The British fanzine True Rat once published a definition of science fiction which is attributed to Peter Nicholls, and which Dave Langford claims is the only definition of SF that he can ever remember.
Sci-fi can be succinctly defined as speculation, whether based on established scientific facts or on logical pseudo-facts consistent with the framework of the fiction in question, involving smelly green pimply aliens furiously raping or eating, or both, beautiful naked bare-breasted chicks, covering them in slime, red, oozing, living slime, dribbling from every horrific orifice, squeezing out between bulbous pulpy lips onto the sensuous velvety skin of the writhing sweating slave-girls, their bodies cut and bruised by knotted whips brandished by giant blond vast-biceped androids called Simon, and written in the Gothic mode. (True Rat 7, 1976)
Notice the subordinate clause " whether based on established scientific facts or on logical pseudo-facts consistent with the framework of the fiction in question ". Its not a bad definition of science fiction, when you think about it.
The hardest of hard science fiction would, I feel, restrict itself to just the first part of that definition and would require that the story be firmly based on scientific fact. Im going to begin by examining that restricted definition. I feel that it is of limited use, for the number of stories we can point to under this rubric and say "Thats hard SF" are vanishingly small. And those few that do exist often prove to be dull.
Later Im going to extend the definition to include the "logical pseudo-facts consistent with the framework of the fiction in question". I think that small modification will give us much more room to manoeuvre and will allow the definition to embrace a much wider (and hopefully more enjoyable) set of stories without necessarily having to give up the claim that they are hard SF.
But lets deal with the really hard stuff first
The magazine Astounding (later Analog) had a reputation for publishing the hardest of hard science fiction; to such an extent that it actually suffered quite a large drop in its readership as its stories degenerated into trivial wiring diagrams with dialogue. When the whole point of a story is some arcane scientific fact (or implication thereof) things like plot, characterisation and drama tend to vanish and we are left with a thinly coated science lecture. So why not just have the lecture in the first place and ignore the fiction that is grafted on to it so badly that the joins show? Robert Forward was particularly bad at this. His novel Dragons Egg explores the implications of life on the surface of a neutron star. It is stuffed to the gunwales with fascinating scientific speculation, but is so badly flawed as a novel that it becomes almost unreadable.
It doesnt necessarily have to be that way, of course. Hal Clement was notorious for the months of work he put into the scientific calculations underlying his stories. Sometimes he even published those calculations. Many modern editions of his most famous novel Mission of Gravity have a long article called Whirligig World bound in as an appendix. It goes into great and grinding detail about the physical and mathematical assumptions behind the story. The article itself leaves me unmoved I have read far too many tedious scientific papers in my life but the story derived from it is superb. The novel is the hardest of hard SF and it would be a very brave physicist or mathematician indeed who would dare to argue with the science (though the basic premise might raise a few eyebrows). Furthermore I dont believe that anyone could seriously contend that it wasnt also a triumph of the storytellers art. Theres no doubt about it, hard SF in this sense can be made to work magnificently. But it is extremely difficult to do and for every Mission of Gravity there are a thousand novels like Dragons Egg.
So art and science are demonstrably not mutually exclusive. Nevertheless, I still contend that hard SF, in this sense, is almost non-existent.
Lets look at some so-called hard SF writers and see why they fit most uneasily into the pigeon hole we have made for them..
Larry Niven is often assumed to write very hard SF indeed. Even his fantasy stories (of which there are many) have a scientific feel to them. For example, What Use Is A Glass Dagger describes the dying days of magic on the Earth and we learn that the reason magic is dying is that it is powered by a strange force or substance that Niven refers to as mana and it has been over used and is running out. Theres a thinly disguised and rather clumsy message here; a plea for conservation if we dont take care of our non-renewable resources such as oil, they will vanish underneath us before we know it. Shouldnt we be looking for alternatives?
Given that the story deals overtly with magic and makes no attempt whatsoever to introduce a scientific (or even pseudo-scientific) mechanism behind the idea of mana there is no way that we can claim the story for the hard SF canon. Nevertheless Niven shows us his way of thinking here and extrapolates quite rigourously from his initial assumptions, untenable though those assumptions are in reality, and it feels very much like hard SF even though it isnt.
Paradoxically, his "proper" hard SF stories are built on much flimsier ground. Everyone knows, for instance that in Ringworld he had the Earth revolving in the wrong direction. After a schoolboy howler like that, who is ever going to take him seriously again?. Furthermore many of the plot devices are themselves sheer nonsense. A small, though significant, plot element revolves around the use of matter transmitters as teleportation machines. There are huge problems associated with such a mechanism. Even if you can come up with a plausible explanation of how the things work in the first place, practical problems involving the various conservation laws would suggest that the machines are likely to be quite lethal. Niven is fully aware of these practical difficulties he has written articles about them; see, for example, The Theory and Practice of Teleportation in his collection All The Myriad Ways. However despite knowing that the machines are thoroughly implausible, he felt no compunction whatsoever about using them in his story. Consequently there is absolutely no doubt that Niven is not writing hard science fiction as we have defined it. That doesnt make his work any less entertaining and neither does it invalidate the work in any artistic sense. But to claim that it has anything at all to do with science is ridiculous.
The big three Asimov, Heinlein and Clarke are also often considered to be hard SF writers par excellence, but I have my doubts here as well.
Asimov is easy I dont think he ever wrote a hard SF novel in the whole of his life, and on many occasions he cheerfully admitted to the charge. Positronic robot brains? Psychohistory? Neither idea is anything other than enormously clever arm waving with no basis whatsoever in reality as we know it (or, I would venture to suggest, reality as we will ever know it). Fortunately his robot stories arent really about positronic fluxes and his Foundation stories arent really about mathematical psychology applied to historical forces (except trivially) so it doesnt actually matter at all. His other stories and novels are also littered with common SF escape clauses like faster than light travel and time machines neither of which can be taken seriously. Their only real purpose is to move the story along so that something actually happens prior to the protagonists dieing of old age and decrepitude before their immense journeys have properly begun.
Heinlein and Clarke are more problematical. It is easy to point to many of their stories that are indeed firmly grounded in physical and mathematical fact even if sometimes their extrapolations from those facts veer off into (probably) untenable areas. Im very unsure that well ever see Clarkes space elevator, for example. But that doesnt invalidate the speculation and it certainly isnt grounds for claiming that the stories are not hard SF. Extrapolation from known facts, however unlikely that extrapolation may be, is a perfectly legitimate SF mechanism. Indeed it is one of science fictions primary distinguishing characteristics. My problem with both Clarke and Heinlein is that they have a tendency to intermingle this hard SF approach with more than a touch of mysticism.
Clarkes most famous short stories, The Nine Billion Names of God and also The Star both claim quite overtly that the universe is controlled by something that we can only refer to as God. Even novels such as 2001 and its many sequels leave the questions of the origin and motivations of the aliens open to a variety of interpretations, including a religious one. Early novels such as The City and the Stars and Childhoods End flirt with mystical explanations and influences and directly invoke a surprisingly evocative sense of yearning for the numinous.
Heinleins flirtations with the other world are less high-powered (and often much less artistically effective as a result). But stories such as Waldo certainly preach the message that there are more things in heaven and earth than are dreamed of in our philosophy. Specifically, in Waldo, the eponymous hero who is presented to us as some sort of super scientist is shown to invoke magic to solve his problems. Its heavily (and cleverly) disguised with a lot of arm waving. But nevertheless it is magic in the true sense.
However Heinleins very best novels have absolutely nothing whatsoever to do with science per se. Heinlein was far more concerned with politics and sociology and the majority of his mature work concentrates on these areas almost to the exclusion of everything else. And with the best will in the world, neither politics nor sociology can be considered to be a science.
Clarke is particularly famous for formulating Clarkes Third Law which states that any sufficiently advanced technology is indistinguishable from magic. Theres a lot of truth to this. Imagine what people from the end of the nineteenth century would have made of computers and cell phones? And thats only slightly more than a hundred years ago a mere eye-blink of time.
Indeed, my experience suggests that even many contemporary people consider computers and cell phones to be pure magic! Intellectually they may be willing to admit that the devices are proper technological miracles. But emotionally, in their heart of hearts, they have no real understanding of the technology and so it might as well be magic as far as they are concerned.
And this is the corollary to Clarkes law, of course. Any technology that IS distinguishable from magic must therefore be insufficiently advanced.
Both interpretations of Clarkes law can be used as an attempt to justify the mystical (or magical) and therefore to drag Clarkes own work and also Heinleins back into the fold of hard SF. But I think its a specious argument that simply papers over the cracks. Technology is always distinguishable from magic even if only because it never actually claims to be magic. Failing to make that distinction is profoundly insulting to the more abstract notion of human development itself. That distinction is one of the reasons why we arent still in the caves, banging the rocks together guys.
And this is where I take issue with that other modern icon of hard SF, Vernor Vinge.
Vinge and his disciples seem to believe as an article of faith in something they call the singularity. It is very difficult to pin down exactly what is meant by this because Vinge gets very vague when questioned and tends to lapse into incoherence. His grasp on the niceties of English prose is tenuous at the best of times and the essays on his web site that discuss the idea appear to have been written on a word-processor dipped in muddy water. However he does sum up his thinking in two sentences of crystal clarity:
Within thirty years, we will have the technological means to create
Superhuman intelligence. Shortly after, the human era will be ended.
Vernor Vinge. Vision-21 Symposium. 1993
As near as I can pin it down, Vinge seems to think that the human race is rapidly approaching an intellectual crisis where artificial intelligences (probably computer based though he doesnt rule out biological entities) are reaching a state where they are starting to work in areas that they "comprehend" but we dont. It is a crisis of intellectuality that will leave us behind, lost and gasping as the new intelligences grapple with concepts that are forever denied to us because their way of thinking is so foreign to us. The old models are thrown away and a new reality rules. The forces of change (both technological and societal) may come to be measured in hours if current exponential growth rates are any indicator. Neither we nor our current society will be able to cope with such rapidity of change and thinking. Therefore, says Vinge, contemporary pre-singularity fiction cannot be set after the singularity for it is literally unimaginable.
Furthermore, after the singularity has occurred it wont be possible to write science fiction any more. Indeed it may not even be possible to write any meaningful fiction at all because meaning will no longer be comprehensible in terms of the post-singularity world, whether it be us looking at them or them looking at us. The singularity is a great apocalyptic divide.
It seems to me that this is nothing more than a restatement of the old science fictional cliché that closed so many B movies. The hero gazes soulfully across the wreckage of the world that the alien invasion has left behind, and he says to the mad scientists beautiful daughter:
"There are things that man was not meant to know "
Vinge pretties it up with fancy buzzword-compliant phrasing, but stripped to its essence Im fairly sure thats what he means.
Its a complete failure of nerve. After all, every generation has at some point claimed limits on knowledge. The nineteenth century scientists were certain that every scientific principle was known and the science of the twentieth century would consist of nothing more than I-dotting and T-crossing. And when quantum theory and relativity blew their smug certainties to shreds, many denied the validity of the ideas.
"New fangled rubbish, never catch on."
Many of that generation went to their graves not accepting and not understanding the new ideas. In one sense this is a perfect example of Vinges singularity and a complete vindication of both it and of Clarkes Third Law.
But the generation that followed them, the generation who grew up with the new ideas, had no difficulty at all appreciating them. In many ways the ideas became a common coin which fed further revolutions. They were just a base on which to build. No singularity outlasts the generation to whom it happens and to claim that the post-singularity world remains forever beyond our grasp is demonstrable nonsense. You and I as individuals may not appreciate it, but those who come after us will. And they wont be divided from us any more than we are divided from our great grandparents. We didnt lose our understanding of history because of the change in our understanding of science.
Despite this, the idea of the singularity is often used as an excuse to shoehorn patent nonsense into the hard SF camp on the utterly specious grounds that it is too rarefied for us to appreciate it properly since it belongs to the post-singularity world. And this includes Vinges own writing of course. Hows that for self-glorification!!
In a way we are pursuing a willo the wisp. The above examples show that you simply cant bend the phrase "hard science fiction" to make it mean what at first glance it appears to mean. Scientific fiction is practically an oxymoron somewhat akin to military intelligence. Both are believed to exist, but the evidence is thin on the ground and counter-examples are legion.
However the phrase is used far too often to dismiss it out of hand, and it does appear to be a useful category, even if we are not completely sure what belongs there. So it must mean something. Perhaps a looser definition would help. And this is where the "logical pseudo-facts consistent with the framework of the fiction in question" come into play.
Based purely on empirical evidence, I would suggest that we consider the phrase hard SF to be just a yardstick that allows us to measure scientific philosophy rather than strict factual accuracy. Thus hard SF stories are stories that recognise that the universe is strange but which require that ultimately it be understandable in its own terms. If you like, hard SF is SF for the rationalist. And Asimov was a rationalist par excellence. His positronic fluxes and psychohistorical insights were simply window dressing.
In many respects it doesnt actually matter at all what the explanation turns out to be. It is sufficient that there IS an explanation with some degree of convincing rigour about it. Nobody requires that you have to believe it, and it doesnt even have to make sense in any realistic way. If it is done sufficiently well, we are generally more than willing to suspend our disbelief. And that, of course, still gives the mystics room to manoeuvre which finally allows Clarke and Heinlein back into the fold. Thank goodness.
Vinge is still left out in the cold but I really cant see any way of incorporating his passive fatalism (and huge inferiority complex) into the mix without destroying whats already there. Despite the cries of his vociferous fans, I dont think he is writing hard SF at all, however you want to define it.
I dont seriously expect that the answer to life the universe and everything will ever emerge from a science fiction story. Almost by definition fiction is derivative. Stories seldom if ever innovate in any scientific or philosophical sense. What they do is illuminate.
Hard SF is a way of thinking about things in a rigourous manner and dramatising those thoughts in fiction. And for once the devil is most definitely not in the details because they only get in the way (and sometimes invalidate a beautiful thought). But even though the details dont matter at all, the thinking behind them, the philosophy that drives them, the approach to the structure of the universe that the story takes, all matter a very great deal.
In the final analysis, who really cares if the Earth revolves the wrong way in Nivens novel? Who cares that Asimov wrote stories in which the planet Mercury never revolves at all? Who cares that Clarke said "There are no mountains on Mars" in his novel The Sands of Mars? What difference does it make that Valentine Michael Smith could not have been brought up by Martians and could never have developed all those psychic powers?
Despite their factual errors and sometimes the complete implausibility of their ideas, under this broader definition, Niven, Asimov, Heinlein, Clarke et al really are writing hard SF. And so are Clement, and Forward, of course.
Only the pedants who continually argue about the minutiae of scientific detail in many SF stories really care about the first, much narrower definition of hard SF and I firmly believe that those pedants are completely missing the larger, more important point.
But then, pedants so often do.