I am sure that most people have heard this definition before, but here goes: science fiction is set in our universe, while fantasy does not have this restriction. This definition is reasonably simple, but it has quite complicated consequences. I will explore some of them by looking at some of the considerations that I think good science fiction must take into account. First I want to look at an indirect approach to understanding future technologies, by reference to two examples.
I introduce the first example by making a bold claim: all the naysayers who said that heavier-than-air flight was impossible were right. Of course, this is obviously not true, but I think their mistake sheds light on a broader topic, because what they should have said is, "Unless you invent a power source whose mass-to-power ratio is a lot better than anything we have now or materials which are much lighter than steel but still strong, heavier-than-air flight is impossible." With the technology available at the time (I'm thinking about most of human history up to the mid-1800's, when coal power and human muscle power were about the only types available to a flying machine), what they said was true: with their materials and power sources, you could not build a heavier-than-air flying machine.
The flaw in their reasoning, was that there was a power source with a far better power-to-mass ratio which would be discovered towards the end of the 1800's: the internal combustion engine (and later, carbon fiber and other materials which were light and strong, but those were discovered later). This technology, while not very efficient and not very clean, still has one of the highest power-to-mass ratios both in terms of fuel and engine weights available, which why it is used so extensively almost a century later. The predictors of the impossibility of heavier-than-air flight had been tripped up by their lack of imagination. From another perspective, available at the same time, one could have said, "Heavier-than-air flight seems hard, but there are no laws of physics which prohibit it. Our known chemical energy sources have the energy density necessary, thermodynamics allows enough efficiency, oxidation reaction rates aren't too slow to liberate enough power quickly, and aerodynamics explains how we can convert power into lift; moreover, since birds are heavier than air and they fly, it's a problem of degrees, not fundamental limits. We just need to figure out an engine and a external structure which combines all these elements appropriately."
Now I want to look at my second example, where I use a similar type of reasoning but based on sounder premises to reach the conclusion that FTL (faster-than-light) travel (or communication, which is more or less the same thing) is extremely difficult. I will start by noting that the laws of physics as we understand them contain the special relativistic light-speed-limit wholly. Neither relativistic quantum mechanics, the basis for the Standard Model, or general relativity ever permit any violations of this speed limit (well, general relativity does permit them, but only under special circumstances which I will discuss later). But, you say, "Aha! Since our laws of physics may not be complete, and in fact, probably are not, it might be that the real laws of physics actually do permit FTL travel."
The way to rebut this argument is to look at how future physical theories are likely to include our current understanding of physics: precisely, as an approximation at sufficiently low energies/large distances/long timescales. Our current laws of physics have been verified extremely well for the situations we can create in our labs and observe clearly in the solar system and sometimes even other stars. I think the Kuhnian emphasis on the conceptual paradigm shifts which underlay the transition to modern physics obscure the importance of the simultaneous continuity: both quantum mechanics and special relativity contain Newtonian mechanics as an approximation, for sufficiently high mass (for quantum mechanics) and low speed (for relativity) phenomena. Likewise, general relativity contains Netwonian gravity as a low-field-intensity approximation. Thus, I argue that because our current theories are supported so well by experiment, then historical experience would lead us to believe that whatever new theories of physics emerge, string theory or M-theory or whatnot, those theories will contain our current theories as an approximation.
What does this imply for FTL travel? It means that to exploit new phenomena which permit FTL travel, you have to go to energies, sizes, and timescales which extend beyond what our theories can predict. What kind of energies? As a back-of-the envelope idea, consider the Planck energy density (which is formed by using dimensional analysis on G, hbar, and c), G^-2*hbar^-1*c^7. G and hbar are small, c is big, the resulting energy density is astronomical: 4.63e113? J/m^3. For comparison, the energy density of ordinary water at STP is "only" 8.99e19? J/m^3. The matter inside the dense region of a neutron star achieves an energy density of "merely" 3.60e31? J/m^3. To get these kinds of energy concentrations, to exploit new laws of physics, requires either titanic energies (to bring a meter cube to this energy concentration, you would need the equivalent mass of 10^66 neutron stars transformed into energy, which is probably more than the number of stars in the observable universe) or enormous energy focusing in both time and space, or more likely, a combination of both massive energies and tight concentrations.
This is what I mean when I say that FTL travel is hard. Remember the FTL solutions from general relativity? They back me up: besides requiring negative energy densities (not antimatter, but actual negative energy, which may or may not exist), these solutions all require energies which are larger than the amount of energy in the available universe. Obviously, my back-of-the-envelope calculations are not a proof of exact numbers, but rather are an indication of the scale. The difference between the energy densities available to the first members of Homo Sapiens and to us is dwarfed by the distance between us and a civilization capable of FTL travel. I won't fall into the trap of saying, "FTL travel is impossible," because our current understanding of physics is not complete enough to allow that statement, but I do believe our current understanding of physics allows us to see that FTL travel will take many iterations of technology to reach, if it is possible.
There several layers to this pair of examples which I wish to explore. The first is the one I have already introduced: how it was the faulty premises when thinking about heavier-than-air flight that led to an incorrect conclusion, not the reasoning process itself. The second is the flavor of a type of argument about future technology. I have, to some extent, chosen a facile example: because FTL travel is so firmly and clearly contradicted by our understanding of physics, in a very precise manner, it is easy to use this type of reasoning to reach a strong conclusion. However, I think this type of reasoning has broader applications which, while producing results with more uncertainty, may still be helpful to give an idea of how hard technologies are. Another example: the rough calculation of the computational capability of the human brain, which indicates that our current computers are still several generations behind that capability (I don't remember the details off the top of my head, unfortunately). Because human intelligence and the functionings of the human brain are so much more poorly understood than the light speed limit, this kind of calculation has a great deal more inherent undertainty, but I still argue that it gives us an idea of what to expect. Moreover, by knowing how hard technologies are, while we will never to be able to pick dates, we might be able to establish an order in which they emerge.
FTL is a favorite stalking horse of mine for exactly this reason. There are countless science fiction novels which introduce FTL travel without introducing the countless other technologies which are vastly easier than it. Because FTL is so hard, I'm not sure that we are even capable of imagining what a society with technology that advanced would be like: the closest analogy I can imagine is one of our ancestors telling a story including airplanes and the Internet, only the gulf would be much bigger. Moreover, because FTL violates the laws of physics in such a fundamental way, understanding what a world which includes it is like may also be a task beyond our understanding: FTL travel permits causality violation, and our understanding of physics is silent about what happens when you try to do such a thing.
Here I wish to unpack one more layer of this pair of examples. FTL travel is a unique technology because it is extremely easy for us to envision, but technologies at a similar level of advancement are hard for us to grasp; frankly, I can't even think of any. Flight is another such technology: while humans have dreamed of flight since the earliest stories which are recorded, the actual execution of it is rather different than they foresaw, and our Paleolithic ancestors would have had tremendous difficulty understanding even the simpler components which go into making our flying machines, and the simpler concepts which underlie their functioning. The gap between us and FTL is almost certainly bigger, so how much validity will our speculations have?
Now to step out again to the question of the definition of science fiction. I think one of the flaws that show up in novels which want to be science fiction, but aren't by my definition, is a confinement of the consequences of technology and culture. This happens from both sides of the hard/soft sci-fi divide: hard sci-fi writers tend to concentrate too much on the technology while neglecting the influences of the technology on the culture and, even more, the culture on the technology. Meanwhile, soft sci-fi writers build a culture and imagine the technologies which make it work, but confine the influences of those technologies other elements of the culture don't change. All settings have a history, and it is just as important to understand that as to understand the time which is the central focus of the story. Otherwise, you might as well just say the world was dreamed up out of whole cloth, because that's exactly what happened. This importance of history is the reason I placed such emphasis on understanding at least rough ideas of how hard technologies are, because it implies that certain technologies will probably be invented before others, and certain technologies are almost certain to be invented before others. For instance, historically, I'm damn near certain that ironworking would be invented before flight in all or nearly all possible worlds. Similarly, I'm expecting practically every technology we have ever envisioned to happen before FTL, and in my opinion, this type of consequence must be dealt with.
Now I want to discuss again some more specific examples. I've been beating on FTL only because it is so common and yet so poorly handled; while I would tend to shy away from including it in any science fiction story unless one really wishes to grapple with that amount of change, it could be introduced without causing so many problems. The simplest and sleaziest way I can think of is to simply not have the technology in control of the point-of-view characters: i.e., aliens came and dropped this technology on the human race, we don't understand how it works, but we're glad to take advantage of it (even that has the problem of some clever physicist trying to use the technology to violate causality to see what happens, but writers can take some poetic license and simply ignore that). I'm sure there are better ways, but my point is that my problem is not with FTL in itself, but the enormous consequences it almost certainly implies yet are so rarely even contemplated.
Another favorite example of mine is psychic powers. I have seen truly offensive numbers of novels classified as science fiction which contain psychic powers: this goes back to that old publishing genre definition, where if you call moving things without touching them "psychokinesis," you're writing science fiction, but if you do it by "magic," you're writing fantasy. A lot of psychic powers have problems with fundamental laws of physics (if something violate the laws of physics as we know them, then that something has the same problems as FTL for the exact same reasons) such as FTL communication or conservation of momentum/energy. I can imagine plausible ways psychic powers could be explained: for instance, if we each had a radio set wired into our brain, it would act a lot like telepathy. But, these too have consequences, and I don't think the way psychic powers are usually introduced (humans just spontaneously evolve them) works unless there are some very creative explanations.
All right, let's get down to the nitty-gritty and start discussing specific novels and authors. I'll borrow the examples cited on the Dune page and try to explain further what the hell I'm talking about. Quoting Ari,
I believe this is exactly what I'm talking about. Dune is placed 20,000 years in the future, yet contains entirely too much that is familiar. Humans have their same familiar biology, they shoot at each other with guns except when they stab each other with knives, communications are carried by electromagnetic waves, food is grown, they use nuclear weapons, and have a social system which would be familiar to anyone on Earth, if only from history books. Oh, and they have FTL travel and shields. I agree that our knowledge of science is incomplete, but I think that a cilivization which can harness the energies necessary for FTL travel should be able to use unharnessed energies to create weapons which make nuclear weapons vastly more obsolete than swords are today.
Now quoting Matt Brubeck,
I have no problem with the soft science fiction attitude of "culture-first". What I disagree with is that he only takes the interaction one step: he considers a culture, then he shapes the technologies around it. But he never considers the feedback effect of technology on culture again. They should have more technology than they do, because they have FTL travel. I do not regard the progress of technology inevitable, so I can believe in something like the Butlerian Jihad. But when you look at technological reversals in our history, they occurred under specific circumstances, and one of those circumstances was a single authoritarian government whose political power would be eroded by technological change. Historically, situations like Dune's, with many Great Houses scheming against one another, fed technological progress because having the upper hand in technology meant greater success in the wars between the different political entities. This is an example of what I mean about Herbert not fully thinking out the interactions between technology and culture.
The Brubeck's criticism of my judging Herbert too harshly is apt. It is a worthwhile point that science fiction needs to be judged, to a certain extent, in the time it was written. But, actually, yes, I am criticizing science fiction that doesn't deal with the consequences of many technologies. Not all technologies need to be a major theme, but they need to be integrated in some way. I don't believe, for instance, that when cyborgs arrive that artificial intelligence will be that far off, and vice versa; and moreover, genetic engineering will be probably be advancing based on the understanding of intelligence and computational power concomitant with artificial intelligence. And so on. Technologies feed off each other and culture, culture feeds back into technology, and the whole system displays behavior that you don't get in isolation. Is this hard? Yes. I still think that good science fiction needs to grapple with these complexities.
This is an example of differing opinion, but I believe that while prognostication is difficult, you can make some reasonable arguments about it. I should also be more careful about my terminology: when I say "artificial intelligence," I am being more inclusive than meaning strictly an AI that can pass the Turing test. Computers will become better at many tasks which are now human-only, such as natural language recognition, and while they may not be able to pass the Turing test, in their own realms they will be as intelligent or more intelligent than any human. As for exactly how hard building a true Turing-test AI will be, I don't know exactly. Because the requisite sciences are so much less mature than physics (as well as being innately less predictive), it's simply much harder to make definitive statements. Once you have the requisite processing power, which it is difficult to calculate precisely, then you have the question of algorithm design and how humans actually go about the many tasks they are capable of. I won't provide a timeline, since I think that guessing time is so hard as to make most estimates totally useless, but I would put it a comparable challenge level to sophisticated genetic engineering, and still vastly easier than FTL travel: it can be achieved using solar energy and ordinary matter, since evolution did exactly that. In fact, advanced genetic engineering and hard AI (Turing-test AI) are surprisingly similar in that they are both essentially analyzing evolutionary solutions to complex problems; while hard, they are not even in the same league as FTL travel.
I think that nanotechnology is something which has to be understood very carefully, and on a case by case basis, simply because it has been used in so many different ways. I've seen uses of nanotechnology which are plausible, and others which simply aren't. For instance, hoping on nanotechnological magic to restore your frozen head to a body centuries from now is ludicrous: whatever information exists in your brain will almost certainly not be preserved because the brain is not chemically stable, and freezing it permanently destroys tissue. On the other hand, I've seen nanotechnology used to achieve tasks which are basically permutations of biology: regenerating limbs (frogs do it, are there in-principle reasons we can't?), grow complex systems, etc. I think there's a lot of bunk there, but mixed with some things which are plausible. Again, it is possible to get an idea of what is and is not possible by noticing that physical principles like conservation of mass must be observed, looking at how much heat generation there is going to be, and the like. What biology has been able to accomplish is an excellent guide; when going beyond it, care must be exercised.
The criticism that I don't have many good examples of science fiction which do fit my definition is also apt. When I first started to give this issue serious thought, I was attempting to understand what I disliked about many works of classic science fiction. I isolated and broke it down into this definition and attendant explanation. Since then, I've only seen a few things which I consider to be good examples of science fiction (not necessarily good, though). And yes, mainstream literature does fall under this definition, but I would exclude it because it is not speculative fiction (there is another half of this essay which has yet not been written). Kim Stanley Robinson's stuff and what I've seen of Gibson (except occasionally in the specifics) are about all that come to mind immediately. Granted, I haven't read a lot of modern sci-fi (I haven't too many classic sci-fi novels, though I have read quite a few short stories many of which I no longer remember, unfortunately). Also, because of historical perspective, I think that there may be some older things which may deserve to be included. For instance, I think you could argue some of H.G. Wells and Jules Verne's works because while their vision of science is quaint at best, their worlds did maintain an essential consistency.
To summarize, the difference between science fiction and fantasy in my mind is a difference of reality. Fantasy deliberately breaks with our reality in subtle and unsubtle ways; pure science fiction, on the other hand, should aim to maintain a palpable sense of reality. Obviously it is possible to combine science fiction elements with other genres, but I think using the publishing genre definition is ultimately a disservice, because there are too many things classified as science fiction which use imaginary technology as plot or setting deus ex machina, without bringing out the full flowering of consequences.
First of all, human-powered flight is possible. The Daedalus 88 has flown 115 km in a single flight- hardly a "brief hover"! Who could have foreseen an aircraft with a 34 m wingspan that weighed less than half as much as the pilot?
The "inconsistencies" in Dune could easily be found in virtually any other science fiction novel. It's easy to say, "Their society wouldn't be like that!" or "How come they have technology X but not technologies Y or Z?" A visitor from the 1950's might ask why we haven't yet landed on Mars, but are so interested in tinkering with this "DNA" stuff. Suppose that, in 1925, someone had published a novel about the atomic bombing of Hiroshima. Scientifically plausible? An explosive that powerful, in 20 years? Don't make me laugh- such a weapon is merely a deus ex machina for the sake of allegory. What about a book describing modern communications technology being written before the discovery of electromagnetism? I suspect that the idea of invisible information waves that can traverse the earth in a few seconds would sound as plausible to people five centuries ago as FTL communication sounds to us.
We cannot predict the science and technology of the future. We can speculate on what is likely to be true, but the possibilities of the future are not limited by what scientists believe at any one time. Furthermore, if fiction were limited to the likely and probable, who would set foot inside a bookstore?
Also, scientific plausibility is highly subjective. Your idea of what constitutes scientific plausibility is based on a far greater knowledge of physics than most people have. Who gets to decide what's believable?
Even if one could somehow decide what fiction could take place in our universe, this definition, as applied above, eliminates most of what are considered science fiction classics. I don't see the worth of such a definition.
Personally, I'm fine with the elves vs. cyborgs definition. Either that, or lump them both into speculative fiction. I don't quite agree with EvilSouthie's claim (cf Dune) that the two genres are actually isomorphic (a fantasy version of, say, Snow Crash would certainly lose much of its zing, and it's hard to imagine a science fiction Neverwhere), but the overlap is significant. --AriNieh
I guess my main question is: what on earth do you consider to be science fiction? Any particularly good novel-length examples? I'm trying hard to think of anything I've read that qualifies under your definition and failing rather miserably. I guess I don't recall too much in The Moon Is A Harsh Mistress that would invalidate it as SF under your definition, but it'd probably be "bad" SF...
--AndrewSchoonmaker (who notes that the SF version of Hunter would be downright scary)
Easy. Mainstream literature (Elmore Leonard, Agatha Christie, Charles Dickens, John Grisham) falls within CurtisVinson's restrictions on sci-fi. With just a few exceptions, speculative fiction (Asimov, Clarke, Niven, Stephenson, Gibson, Brin) does not. I'm curious to see how Curtis excludes "non-speculative" works without resorting to genre-based judgements like the fantasy/scifi distinction he decries above.
Well, one could draw the line based on whether the technology in the story already exists at the time of its writing.
What Curtis is defining does exist, but it is not what anyone else means by the words "science fiction." It is a useful distinction, but why not invent a new term instead of using an one already established under a different meaning? Like it or not, "science fiction" describes a genre of literature. Like other genres it is set apart by its conventions and culture, not by rigorous analysis of the worldviews presented.
I'd also like to respond to Southie's quote from Dune:
This is true for a large subset of fantasy and sci-fi, but I claim it is untrue for some of the best examples of each genre. In some of my favorite science fiction, the scientific concepts are not just devices to further a plot; they are the central point of the entire story. You can place the characters and events of RedMars in a magical setting, but it is no longer a speculative history of the human race exploring the actual planet of Mars, and so (I argue) it is no longer Red Mars in any meaningful way. The same may be said for much of Gregory Benford's work (like Timescape). (Interestingly, these are also among the few SF works I know that fit Curtis's definition.)
Similarly, I don't think that Tolkien's work can be removed from its setting in Middle Earth and still contain any of its core characteristics. The Lord of the Rings is not a story of a quest to destroy a weapon and defeat a warlord; it is the story of the Third Age of Middle Earth. The setting is the story, and the magic makes the setting.
Well, two points. One, you didn't really answer the question I asked, which is "What does Curtis consider to be science fiction?" and not "What does Curtis' definition of science fiction include?" ;-) Two, I guess my question would be better phrased as "Is the intersection of what I consider to be sci-fi and what Curtis considers to be sci-fi nonempty? Or even close to nonempty?"
Yeah, my reply above was really more toward Curtis than toward you. As for genre sci-fi that also falls under the proffered definition, I think that RedMars and sequels fill the bill, and so does Timescape and some other Benford stuff. -- MB
I'll add some discussion which I felt didn't have a good position in the main thread.
MattBrubeck's citations look good. I haven't read Timescape, though I did pick up a copy based on this discussion. I recognize that I am being anal-retentive about this, but I am trying to make a point about philosophy. Knowing MattBrubeck's biases, I'm not surprised he disagrees with me. And I also recognize that MattBrubeck is correct about the genre being defined, in practice, by culture and conventions, not by philosophy. Nevertheless, I prefer in-principle classification schemes based on philosophical distinctions rather than historical happenstance.
As for EvilSouthie's discussion of the isomorphism between fantasy and science fiction, I also agree with MattBrubeck that many of the best of examples of both genres cannot be changed into each other without losing something essential, but I recognize that EvilSouthie is quite correct about many cases in both genres. I think that this point is getting closer to the essence of the understanding I am trying to clarify with this definition: good fantasy is fantasy, good science fiction is science fiction, and things which are in-between are something else entirely. This is a deep point, and I think there's important understanding to be gained from it, but I don't grasp it yet. I'm going to do some thinking on the issue and eventually I'll get around to posting my musings here.
Ari is dead right about human flight. This should be the 10th-millionth reminder to everyone to fact-check carefully, which is much easier now with Google around. My source was out-of-date and probably just plain wrong (I was relying on memory here, rather than on a direct citation as I should have). My apologies.
In response to something Matt pointed out earlier, I'm curious as to where you (Curtis) make the destinction between SF and "normal" fiction. Crichton books like Disclosure and Andromeda Strain come to mind as questionable ( though, IMHO the only Crichton book that is *good* SF is Andromeda Strain, and even that deals with mainly human interaction issues from what little I remember ).
I want to make a comment on your opinions of cyborgs. You claim that cyborgs and artificial intelligence will come nearly hand in hand. I think this is untrue, in that I would claim we already have cyborgs and we do not already have artificial inteeligence... or even nearly have it. I believe that you are falling victim to a stereotypical portrayal of cyborgs in having connection to neural functions and computer functions, but this is not what makes a cyborg.
If you want to go with the definition that ChrisLundberg and TitusWinters use, then we have had cyborgs for hundreds of years. Their argument is that any enhancement to a human counts as making a cybrog, and thus all of you that wear eyeglasses to enhance your vision are cybrogs. Now, you might say that that's not a good definition so I refer to Merriam-Webster for a good definition of cybrog and I get:
Main Entry: cy-borg
Okay, that's not very helpful. So, what does bionic mean?
Main Entry: bi-on-ic
1 : of or relating to bionics
2 : having normal biological capability or performance enhanced by or as if by electronic or electromechanical devices
Now, at the very least the second definition covers many other cyborgs that we have today. I'm referring to individuals with artificial limbs. Granted, not all artificial limbs would qualify (a wodden leg certainly wouldn't) but after a google search for "artificial limbs electronics" I found out that there are static prostheses and dynamic prosthesis based on whether or not they contain electronics. One of the methods the dynamic prostheses use is called myoelectricity whitch senses the electrical signals in your muscles, and interprets those signals to command the artificial limb to move as intended, which is powered by a battery. These clearly fit the second definition of bionic, which in turn shows that we have cyborgs today.
To top it off, I recently saw (like June '02) on the news the next step in cyborgs. They have successfully created a device (and implemented it in like 9 people) that uses a camera, a computer and a hardwire connection to the person's brain to allow blind individuals (that used to be able to see) to see again. Granted, it's not like normal vision, but it's enough that they were able to back up their car and see when an object was in the way, and go around it.
Anyways, this really doesn't fit in very well with the rest of the debate on sci-fi, but I just wanted to comment on the belief that cybernetics and artificial intelligence are likely to come (nearly) together.
Odd that this discussion should pop into my head today and then, when I come back to read FunWiki on my wonderful 36k connection, that NateCappallo's commentary should have been added today. First of all, the original rant needs refactoring. This page could probably also use some refactoring in general, though trying to make this page something other than a discussion would destroy much of its interest, in my opinion (refactoring a page completely out of thread-mode works very well for technical discussions which share more common assumptions and facts, but not so well for philosophical debates). Now, on to my comments.
First, my thought today was that there is another broad class of ficiton which fits into my definition: alternative history. Alternative history includes works which are speculative (in the sense that they happen according to our physical laws, but not in our universe); I believe that Harry Turtledove writes books which are examples of the genre (DISCLAIMER: I have not read anything by Turtledove and don't plan to start soon; I have no idea if his works are good literature or not, or even whether they hew strictly to alternate history without including alternate physics, but I believe that they are in some respects representative of the genre). A specific book which fits into this genre (and my definition, incidentally), is The Difference Engine, a collaborative work between Gibson and Sterling. I would describe the book as turgid and confused, so don't rush out to read it, but it does fit my definition.
Now, to make a broader point: one of the characteristics of the "cultural" genre definition is that sci-fi is placed in the future. Most sci-fi, for better or worse, is about the future (in fact, by my definition, all sci-fi which is not futuristic is alternative history, but whatever). But time flows (at least from our perspective), and so what is once future becomes past. So, all science fiction turns into alternative history. I think that if, in whatever future it meets, a piece of science fiction doesn't become a plausible alternative history, then it fails my definition. It's fantasy, and isn't part of our reality.
As for Nate's points about cyborgs, they're well-taken. If you hold with TitusWinters' and ChrisLundbergh?'s definition, cyborgs have been around a long time. I, frankly, don't, for the practical reason that, if you start including things like glasses, you have to wonder about things like jackhammers and swords and spearthrowers and eventually, flint and obsidian hammers, and then the word "cyborg" doesn't mean very much. I think a more practical definition is that a cyborg implant should require major surgery to remove. Nate is right. Cyborg technology moves quickly. It may well arrive far before Turing Test AI. I could be wrong, and AI could be much, much harder than we had thought. The study of human cognition is still pretty primitive. However, the cyborg technology we're seeing now hasn't gone very far. Maybe, because we don't know much about AI, it's much closer than we think, and will quickly overtake the seemingly fast-advancing cyborg technologies. I personally think there is a connection, because the integration of computer chips with the brain will require the same kind of understanding of human cognition as AI, but I could be wrong. I don't know. I can't see the future. I do, however, think there is a plausible connection.
Oh yeah, I'll give a quick definition of speculative fiction. Science fiction is separated from ordinary fiction by its speculative nature. Simply put, speculative fiction is fiction that is not placed in our past or present; these settings are known, ignoring borderline cases such as the prehistoric past (Jean Auel's series comes to mind as a borderline case). Speculative fiction is placed in settings which are, to a greater or lesser degree, the whole creation of the author. That's the difference between science fiction and Jane Austen.
I do continue to think about this discussion every so often. One of my more recent conclusions is that I may have been unduly harsh in my criticism of FTL. I chose the Planck energy density based on theoretical reasons: the unification theories of physics indicate that those are the energy scales required to unify the three forces we have quantum field theories for with general relativity. Both quantum field theory and general relativity obey the light speed, so a new theory which doesn't always respect it would require a merger of those two. However, our experimental evidence is less restrictive. Our accelerators don't scale up to energies that are nearly so high, and interpreting the limitations imposed by cosmological and astronomical data is harder. Of course, scaling up beyond the energies available in our most powerful accelerators to create macroscopic FTL travel will still involve energies much beyond our control, but they aren't quite so huge as the theoretical predictions. The experimental predictions embody a looser but firmer bound.
I've given some thought to MattBrubeck's suggestion for new terminology, and I haven't been able to come up with anything appropriate. Are there any neologism coiners who can?
I read Timescape some time ago, but never got around to writing a review. Frankly, I didn't that it was that outstanding, but Benford did at least try to deal with some of the consequences of his science. He proposed a mechanism for dealing with communication from the future to the past, and one for the global ecological disaster he posits. I found the latter unconvincing: he proposes a pollutant that encounters a virus which catalyzes the production of said pollutant, resulting in unfortunate consequences. My sense is that the chances of this happening are very small: catalysis in most cases requires some sort of design, which presupposes either intelligence or some sort of selection mechanism ala natural selection. That said, as I said, Benford at least tries; his sense of the science may be different than mine, but his work is a cut above the ruck and run of science fiction in this domain.