| SF Reviews by Aaron M. Renn | By Author - By Title - By Date Reviewed |
Conclusion: Incomprehensible but Worth Reading
This is a science fiction novel that combines a number of elements that should satisfy many an SF fan: advanced AI, galactic exploration, a vast sweep of time and space, etc. Indeed, there was a lot that I liked about the novel. However, it suffers from way too much pseudo-scientific gobbledygook. There's a quote at the front from the San Francisco Examiner that says "One gets the feeling at times not of reading a novel, but of witnessing an extended conversation Egan is having with himself on topics ranging from biotechnology to particle physics to social theory." Somehow I don't think that was meant to be a compliment.
The novel opens sometime near the end of the 21st century. Humanity has splintered into disembodied AI's living in cyberspace polises (some of them actual uploaded humans), people who live in robot bodies, and those who remain relatively human, termed fleshers. I say relatively because most of these humans have radically altered their genetic structure in some way. The theoretically impossible collapse of a nearby binary neutron star system sends massive doses of radiation to earth, severely harming the biosphere. This brings home to the non-fleshers that they too might prove mortal. So they decide to try spreading across the galaxy in an attempt to improve survival chances, hence the title.
Right off the bat we're treated to an example of what is wrong with this book. This is in the form on an extended description of how one of the polises goes about constructing a new AI citizen. This could be considered an interesting bit of speculation to some fans, but to me it was just superfluous. This fascination with scientific speculation, as opposed to the normal plot, character, setting etc of the mainstream novel, is one of the biggest weaknesses of the genre, IMO, both from a fan and writer perspective. Every SF novel needs what I call its "SF premise", some bit of speculation that informs the story or setting such that the book is SF instead of a vanilla novel. But taking this too far is, again IMO, a big mistake. Egan crosses this line into scientific speculation for its own sake.
But the description of AI replication is only the beginning. I'd estimate about 30% of the book is devoted to extremely dense, virtually incomprehensible cosmological theories. While I guess these are necessary to advance the plot as such - much of which revolves around trying to explain things such as the neutron star system collapse that were theoretically impossible under existing theories of the universe - it really makes the book difficult to follow. I'll admit, I took to skimming the pages when Egan went too deep into this stuff. A) It isn't real, so it's not like I'd really learn much B) the details aren't relevant to the story, and C) you practically need a masters degree in physics to understand it, or at least Egan perpetrated enough of a snow job to make you think you do. It sorta reminded me of a Dilbert cartoon: "You'd be a fool to ignore the boolean anti-binary least squared approach!" Yeah.
Another problem with the book was the fact that it just did not feel like a unified narrative. It has an "episodic" feel to it. I was all prepared to use that term when I saw on the inside cover a quote from Science Fiction Weekly that likewise referred to it as "episodic", so I guess I'm not the only one that felt Egan really didn't completely knit all of these disjointed events into a smooth whole.
But there was some good stuff too. Egan threw in some descriptions of alien life that were, strictly speaking, gratuitous. But they added flavor to the novel. I found that much more interesting than the pure physics rambling. Perhaps because I could really understand it and because it didn't take up huge volumes of space. While the description of how an AI was created in a polis was bland, the first few experiences of the fetal AI in question were interesting, and I thought the characters were generally well done. As usual, Egan was rather economical with space. Despite the scope of this novel, it only ran to 400 pages. I'm sure that once you grant Egan the necessity of including his scientific theories, he didn't use a lot of bloated prose to get them across.
Obviously people who are fans of pure scientific speculation are going to love this book. I like "hard SF" as much as the next guy, but I'm obviously not into science for science's sake. A little less pseudo-science and a little more work on plot integration would have gone a long way towards making me like this novel more.
BTW: I read a few other reviews that slammed the ending of this book. I disagree with them and rather liked the ending myself.
%A Egan, Greg %T Diaspora %I HarperPrism %D 1999-11 (original publication 1998) %G ISBN 0-06-105698-3 %P 403 pp. %O mass market paperback, US$6.99
Reviewed on 2000-07-30
Just Say No to Frames, Ads, and Animated GIF's