overripe fruit

[psocoptera]

So apparently Gaiman is writing a new Sandman story. Is this actually good news? I feel like creators going back to their most well-known early universes rarely produces anything up to the quality of their early works - in fact, I can't offhand think of an exception to this. Anyone?

Here are some of the cases I've already thought about:
Orson Scott Card's "Shadow" books revisiting the Ender universe
Asimov's later Foundation books vs the original trilogy
Vinge's recent Tines book
David Brin's later Uplift trilogy vs Startide Rising and Uplift War
George Lucas's Star Wars prequels
Indiana Jones and the Crystal Skull

The only counterexample I've come up with so far, after more thinking, is Jo's Boys, which was published fifteen years after Little Men. And I guess one could count the Lord of the Rings vs the Hobbit, although honestly, though it borders on blasphemy, I think a certain amount of Tolkien's later History of Middle-Earth work fits the pattern of "bloated and unnecessary".

Comments

[irilyth]

https://en.wikipedia.org/wiki/Robot_series_%28Asimov%29#Robot_novels says "The Caves of Steel and The Naked Sun are both considered classics of the genre, but the later novels were also well received, with The Robots of Dawn nominated for both the Hugo and Locus Awards in 1984, and Robots and Empire shortlisted for the Locus Award for Best Science Fiction Novel in 1986."

(I also haven't read them in a while; I don't recall being aware at the time that they weren't written in more or less the same time period.)

[q10.livejournal.com]

the tip-offs are that Robots of Dawn is significantly longer than the first two and is more explicit about sex.

also Robots of Dawn contains a reference to Bicentennial Man, and more generally does a lot of work to retroactively situate the Robot novels in some kind of continuity with the other Robot work, all of which is more of a late-Asimov thing to do (cf. the various not-so-great revisitations of the Foundation series).

Robots and Empire of course devotes a lot more time to continuity-merging and associated retconning. (and, i would be inclined to argue, is a worse book for it, although again my memory is rusty.)