In a recent comment thread over at Torque Control I said "There seems to be a common feeling with people coming into SF that you need to know real science to write good SF. Which is of course rubbish." The comment was in response to Catherynne M. Valente posting about writing Science Fiction.
Following that Athena Andreadis wrote a post titled Science Fiction Goes McDonald's: Less Taste, More Gristle on The Huffington Post and said she was surprised at the agreement with which my statement was greeted.
I think the key point here is that there are degrees of accuracy.
As an example Dr. Andreadis rephrases my statement to use historical fiction as an example:
Let me rewrite that statement for another genre: "There seems to be a common feeling with people coming into historical fiction that you need to know real history - or at least the history of the era you plan to portray - to write good historical fiction or alternative history. Which is of course rubbish."
Cell phones in a Renaissance novel? Tudor court ladies on mopeds? Why should anyone notice or care?
So first of all we'll make the assumption that the author of the historical novel is not attempting anything modern or funky or surreal and is going for an accurate historical novel. I agree that if I was expecting an accurate historical novel then phones and mopeds in a Tudor court would jar with me. However, I am not a scholar of history, or even very well versed in many periods of history. But then who is? Who is an expert in every slice of history? So when it comes to a historical novel such as The Baroque Cycle, guess what? I have no idea what Neal Stephenson made up and which bits were taken from primary sources. And I don't really care. The novel evoked the period to me, I loved the story and I could, if I wanted to, try and find more academic sources to read.
But even those academic sources have elements of doubt. Presumably much of history cannot be pieced together form primary sources? I don't know, I'm out of my depth talking about history already.
The same is true even when moving to science: this year I read A Quiet War by Paul McAuley, there's some interesting science in the novel: biology and ecology. To be honest I had no idea whether it was accurate or not. Like history, I know very little biology. It didn't really impact the story, McAuley sounded like he knew what he was talking about.
So let me switch to something I (used to!) know about, physics.
Say you want to write a Hard SF novel. You research the state of the art in string theory, or brane theory, or whatever. You stick to your exacting research. The number of people who will know whether you are accurate or not is tiny, a small group of physicists who have spent years buried in maths. And let us not forget that those theories are just that, theories! They could be disproved within a year if the LHC delivers some nice data.
Even with experimental physics, nothing is ever really proved to be the absolute truth. Physics is just a model of the universe that we can conveniently use. You can never discount that some experiment will come along and highlight a flaw in your glorious model. Yes, of course some models have gathered such impressive experimental proof that they can be taken as "correct", but my point is that it's not black and white.
So as a SF writer you have to decide how deep you want the facade to go. The deeper you go the narrower the audience that will be able to spot your flaws. If you want a shallow facade it does not mean you can't write Science Fiction, it is after all, fiction.
The big point you make here is one I was thinking of when I read that original article:
It doesn't matter how good your research is (in regards to science, anyway), the vast majority of readers won't know what's correct or not anyway. Most Americans, for example, barely understand the basics of physics or the other sciences.
I've always held that science fiction needs science, but it doesn't necessarily have to be accurate. What makes science fiction stories work is that the stories feel and read as though they are real, with good characters, a logical plot, etc.
So, yeah!
I started writing a historical novel in 1994, drawn by the possibilities of the period (17th century London ... smells ... executions). Finally finished it 14 years later. I found that in order to know where I could make stuff up, I needed to know about the stuff that many readers would expect me to get right. In working the book with readers and editors I discovered that a lot of readers of historic fiction (not all of course) absolutely demand you know your stuff. Fortunately I found the journey fascinating! Finally, in 2009, The Sweet Smell of Decay was published!
Yes, I understand, I'd still bet that the degree of accuracy is inversely proportional to the people who would notice the accuracy.
The key points, which I make in the part of the article that you did not quote, is that 1) fiction is indeed the dominant partner in SF, as it should be in all literary works, 2) accuracy doesn't matter as much as consistency and quality of imagination and 3) the science in good SF is integral to the story, not an extrinsic, optional gimmick to impress non-scientist readers.
Hello. Sorry, I didn't mean to misrepresent you. I completely agree with 1 + 2.
With 3, I agree that the science shouldn't be bolted on as an extra to "make it Science Fiction", and that it's integral to the story. However I (currently!) don't feel the need for that science or technology to be deeply accurate.