Another answer: in my teens I read Randall Jarrell and William Empson and Hugh Kenner and (by that time I was enrolled in her courses) Helen Vendler. Even before my teens, if I remember rightly, I was reading popular science explainers and language mavens and other explainers of complicated things in clarified, non-esoteric language. I was storing up models, without knowing why.
I have been fortunate enough to be taken up by congenial editors early. Not all critics, not all reviewers, get that.
I read a wonderful interview you did where you said the following: “Reviewing, like all other literary criticism, like the making of chairs, like the making of film scores, is an applied art: it’s heteronomous, serving ends outside itself, and should not let its own artfulness detract from those functions.” What function do your reviews serve? And how do you know–can you know?–if you’ve succeeded?
Time to quote Auden! “What is the function of a critic? So far as I am concerned he can do me one or more of the following services: 1. Introduce me to authors or works. 2. Convince me that I had undervalued an author or a work because I had not read them carefully enough. 3. Show me relations between works of different ages and cultures. 4. Give a ‘reading’ of a work which increases my understanding of it. 5. Throw light upon the process of artistic ‘Making.’ 6. Throw light upon the relation of art to life, to science, economics, ethics, religion, etc.” (This and much else from the essay “Reading,” at the front of Auden’s The Dyer’s Hand.)
Reviewers can do all those things. I hope that I have done them, now and again.
Reviews can also clear space for the appreciation of neglected, undervalued or misunderstood art by dispelling bad arguments about art, or by trying to clear worse art out of the way when it seems to be obstructing the view of better art. (Sometimes the better art and the worse art are by the same artist.)
As for how you know when you have succeeded, W. S. Merwin recalls in a poem that he once asked John Berryman whether and how he knew his poems were good, and Berryman replied “You can’t you can never be sure/ you die without knowing.” You can’t quite know what you have done inside a reader’s mind.
On the other hand, reviews can get books more attention—and that’s something you can know. If I review Jane Doe’s second book, and then her third book gets more attention than her second (more reviews, more people reading it, even more copies sold), perhaps at the margin I had something to do with that. Reviewing is like writing poetry in some ways (it’s an art) but it’s also like making chairs (see above) and it is in a third set of ways like voting: an individual contribution to a necessarily collective effort.
How do you sidestep the peril that is reviewspeak (oh, all those “luminous,” “lyrical” “tours de force!”)–and formulaic review structures?