Hard sci-fi is not better sci-fi, consistent sci-fi is better sci-fi. You can have teleporters in your science fiction setting, but you need to decide how they work and stick with it. Don’t pull a Star Trek and have them both able to teleport through shields and not able to teleport through shields, create doppelgangers and not treated by anyone like a device that creates doppelgangers.
This has been an issue from the Golden Age on. The first major in-field critics, Damon Knight and James Blish, fought for more than good writing, deep characterization, and solid plots; they wanted science fiction to be about science. Knight, the founder of the Science Fiction Writers of America, felt so strongly that he didn’t want fantasy writers to be a part of it.
Looking back, though, that era included major writers like Ray Bradbury, Kurt Vonnegut, C. L. Moore, Leigh Brackett, Richard Matheson, Fredric Brown, Frederik Pohl and C. M. Kornbluth, William Tenn, Theodore Sturgeon, and a dozen others whom I should be remembering. None of them specialized in hard science fiction; most hardly wrote any at all. Of the titles picked for *American Science Fiction: Nine Classic Novels of the 1950s *, how many of them are hard, how many soft, and how many a sort of blend?
[ul]
[li]Frederik Pohl and C. M. Kornbluth, The Space Merchants[/li][li]Theodore Sturgeon, More Than Human[/li][li]Leigh Brackett, The Long Tomorrow[/li][li]Richard Matheson, The Shrinking Man[/li][li]Robert A. Heinlein, Double Star[/li][li]Alfred Bester, The Stars My Destination[/li][li]James Blish, A Case of Conscience[/li][li]Algis Budrys, Who?[/li][li]Fritz Leiber, The Big Time[/li][/ul]
We don’t think of dividing them in that way, usually. They’re all classics and lumped together.
You might argue that these are literary titles and rated on a different scale. That’s true in many ways. A large segment of readers have always preferred gosh-wow books without caring in the least whether they had good writing, deep characterization, and solid plots. That’s just as true today as it was in 1926 when Hugo Gernsback started Amazing Stories.
It’s also true that both writers and fans thought those nine titles to be great at the time they first appeared. And it’s true that during the 1950s, *Astounding *magazine cared less about characterization than about ideas, but the new contenders, F&SF and Galaxy, deliberately sought literary and social values. *Astounding *always had more circulation, though.
My personal tastes run to soft or a blend. Hardness needs so much continual structure and reinforcement that it gets in the way of good writing, deep characterization, and solid plots. The writers lumped, often mindlessly, together as the New Wave of the 60s saved the field. (No, really. Read the fanzine commentary from 1960, especially the one called Who Killed Science Fiction?) Zelazny, Delany, Ellison, Disch, LeGuin, Brunner, Silverberg, and those dozen should-be-named others did soft and blends that were fantastic in lots of senses including how wonderfully written they were. As you implied, MrDibble: just write the best stuff you can without putting up artificial walls to scale.
QFT. “Hardness” is a bug, not a feature.
Yes, it could. 
Welsh rabbit, by definition, contains no actual rabbit. While only hardcore purists insist that science fiction must be all about science and scientists, it’s perverse to insist upon the opposite: that it can contain no actual science.
Reading the reactions and thinking about this some more, it occurs to me that people have a hardness limit. Someone who is interested in and knowledgeable about science and technology will appreciate those things done well in stories, while someone who knows how to operate a light switch but doesn’t have a clue how it works won’t see a difference been Star Trek technobabble and The Martian sciencing the hell out of the situation, and thus won’t like it that the latter spends a good amount of time on that.
Within that limit, a story needs to be told, so liberties can and should be taken. After all, nobody is interested in a scientifically accurate retelling of an average day in the life of an average human. SOME unusual and therefore implausible situation is needed to make the story interesting.
If a story solves a plot point using real science, that’s better than plausible science or technology which in turn is much better than nonsense science or magic for the part of the audience whose “science bar” is high enough. And as we’re doing science fiction, that will be a non-negligible part of the audience.
In other words: nobody gets turned off by the science being too accurate, but tons of people get turned off by the science being inaccurate (assuming the length and dryness of the explanation is the same).
In that sense, harder is always better. But that doesn’t mean that boring science trumps exciting non-science.
Well, it’s perverse to insist that it must contain no science. That it can contain no science is proven by the existence of Ray Bradbury.
Excellent post, iljitsch. It occurred to me while reading it that this:
can also work the other way around. That is, there are people who have been turned on to science by reading science fiction. It’s something that (some!) good science fiction can do: you read a story, and it piques your interest and makes you want to learn more about the science involved. This isn’t something that all SF can or should do, but when it happens, it’s a plus.
Well, yeah; that’s what I meant—thanks for the clarification.
[Martin Prince]I’m aware of his work…[/Martin Prince].
I have read arguments that Bradbury’s stuff shouldn’t be called “science fiction,” because it contains no science. I don’t subscribe to this view, but I am somewhat sympathetic toward it.
By the way, I noticed the OP never actually uses the words “science fiction,” but always the ambiguous abbreviation SF, which could stand for “science fiction” or the broader “speculative fiction.” I wondered whether that was deliberate; though the claim that “harder is better” really doesn’t hold up if you’re looking at all speculative fiction.
I think San Fran’s gone soft (seriously did not know what the OP was talking about until the subsequent posts and the title definitely didn’t say it all).
More on-topic, a mix of practical/magical tech makes for interesting reading that the reader can connect with the present day but still suspend disbelief around is always a plus in my book. The whole **Saga of the Seven Suns **got too whack-a-doodle for me with its deus ex magica elementals (let’s just put everything together and call it epic!!). Piers Anthony’s **Apprentice Adept **series however seemed to blend magic/sci-fi very well while keeping clear delineations.
OP’s question, please. The OP’s theory is that it doesn’t really matter, which seems to be the general consensus here.
Which does lead me to wonder why people do sometime use “hard” as a synonym for “worthy”
This is the Cafe, if I wanted to talk about the Castro, I’d do it MPSIMS or IMHO 
If that is your reading of the consensus, then we are in consensus about what consensus means. (I.e., something different than what the dictionary says.)
I remember a Phillip K Dick (short?) story set in the distant future, where someone recieved a message via courier. The message was on a reel of film, which he placed in a film projector that made the typical “click-clack” 16 mm projector sound but projected a holographic image of the sender that spoke as if he were right there.
What? He could imagine a hyper-realistic holographic image but assumed it would just be an incremental improvement on the movie projector he was familiar with?
Is there a name for that kind of thing?
TV tropes always has the answer.
My apologies for the mischaracterization and mispunctuation.
Some factor of it is the overlapping range with Sci-Fi type fiction and fantasy. The desire for a purer form of SF tends toward the ‘hard’ end of the scale. Part of me likes the hard stuff, but like the others it I see it as often limiting the work instead of enhancing it. There is some area that’s the sweet spot for me where the science that is there has a solid base in reality while still including some flexibility for convenient concepts like FTL that don’t need to be fully explained.
The Martian Chronicles was published in science fiction magazines, and contains sciency stuff for the day. Bradbury clearly didn’t understand it and much of it was laughably wrong (by which I mean all of it) but I don’t think you can exclude it.
He also wrote lots of pure fantasy.
Thanks! That was fun.
Speaking of abysmally bad YA sf from the 50s and 60s…
I have a bunch of the original Tom Swift books, written by the guy who wrote Uncle Wiggly. If you ignore their background racism (though less racist than early Nancy Drew) the science in them, such as it is, was reasonably good. Very few total howlers.
Compare to the Tom Swift Jr. books from the '50s and '60s, which had absolutely no concept of conservation of energy to mention a mild problem. Quite dreadful. I’m glad that I never read them as a kid, I would have been ruined for real science.
Yet superficially they seemed harder than the original ones.
Psi. Yeah. FTL? I think Baxter counts as hard sf, and he uses FTL. The difference between hard and soft use of FTL is that hard sf understands why FTL is impossible with our understanding of physics, and finds some way around this - with varying degrees of plausibility. Soft sf just ignores the problem.
For its time ST:TOS was relatively hard for TV sf since it knew that FTL travel could not be achieved by pressing harder on the accelerator. It also never had a small asteroid with an atmosphere.