One of the most common critiques I hear for short story collections is that they're "uneven." I don't hear it very often for novels, and only occasionally as a critique of an author's oeuvre.
A few brief samples:
- Publisher's Weekly called David Foster Wallace's "Brief Interviews with Hideous Men" uneven.
- Seattle Times called Evan S. Connell's "Lost In Uttar Pradesh" uneven.
- Martin Amis's "Heavy Water and Other Stories" called uneven.
- In the LA Times, John Freeman called John Updike's "My Father's Tears" an uneven collection.
It's not exaggerating to call it the most frequent criticism leveled at short story collections. But I wonder whether this critique is effective or informative.
I think the "uneven" critique is particularly prejudiced against short story collections that embrace a wide variety of forms, such as pairing postmodern meta-fiction along with Carver-type realism and throwing in some genre-inspired work.
Almost inevitably, the reviewers tastes will lean towards one style or another, and they'll laud half the collection and slam the other half. So the critical matrix of short story reviewers (where "uneven" or "even" is used to judge collections) encourages a form-based, limited type of "unity" to collections, and discourage a thematic or innovative type of unity.
"Uneven" also says more about a reviewer's taste than about the content itself. I know this is tricky waters — how would one separate the reviewer's taste from their evaluation of content? — but I feel as if the word uneven is shorthand for "I liked some stories and I didn't like others," which doesn't tell me much about whether I would like the same stories or dislike the ones disliked.
Also, saying that some stories are liked and others disliked is a bit of a cop-out, review-wise. You could direct this criticism at virtually all collections — aren't collections, by their nature, created so some stories stand out of the pack, and seem better than others?
Uneven also means the reviewers are judging the collection as a mosaic of pieces, rather than as a unified whole. For some collections, this is the appropriate approach, but for others, it might be better to judge it as a cohesive beast, the same way one might read and review a novel. A good collection accomplishes a certain goal, and the reviewer should pay attention to the degree to which that goal is attained, talking about the collection as a single entity.