Brian Aldiss and David Wingrove in Trillion Year Spree offer us this:
Science fiction is the search for a definition of mankind and his status in the universe which will stand in our advanced but confused state of knowledge (science), and is characteristically cast in the Gothic or post-Gothic mode.I have limited patience with this kind of literary jargon (I'd be pleased if someone would contribute an explanation of Aldiss' definition that makes sense to a benighted engineer; however, most of the book is quite understandable). But I have less patience with people who think ScF means any book that has a spaceship on the cover, and none at all with those who say it is "what we point to when we say `science fiction'" (this, attributed to Damon Knight, is IMHO just a refusal to explain how he defines it -- see also the passage about teaching `quality' in Zen and the Art of Motorcycle Maintenance).
Poul Anderson suggested, in an afterword to one of his books, that an ScF story should be allowed only one "fantastic" assumption, by which he meant one that contravened the laws of science (as known to the author, presumably). In that book, FTL travel was his assumption, and he offered arguments as to why the other futuristic elements of the story were plausible, or at least not utterly infeasible. I neither agree nor disagree with this; it helps keep things interesting.
In Robert Silverberg's World of Wonder Silverberg says that ScF requires:
- An underlying speculative concept, systematically developed in a way that amounts to an exploration of the consequences of allowing such a departure from known reality to impinge on the universe as we know it.
- An awareness by the writer of the structural underpinnings (the "body of scientific knowledge") of our own reality, as it is currently understood, so that the speculative aspects of the story are founded on conscious and thoughtful departures from those underpinnings rather than on blithe ignorance.
- Imposition by the writer of a sense of limitations somewhere within the assumptions of the story. The magical ability to turn anything into anything without effort all too easily vitiates the tensions necessary to good fiction of any sort. On the other hand, subjecting the powers of magic to the laws of thermodynamics turns a story from fantasy to science fiction, though that story may be full of sorcerers in pointy hats.
The Australian author and critic Damien Broderick has something to say about this in an interview in Issue 14 of Ibn Qirtaiba, to the effect that ScF is a "mode" rather than a genre. Perhaps I'll re-read this some time and see if I understand it.
James Blish wrote in 1973 or 1974, under his critical pen name of William Atheling, Jr., an article in which he defined science fiction, adding that "it has thus far produced no towering literary masterworks, [and] no such work can be expected of it in the future." To understand his definition, though, requires understanding several pages' worth of Spengler's view of history, which Blish explains in the article, but which I choose not to repeat in full. The kernel of it seems to me to be this part of the description of the winter or death of a culture:
Technology flourishes ... but science disintegrates into a welter of competing, grandiosely trivial hypotheses which supersede each other almost weekly and veer more and more markedly toward the occult. Among the masses there arises a 'second religiousness' in which nobody actually believes; an attempt is made to buttress this by syncretism, the wrenching out of context of religious forms from other cultures, such as the Indian, without the faintest hope of knowing what they mean. This process, too, leads inevitably towards a revival of the occult, and here science and religion overlap, to the benefit of neither.The definition itself is: "Science fiction is the internal (intracultural) literary form taken by syncretism in the West."
I'm indebted to George H. Williams for drawing my attention to Blish's article.