Avalanche Soldier, by Susan R. Matthews

Second paragraph of third chapter:

Stepping carefully between berry-bush roots and fronds of fern and the long thin branches of ground-level shrubs, Salli smiled to herself. It was Fenka. She liked Fenka. She kept her mouth shut and walked on.

I bought this on seeing a recommendation from someone back in 2011; and I cannot not find whose recommendation it was, or why. I thought at first it might have been Liz Bourke, but her review is from 2012, after I had bought it. Lisa DuMond wrote it up in 2000, the year after it was published, but I don’t think I saw that either.

It’s a book about the intersection of religious doctrine and security politics, in a society where a Reformation is up-ending traditional power structures. Unfortunately I never quite understood what was going on, and was particularly thrown at the very beginning when the central character’s brother, a serving member of the security forces, fears he may be in trouble because he accidentally killed a member of a minority group during a riot. This doesn’t seem very plausible. I note that Bourke and DuMond, who are otherwise boosters of Matthews’ work, are frank about the imperfect execution of this one. I haven’t read anything else by her, and I’m not going to look out for it based on this. But if you want, you can get it here.

This was the unread sf book that had lingered longest on my shelves. Next on that pile is The Nebula Awards Showcase 2011, edited by Kevin J. Anderson.

1 thought on “Avalanche Soldier, by Susan R. Matthews

  1. I think you’re right, population growth is too significant a factor to ignore, even in the context of the hilariously oversimplifying assumptions I was making elsewhere. If I’d extrapolated back further on the same basis, we’d have Mitochondrial Eve in around 250AD. Oops. Also the result of another basic flaw, relating to “robust” lineages with many survivors in a given generation, but I’m not quite so sure how to readily correct for that one.

    If we assume each woman has exactly 2.3 children (yes, I know!) that’s a matrilineage survival chance of more like 80% per generation, and gives about 9m ancestors for 18 generations. However, I don’t think 28 is a realistic generation length as a global estimate for that time interval. I can imagine that there’s the sort of difference in the post-med period by class as there presently is by country, tending to account for what you’re seeing in your sample data. For comparison it’s 5m for 20 generations, 3m for 22.

    A slightly less crude model might approximate family size with a Poisson distro, and keep a count of the representatives for each lineage. That’d need actual code, though, not just futzing around in a calculator… Maybe if I get a bad dose of cabin fever around Boxing Day or so.

Comments are closed.