I had a vaguely insightful thought about why I tend to like books in a series less in the terminal volume than any of the other component parts.
I think it’s because they’re basically splitting the difference between series and book, as I read them. I’d need to look more into this to confirm, but I tend to have this problem when you have two (or more) basically independent episodic books, and then a third book that focuses on a boss fight and resolving the individual storylines. Books that tell one end-directed adventure that happens to be divided into thirds don’t bother me. Books that are strictly episodic, like most mystery novels, don’t bother me. It’s when you have one strategy and switch to the other that the readers’ expectations are dashed.1
When one story is broken up into parts, all the disappointment of the ending is pushed forward into one chunk that we identify as discreet, and that’s why we think of the series as “sloping downward.” If it was all one book, we would remember it as awesome. And I think that’s why you can wholeheartedly recommend a series even if you think each individual book was kind of just ok. I talked about this a little in terms of Conjuring of Light in my training post earlier this week, but the more I think about it the more it does seem like a general rule.
The series I find most successful are actually the more formulaic ones, so that each unit in the series necessarily has all the elements I liked about previous books. I’ll never get tired of reading Montalbano books, for example, because they’re basically television show made of novels. I think there’s a balance that you can achieve with this, though, that veers slightly more towards the continuing storyline than Montalbano. Camilleri’s use of throw away women really pisses me off.
I’d like to write a series someday but analyzing other people’s work always psychs me out, so we’ll see if I just analyzed myself out of doing it.
- Not really related, but Mieville actually had a really interesting point about endings in mysteries that I think applies here, in a roundabout way. I saw it originally on Whatever some time ago, and I always think of it when I’m doing any kind of long-form plotting.