Soft vs. hard science fiction (definitions)

[J Riff]

Mushy/Very Soft/Soft/Medium/Firm/Plausibly Hard/Very Hard/Ultra(diamond)Hard
oh, and Science Fantasy.
Now we know.

 

[Vertigo]

I must admit I tend to ignore hard and soft in descriptions of SF books as almost everyone seems to have slightly (or hugely) difffering interpretations. As shown by this thread.

However, internally if you like, I tend to consider any sf book that has a stong emphasis on the technology as hard and if the technology is secondary to, well, "softer" themes then I consider it soft.

Of course, as with any pigeon-holing, almost every sf book out there falls somewhere between the two, I doubt there is any book that could truly described as purely soft or purely hard.

 

[J Riff]

Just ran into a Hal Clement essay "The Creation of Imaginary Beings' Here's Hal:

I am confining my remarks to the rather narrow limits of "hard" science fiction, where I am qualified to hold a professional opinion. It has been charged that in restricting ourselves to "scientific accuracy" my colleagues and I are narrowing the scope of usable story ideas available to us. My answer, mathematically rather horrible but defensible under literary standards, is that the square root of infinity is not really that much smaller than infinity as far as resource material goes. Our main point is that for many modern readers, a violation of the laws of thermodynamics by the author can spoil a story just as effectively as having Abraham Lincoln changing a set of spark plugs in a historical novel.
Therefore, if we travel to Mars in a story, the vehicle must operate either along physical laws we currently think we know, or at least on more or less convincing extrapolations of those laws. Furthermore, when we get there the Martians, not to mention their lapdogs, saddle horses, dinner steaks, and rheumatism, must not strike too jarring a set of notes against the background which author and reader are, it is to be hoped, visualizing together. It is permissible and even desirable to take the reader by surprise with some of these details, of course. However, his reaction to the surprise should be the urge to kick himself for failing to foresee the item, rather than resentment at the author's ringing in a new theme.
It follows that the "hard" science fiction writer must have at least an informed layman's grasp of biochemistry and ecology.

It goes on in great detail... really worth a read for anyone wanting to write SF.

 

[The_African]

Just ran into a Hal Clement essay "The Creation of Imaginary Beings' Here's Hal:

I am confining my remarks to the rather narrow limits of "hard" science fiction, where I am qualified to hold a professional opinion. It has been charged that in restricting ourselves to "scientific accuracy" my colleagues and I are narrowing the scope of usable story ideas available to us. My answer, mathematically rather horrible but defensible under literary standards, is that the square root of infinity is not really that much smaller than infinity as far as resource material goes. Our main point is that for many modern readers, a violation of the laws of thermodynamics by the author can spoil a story just as effectively as having Abraham Lincoln changing a set of spark plugs in a historical novel.
Therefore, if we travel to Mars in a story, the vehicle must operate either along physical laws we currently think we know, or at least on more or less convincing extrapolations of those laws. Furthermore, when we get there the Martians, not to mention their lapdogs, saddle horses, dinner steaks, and rheumatism, must not strike too jarring a set of notes against the background which author and reader are, it is to be hoped, visualizing together. It is permissible and even desirable to take the reader by surprise with some of these details, of course. However, his reaction to the surprise should be the urge to kick himself for failing to foresee the item, rather than resentment at the author's ringing in a new theme.
It follows that the "hard" science fiction writer must have at least an informed layman's grasp of biochemistry and ecology.

It goes on in great detail... really worth a read for anyone wanting to write SF.

I don't understand why a story should focus on scientific detail/accuracy rather than interpersonal relationships, the characters and their struggles, social themes and the plot. I thought stretching one's imagination and relating with characters was the whole point of storytelling, especially speculative storytelling.

Given that most sf involves some form of literary device - AI, FTL, time travel, etc. - that "breaks" the laws of physics, that would make the genre a bit heavy on the soft side.

You could have a hard science fiction story that dealt with AI, couldn't you?

You'd be better off to stick to the accepted definitions.

What are the accepted definitions?

I doubt there is any book that could truly described as purely soft or purely hard.

I think there are but I agree that the classifications aren't black and white, they're a matter of degree. Not all stories can neatly fit into 'fantasy' or 'science fiction' either and even some non-speculative stories contain speculative elements.

 

[J Riff]

True, but most stories in any genre have to be based on interpersonal relationships, the characters and their struggles, social themes and the plot, or they won't get published.
It's only the definitons, originating with people like Gernsback or Clement that set SF apart.

 

[Teresa Edgerton]

Ursa, Great site. This is exactly what I was aiming at, but would never have taken the time or trouble to sort it out. Thanks!!

Quite beside the point. It doesn't hinge on what is realistic. Is a horse less realistic than a spaceship?

Say that a story is about a generation ship heading for a earth-like planet orbiting a distant star. Sixty years into their mission (or any number of years you prefer), certain systems on the ship malfunction, and the story is largely about how the people on the ship race to solve the malfunctions based on known scientific theories and technologies. There is a little bit about the toll the effort takes on those who are engaged in solving the problems, and the toll on their families, but the vast majority of the book is given over to discussions about what they can do, to their first attempts to put their plan into effect, unforeseen difficulties, more discussions, new solutions, new attempts. Finally they are successful and the ship continues on. At the end of the story, the grandchildren of the previous characters see the surface of the planet appear on their computer screens. Hard science fiction.

Another book is about the aftermath of a similar ship landing on a similar planet. As well as humans born on the ship, there are frozen embryos of humans and animals, all of which are used in establishing a colony. For one reason or another (those in the colony would never need to know why) they lose contact with earth. A thousand years later, the colony has evolved (or degenerated) into a nomadic society of loosely allied clans who travel on horseback. The horse holds a special place in their culture, as transportation, status, totem, etc. The story is about that society. Soft science fiction.

The question was, how are the terms "hard" and "soft" science fiction defined. For a good many years and over a wide community of science fiction readers they've been defined in a certain way, and the definitions don't become disputed just because those of us here in this thread can't agree on them.

As for Tom Swift and the Megascope Space Prober, it may not be good science fiction (I haven't had the pleasure of reading the book myself, so I can't say for certain, though I suspect not), but whether something is hard science fiction or soft science fiction isn't about the quality ... although those who prefer the former might like to think so.

 

[iansales]

What are the accepted definitions?

hard sf = science fiction based on the "hard" sciences, eg, physics, chemistry, cosmology, celestial mechanics, etc. Such as Mission of Gravity by Hal Clement.

soft sf = science fiction based around the "softer" sciences, eg, anthropology, psychology, perhaps even politics and economics. Such as The Left Hand of Darkness by Ursula K Le Guin.

Obviously, the terms are not all-inclusive. It isn't an axis on which all sf can be placed. The terms are only generally descriptive, although in the case of hard sf it's often treated as a subgenre in its own right.

Some people seem determined to poison the waters by using their own private definitions for the two terms. The above are what have been generally accepted to be their meanings for at least the last thirty-five to forty years. People may argue how hard, or how soft, a particular text is, but that's not in reference to the accuracy of the science in the text. Science fiction does not mean fiction about science. Scientific accuracy is not a defining characteristic of the genre, and never has been.

 

[Parson]

Quite beside the point. It doesn't hinge on what is realistic. Is a horse less realistic than a spaceship?

Say that a story is about a generation ship heading for a earth-like planet orbiting a distant star. Sixty years into their mission (or any number of years you prefer), certain systems on the ship malfunction, and the story is largely about how the people on the ship race to solve the malfunctions based on known scientific theories and technologies. There is a little bit about the toll the effort takes on those who are engaged in solving the problems, and the toll on their families, but the vast majority of the book is given over to discussions about what they can do, to their first attempts to put their plan into effect, unforeseen difficulties, more discussions, new solutions, new attempts. Finally they are successful and the ship continues on. At the end of the story, the grandchildren of the previous characters see the surface of the planet appear on their computer screens. Hard science fiction.

Another book is about the aftermath of a similar ship landing on a similar planet. As well as humans born on the ship, there are frozen embryos of humans and animals, all of which are used in establishing a colony. For one reason or another (those in the colony would never need to know why) they lose contact with earth. A thousand years later, the colony has evolved (or degenerated) into a nomadic society of loosely allied clans who travel on horseback. The horse holds a special place in their culture, as transportation, status, totem, etc. The story is about that society. Soft science fiction.

The question was, how are the terms "hard" and "soft" science fiction defined. For a good many years and over a wide community of science fiction readers they've been defined in a certain way, and the definitions don't become disputed just because those of us here in this thread can't agree on them.

As for Tom Swift and the Megascope Space Prober, it may not be good science fiction (I haven't had the pleasure of reading the book myself, so I can't say for certain, though I suspect not), but whether something is hard science fiction or soft science fiction isn't about the quality ... although those who prefer the former might like to think so.

T.E. I feel duly chastised. Your point is excellent. Realism can be a strong part of a hard or soft SF story. And perhaps the scientific possibility of what is being described is quite irrelevant. It certainly is irrelevant as to whether it is a good story or not. I've read both kinds and been blown away by the story, slogged through the story, or (rarely) threw the book down in disgust. My reactions haven't depended on the level of the realism of the science; although I might have a bit more patience with someone who is dealing with more realistic science because as someone whose degrees are in social science I have to be brought up to speed on esoteric physics and the corresponding math.

I also agree that any story, whether it deals realistically with science or resorts to what can only be understood (at this time) as magic, rises and falls on the interpersonal relationships and the way the protagonists deal with the crisis that forms the center of the story line. This is why I still am hard put to accept what Ian says are the "accepted definitions."

The furtherest I can see myself moving in that direction is to say that the nomenclature is irrelevant. Leaving only SF which is more or less dependent on what we know at this time to be realistic science.

 

[iansales]

I'm not sure I follow. There's good sf and there's bad sf. And that can mean well-written sf, or poorly-written sf; as well as meaning the sfnal component of the story is good, or the sfnal component of the story is bad (ie, "skiffy"). I'm not sure scientific accuracy factors into that. Alastair Reynolds' novels, for example, are both "good sf" and "good sf" - they're well-written, with dramatic plots and well-rounded characters. And while there are things in his books which do not slavishly follow known scientific principles, the stories very much hinge on their science-fictional components.

I suppose it's confusing to refer to something as "good" in reference to two entirely different and unrelated criteria, and we probably shouldn't do it. But that's a separate distinction to hard/soft sf, or even to the argument as to whether a story must contain realistic science (which would actually make it Mundane SF).

 

[Parson]

Ian,

For me whether the center of a plot revolves around one of the so called "hard" sciences, or one of the so called "soft" sciences, does not describe SF literature. In a very real sense all literature revolves around relationships, which be your definition would be a "soft" SF. Therefore, there would not be any truly "hard" Sf by your definition; unless it would be some fictional physics text book or the like.

So for me (and I've always assumed) a lot of others, Hard SF related to the realism of the physical science included in the book.

 

[iansales]

But "relationships" isn't a science. I never said it was. A "soft" science would be anthropology, or psychology. Writing a story about two people and their relationship is in no way scientific - because, as you rightly point out, it's something all modes of fiction do. Drag in neurochemistry, however, and speculate how that relationship might be affected by neurochemical interventions by another person, or the state, or an alien envorinment... and then you might have hard sf,

 

[Fried Egg]

Parson, I'm with Ian and Teresa on this one, the difference between "hard" and "soft" SF is not about realism but is about the whether the focus is on the (so called) "hard" and "soft" sciences. That's standard usage anyway.

 

[Ursa major]

Before battle lines are drawn, can we accept that except in some specific circumstances, it doesn't really matter why someone thinks the book they're reading counts as soft or hard SF. Unless the book suddenly changes course some way in, it will mostly be clear from early on whether the book is well/badly written, which scientific fields it concentrates on, and whether that science is plausible/risible.

Surely the strict definition only matters to those in the publishing industry (including authors). An author may need to know exactly what the terms mean when they're considering submitting their book to an agent or publisher who asks to know something about the book (if only to say they don't represent/publish that type of book). The general reader doesn't; if this was important, the appropriate term would be printed somewhere on the cover.


By the way, I've just read The Drowned World by J G Ballard. I was expecting a lot of climate science, if not physics and chemistry, but the main sciences driving the plot were psychology and (a kind of) evolutionary biology. If I'd known, I would still have read the book. That I found some of the science implausible was independent of the particular fields.

.

 

[The_African]

hard sf = science fiction based on the "hard" sciences, eg, physics, chemistry, cosmology, celestial mechanics, etc. Such as Mission of Gravity by Hal Clement.

soft sf = science fiction based around the "softer" sciences, eg, anthropology, psychology, perhaps even politics and economics. Such as The Left Hand of Darkness by Ursula K Le Guin.

I didn't think that whoever came up with the terms 'hard/soft' science fiction meant it literally (ie. fiction related to the soft sciences or the hard sciences). After all, referring to the social sciences as 'soft' is almost pejorative, isn't it?

A physics based science fiction story might involve time travel or parallel universes, for biology, I could write a story about genetically engineered humans or an alien society on another planet or whatever but if I were to write a story about a human society with a drastically different culture or social norms than any we're familiar with but no speculative technology or any kind of event that we think of as impossible, speculative or unlikely, I don't really see how that's 'science fiction' (despite relating to sociology), personally. I haven't read it, but Left Hand of Darkness deals with interplanetary travel, doesn't it? So it has to do with astronomy, which is a hard science, but I'm assuming the book doesn't go into scientific detail and may not be entirely accurate.

Obviously, the terms are not all-inclusive. It isn't an axis on which all sf can be placed.

I agree, however you define hard and soft science fiction.

The terms are only generally descriptive, although in the case of hard sf it's often treated as a subgenre in its own right.

Why shouldn't it be?

Some people seem determined to poison the waters by using their own private definitions for the two terms. The above are what have been generally accepted to be their meanings for at least the last thirty-five to forty years. People may argue how hard, or how soft, a particular text is, but that's not in reference to the accuracy of the science in the text. Science fiction does not mean fiction about science. Scientific accuracy is not a defining characteristic of the genre, and never has been.

bolded : I agree but some people prefer science fiction that is scientifically plausible and/or explains scientific detail. So to specify what kind of science fiction they like, what should they say?

By the way, I think The Man in the Moon by Francis Godwin might be the earliest science fiction novel published in English (1638). Does anyone know of any science fiction novel or story that predates that?

 

[Parson]

Surely the strict definition only matters to those in the publishing industry (including authors). An author may need to know exactly what the terms mean when they're considering submitting their book to an agent or publisher who asks to know something about the book (if only to say they don't represent/publish that type of book). The general reader doesn't; if this was important, the appropriate term would be printed somewhere on the cover.

Now this is something I can understand. These definitions are used in the publishing industry so that the two parties can understand more clearly what one or the other is asking for or offering.

I will accept Ian's "accepted definitions" on this basis. I don't believe that the common public sees it this way. And I suspect that when "Hard SF" is used to describe a book in a book review the common reader would understand it more like my default understanding.

 

[iansales]

I have been reading science fiction for more than thirty years, I have been an active fan of the genre more than twenty. And it all that time, hard sf and soft sf have always referred to the sciences which predominate in a story. This is the way it is. You can't change a definition simply because you alone disagree with it. I might decide the US flag's red, white and blue is actually pink, purple and yellow. That doesn't mean it is. My wants don't change it. Nor does any assumption I might make about what the "common public" might think, or what the originator of the term might have meant. When you make reference to terms in common usage, you don't get redefine them on the spot. Because then a) no one knows what you're on about, and b) you start discussions like this one

They're not my definitions, they're commonly-accepted definitions by genre fans, readers and commentators, and have been for decades - "hard" longer than "soft". There's even an anthology, The Hard SF Renaissance, and it contains a lot of stories which are scientifically inaccurate/implausible. But that doesn't matter, because the editors chose stories whose central conceit revolved around a "hard" science. If you looked, er, hard enough, I suspect you'd find many similar anthologies, going all the way back to the 1940s perhaps.

I don't understand why you have such trouble accepting that these definitions exist. They may not be applied rigorously, and people may well categorise different books as belonging to one or another, but they're in general usage as per the definitions I posted earlier.

Incidentally, it's the predominant science which dictates the label. So even though Genly Ai arrived on Gethen in a starship - via NAFAL interstellar travel - the book is chiefly about Ai's exploration of Gethen's culture (and his relationship with Estraven). Hence, it is usually considered to be "soft" sf.

 

[iansales]

... some people prefer science fiction that is scientifically plausible and/or explains scientific detail. So to specify what kind of science fiction they like, what should they say?

That would be Mundane SF. See here.

 

[Ursa major]

I'm afraid that words (and phrases) mean what people want them to mean. There is no English equivalent to the Académie française**. People use words and phrases to suit their own purposes all over the place. Words and phrases change their meaning over time - not that I'm suggesting that this is what is happening here - or are given different meanings by the cognoscenti and the general public (which is probably what is happeneing here***). For instance, most Classical music isn't; not strictly. But people manage to put up with the looser usage, however irritating it is.

As I said: if this was so very important, it would be printed somewhere in many or most SF books.




** - And I'm afraid you'll have to be around for more than a few decades to be counted as an immortal.

*** - Though I expect a lot of the general public are with you on this, Ian.

 

[iansales]

As I said: if this was so very important, it would be printed somewhere in many or most SF books.

It has been - as per the anthology The Hard SF Renaissance I mentioned. There's also The Ascent of Wonder: the Evolution of Hard SF.

I don't actually understand how you can categorize a book by the accuracy of its scientific content, as that would require every reader to be a scientific genius - otherwise one reader might decide a book was accurate and so "hard", while another had found numerous mistakes and so declared it "soft".

 

[iansales]

Since I am now back at home after work, I have looked up the terms in The Encyclopedia of Science Fiction, which I think all of us can agree is a definitive source.

Interestingly, it does say that hard sf can mean sf based on the hard sciences, but also that it is sf which "uses either established or carefully extrapolated science as its backbone" (Allen Steele). It also says that the scientific spirit is perhaps more important than real science, and that it should eschew "supernatural or transcendental explanations for the events and phenomena it describes".

OTOH, soft sf it clearly defines in reference to the soft sciences - "sf that deals with the soft sciences or to sf that does not deal with recognizable science at all".

It would seem that several definitions exist for hard sf - which is not surprising as it's the older term - but soft sf was coined in response to only one of those definitions. Which means - some humble pie being consumed here - to describe hard sf as "sf that is scientifically accurate" is a valid usage.

But. The original question asked for definitions of hard and soft sf. As since soft sf has a relatively straightforward definition, then the definition of hard sf under discussion would be the one in opposition to it.