Notes from three lectures on literary realism. I'm already so far behind. Terrible, terrible.
"Realist" fiction is ostensibly more "real", but what does that actually signify? The text is still a construct, and its apparent "realism" is the result of specific techniques. We have several occasionally contradictory metrics of "reality":
Empirical: reported by the senses. We believe in what we can see and touch.
Endorsed by authority: "endorsed by science", if you prefer, but that's too narrow. We're willing to accept arguments from trusted authorities, even when they're not supported by our senses, e.g. that crazy talk about little deelies called "atoms". Our notions of the acceptably real vary drastically according to time and place, but in general it's true that an appeal to an unverified or unverifiable authority can add credence to a text.
Detail: the reader is more easily convinced of a detailed setting. Immediate sensory detail is very real, as is detail of the sociohistoric setting. Enough jargon-filled details of science fiction pseudoscience and the reader is likely to assume that it's true, or at least plausible. This is often a form of authority: the reader will lack either the inclination or ability to verify those details.
Possibility: it's possible for a man to win the lottery; it's impossible for him to sprout wings and fly. This may be the most important criterion: anything the reader believes is possible is fair game. Anything the reader can be convinced is possible is fair game. We're already remarkably inured to strings of wild coincidences.
Probability: the man might win the lottery, but it's not that much more likely than the flight scenario. The more probable a sequence of events, the more realistic it seems.
Ordinary/Boring: stories about the mundane and normal seem more real. It's possible and probable that an emir will bathe in a gilded bathtub, but it's not ordinary -- for the reader, that is. A realist novel is more likely to star an accountant than a prince. The mundane can also add a realistic veneer to pure fantasy: the dragon-slaying hero who complains of the blisters from his heavy armour feels more realistic than the one who doesn't.
Truth beneath the surface: realism in the sense of dispelling an illusion. It's the equivalent of the rhetorical trick of setting up a straw man just to knock it down -- the text achieves a measure of reality by comparison with the illusions it purports to shred. E.g., chapter one might give a picture of domestic bliss, while chapter two shows the abusive environment behind closed doors. H.P. Lovecraft's "At the Mountains of Madness" tells the reader that the previously released details of the Antarctic expedition were horribly wrong; the story, of course, has the true facts. The reader is tricked into accepting one lie over another.
Unrealistic fiction can be defined by contradistinction to one or more of these factors; speculative fictions as defined in this course are in contradistinction to the endorsement of science.
Remember Clarke's suggestion that any sufficiently advanced technology is indistinguishable from magic. Science fiction often seems more realistic than fantasy because "the marvels of science" is an acceptable excuse for all sorts of fantastic devices.
I'm not sure what exactly Aristotle meant by "the poet should prefer probable impossibilities to improbable possibilities" (S.H. Butcher (tr.), Poetics), but the point seems to be that the reader can swallow utter absurdities if they're probable in context, but the improbable always jars. The impossible idea of the dead rising from their graves to eat the flesh of the living is more acceptable than the merely improbable suggestion that the rest of the population would subsequently invite them over for tea and scones.
Serious speculative fictions use the realist techniques to make the impossible seem real, but the contrast is also exploited for humour value, e.g. the absurdity inherent in the musclebound superhero Mr. Incredible working 9-5 as an insurance salesman.