one easy way to recognize what constitutes excellence, and how vertiginous the gap is between those things that are written at all (hard enough to do), those that are merely good, and those that are classics. That's the real value of grad school: one reads so much that is written for a wide variety of reasons, and in the social science, with so little emphasis placed on the elegant expression of ideas, that one gains many, many examples of how one can go off-model.
On plans: reading over a list of classics with a long historical scope makes for a good starting point. Doing so allowed me to make a few first tentative cuts into literature: Greek drama, medieval and renaissance literature, poetry, French and Russian 19th century novels. All but the last two eventually faded away, and 20th century literature came on quite strongly, especially British novelists of the second rank and, eventually, Spanish and Latin American literature. But the general idea is to read enough in any niche one favors to be able to offer some elementary comparisons amongst its figures, and some comparison between niches (T.S. Eliot talks about this quite helpfully in one of the first lectures in The Use of Poetry and the Use of Criticism).