Hard SF : About the Genre : SF & Consistency With Science
There are those who strongly object to any criticism that an SF story is inconsistent with our scientific understanding. To me, there's a time and a place for everything. I don't think that every story that might be found on a bookshelf labeled "science fiction" must be consistent with every bit of science. However, I think there are times when it is reasonable to feel a story did not do justice to science. I think that pointing out a story's divergence from scientific fact can be useful, but that does not mean the story isn't worth reading.
I wouldn't say that a fantasy story shouldn't have magic stones that fall up. I don't check the science in fairy tales, Alice In Wonderland or A Connecticut Yankee In King Arthur's Court. Even in books I'd call SF, there are books such as Douglas Adams' books, that shouldn't be taken literally. If the Hitchhiker's Guide had a planet where things fall up, I'd know Adams was just having fun being silly.
The scientific methods used to establish the validity of scientific theories give us reasons to treat those theories as true. However, we also know that some aspects of current scientific understanding are likely to evolve in the future. Not every change in our scientific understanding that a writer might suggest are reasonably plausible, but some would be. A story that clearly speculates about a reasonably possible change in our understanding should not be treated as inconsistent with science.
Consider another example. Richard Garfinkle's book Celestial Matters is based on the idea that Alexander The Great's empire continued up into the space age. It also worked under the premise that scientific assumptions of Aristotle's time were valid. All things were composed of the "four elements" - air, water, earth and fire. The Sun circled the Earth. The Celestial Spheres existed. And so on. The science was clearly inaccurate - and no readers were expected to think it was valid. Presumably, readers were to conclude that this was not taking place in our universe. Within the book and its alternate universe, the world was internally consistent with these alternative rules.
I think internal consistency is a reasonable expectation for a reader (assuming one allows certain books to have a consistent comic disregard for rules). If I were reading a historical novel about Thomas Jefferson, I wouldn't want the historical error of having him type the Declaration of Independence into his PC and make copies with his laser printer. (Alternate history books may be a different matter, but those books are specifically intended to diverge from history as we know it.) Nor would I want the physical error of having Jefferson drop his quill pen and have it fell up. A historical novel takes place in our universe, so our laws of nature should apply.
As I hinted at above, not all speculative fiction or science fiction is presented as taking place in our universe or according to the laws of nature as generally accepted by our scientists. A science fiction story may choose whether or not to take place in our universe. Some examples of that were given above. However, there are SF stories that don't seem to suggest they take place in another realm that has different physics. For those stories, a case can be made that inconsistency with the laws of nature (as currently understood) amounts to internal inconsistency.
Just as a story that we would say is "alternative history" is not internally inconsistent if it does not maintain historical accuracy, a story in another universe (we might call it "alternative science") would not have to maintain scientific accuracy. (It should have consistency with the rules of that other universe.) Knowing when it's reasonable to expect consistency with history requires we distinguish between historical fiction and alternative history. A distinction between "science fiction" (or whatever we choose to call that which does not intentionally diverge from our universe) and "alternative science" would also be beneficial.
In the case of the historical novel in which Jefferson dropped his pen, the conflict with the law of gravity when the pen falls up isn't necessarily as significant as it would have been in an SF story. A historical novel, generally speaking, is not about science. As the genre name suggests, it's more related to history. And, therefore, I would expect historical accuracy would be of importance - to a greater extent than scientific accuracy.
Would it be reasonable for a reader to be disappointed if he read what was supposed to be a serious Western, but found the writer was so uninformed of the subject that the "cowboys" rode cows rather than horses, and gunslingers used slingshots to throw guns at their opponents? Is it wrong for someone who buys a "mystery" to object if the criminal is known from the beginning and the story is merely a long chase scene to arrest him? Or would it be appropriate for a book sold as a "fantasy" to tell how physicists develop affordable fusion power in 2020?
Not everyone agrees that just because sometime in the past SF was given the name "science fiction" that it's about science. I, myself, don't feel a need to exclude "future society fiction" which doesn't deal with science and technology from being grouped with "science fiction". Still, SF is the genre where stories which are about or depend on future technology and science will be found. Whatever you may want to call it, let us consider those works of fiction that seem to take place in our universe (or at least one with the same natural laws) and in which science / technology play an essential role. Why should a reader not expect the science / physics in the story to be consistent with what we understand to be true in the real world?
Realistically, even if we convince every SF author to write their stories in compliance with science to the best of their abilities, there would still be discrepancies with science. SF can involve more science than students generally get from pre-college education. Authors don't necessarily major in the sciences at college. And even if they did, they would not necessarily have a doctorate or a background in multiple science disciplines. So, I understand that a story's scientific errors don't mean the writer is a bad person or is indifferent to science. Regardless of that, I don't think that looking the other way is the best approach. I think it's a good thing to encourage efforts to increase the amount of scientific consistency in these kinds of books and/or among their readers.
SF books usually don't begin with a declaration of what universe or laws of physics the story operates within. Perhaps, there will be times the author was working under the assumption that the story took place in another universe and he thought readers would realize that - but some readers go through the book thinking it is in our universe. Misunderstandings like that can happen. I might mistakenly point out a book's inconsistency with our universe's laws in a case such as that. Please understand that in those cases I'm not questioning the writer's right to have stories in other universes or criticizing his choice to place that particular story in another universe. I'm only trying to promote more scientific consistency in those stories I am under the impression take place in our universe.
I also believe that if a writer does not have the necessary background to know whether a technology is plausible (or if he chooses to include a technology he knows isn't plausible), it is better for him not to give an "explanation" of how the technology works. At best, giving a description of how something works that is not based on actual scientific knowledge draws more attention to its inconsistency with nature – making the book seem flawed to some readers. At worst, it may leave some readers with the mistaken assumption the author was presenting established scientific knowledge.
Perhaps the above arguments are the most important, in that they deal with SF as fiction - and there are limits to how far we should go in treating a genre of fiction as something other than fiction. However, I'm not sure we should entirely abstain from further considerations. When a novel is presented as historical and is written in a way to leave readers with the impression that its historical events are accurate, I don't feel it is inappropriate to point out historical errors. (In a sense, genres of this sort are not exclusively fiction.) Perhaps, such errors are inevitable for even good historical fiction writers, but if we point out the errors, they have the option of making a correction in a later edition, or at least avoid repeating the same error in later books. And some readers will be saved from a mistaken historical understanding. I'm not saying all historical fiction readers will be especially concerned if there are some historical discrepancies in a novel. However, I think it's in the nature of genres such as historical fiction and science fiction that a significant part of the readership does care, and therefore this is a useful service to them. Those readers who are indifferent to historical and scientific accuracy do not have to read material that concerns itself with such discrepancies.
On the whole, I would say that the level of historical knowledge from school is probably stronger than what most people retain from their science classes. In that sense, SF may have a greater potential for mis-educating people. I think there would be a benefit to society if the amount of scientific inconsistency in SF was decreased. And even if the SF fiction itself doesn't change, the availability of accurate science explanations being contrasted with what appears in SF could help those readers who would like to know the truth.
I'd like to raise a related topic. There's quite a bit of SF that portrays people with telepathy or other psychic powers. Science does not claim to have proven such things can't exist - proving something (even fairies) can't exist is tricky. However, "mind powers" in SF troubles me for a number of reasons. For instance, the origin of belief in such things is in ancient myths and beliefs in the supernatural without corresponding historical backing. (For more on these issues, see the Psychic & Fringe Theories section.)
These themes are internally consistent in a "supernatural fiction" story. Such fiction assumes a universe in which phenomena which is at least widely doubted by scientists does take place in that universe. Similarly, if readers know a book is monster-horror fiction, they should assume it takes place in an appropriate universe.
Unfortunately, SF that includes "mind powers" often treats the phenomena in a way that doesn't seem to relate to the laws of nature. Telepathy over huge distance doesn't tend to be limited by the speed of light. Rarely does one read of a hi-tech civilization which has devices capable of interacting with telepathy by using the same physical processes the telepathy does. Telepaths seem to magically communicate only with the individual they want - unlike communicating by radio transceivers where everyone on that frequency can hear what you say. And so on. The laws of nature do not forbid the evolution of a body organ that would function like a two-way radio, but that would operate consistently with the laws of the universe. This consistency with nature simply isn't what one usually finds when mind powers are included in what would otherwise seem to be SF in our universe. In effect, these stories are incorporating magic rather than science.