Considerations on Western SF&F

It was science fiction, and then science fiction criticism that got me interested in literature, literary criticism and (horror of horrors!) continental philosophy. It is then appropriate that it is time to return the favour by writing up my own contribution to science fiction studies. I therefore offer a deliberately speculative response to the perennial question: what is the distinction between science fiction and fantasy? I hope to work this out regardless of how unfashionable it has become. ¹

The difference between science fiction and fantasy is to be located in the metaphysics presupposed by texts in both genres. This metaphysics does not need to be identified with the personal metaphysical position of the authors, since a confirmed Catholic could write science fiction and a New Atheist could write old-fashioned Tolkienian fantasy. ² Science fiction offers a clearer distinction and these commitments, I think, can be understood by unrolling a more expansive definition of science fiction that takes the "science" in science fiction seriously.

Science fiction constructs a world distinct from the consciousness of the narrator that is comprised of imaginative variations of features selected from the real world (where the real world is understood as being scientifically structured). This is a good one-liner, though there are places it can be expanded and qualified. It need not be the case that this world is known by the narrator, or that the narrator provides the reader with reliable reports of the world. More important than this is the kind of world—not fictional world, but the real world—that is presupposed by this literary activity.

It need not simply be the hard sciences like physics and astronomy, though that is an extraordinarily common occurrence. Nevertheless, in the case of soft science fiction ³ the operative science here can be the social sciences, like linguistics, anthropology, or sociology. Greg Egan's later novels (like The Clockwork Rocket, or Dichronauts) are unusually thoroughly worked-out examples of the former, since these works sketch out alternate universes with their own distinct laws of physics.

The idea that science fiction is not principally about the future but rather distorts elements about the present is not new to me; this is Samuel R. Delany's thesis, and it's especially well set out in his essay "Some Presumptuous Approaches to Science Fiction." ⁴ Delany points out that science fiction "uses the future as a narrative convention to present significant distortions of the present," and that while science fiction is about "the current world—the given world shared by writer and reader," this relation is not metaphor or metonymy. There are two subsequent passages in another essay of his, "Dichtung und Science Fiction" that can be quoted at length:

The point is not that the science in science fiction is “usually inaccurate,” but rather that some of the science in any given SF story must be inaccurate—or it isn’t science fiction. To read an SF story properly one must have some notion of just where the science is distorted. I said that science fiction presents a distortion of the world. One of the things it distorts in the world is present science. The distortion is usually a matter of taking something scientists claim is possible, but which currently is not proven to be the case, and writing about it as if it were the case. This is a distortion of science. At other times it is a matter of taking something that scientists claim is impossible and writing about it with the kind of language that would accrue to the topic if that impossibility had been scientifically disproved—as so many scientific “impossibilities” have been disproved over the years. This is also a distortion of science. But there are other ways science can be distorted in science fiction: writing about a future with less technology available than we have today is also to distort the range of present science. Writing about a future in which technology has remained exactly the same as it is today is also a distortion of science—because the currently reigning scientific belief is that technological recomplications will continue to increase, at least for a while. ⁵

Similarly, in science fiction, especially that science fiction written from the arrival of Campbell at Astounding in 1937 through the Second World War to the establishment of Galaxy by Gold in 1950 (sometimes called classical science fiction, more often Golden Age science fiction), the real science of the time represents a kind of base pattern from which the SF writer deviates to produce an interesting effect—again, sometimes in formal, conventional ways (e.g., the “hyperdrive” or “hyperspace,” which takes Einstein’s theoretical upper limit of velocity [i.e., the speed of light] and violates it; or the assumption of real ESP), sometimes in more subtle or more personally innovative ways. ⁶

Here my attempt if anything is to reconcile my account with what Delany says about "science" and "the present" and therefore worldhood. It is important to note that Delany here is talking about style, not in a high-flown sense but the reading protocols involved in parsing clichéd science fiction. Therefore my discussion of metaphysics might seem as though it transgresses the bounds of Delany's literary criticism. ⁷ I think Delany here is not quite working out the relationship between science and "the present" (and what sort of world that "the present" has to be in order to be distorted). It is for this reason that the essay turns to metaphysics and attempts to work out that relation in detail.

Here is my second assertion: In science fiction, the real world in its totality is posited as being equivalent (either epistemologically or ontologically) to its scientific description and formulation. ⁸ If we consider Heidegger's texts, "The Question Concerning Technology" and cybernetics in "The End of Philosophy and the Task of Thinking," we could even apply Heidegger's analysis of technology to science fiction itself. In the former text, Heidegger says that the essence of technology is not itself technological, and that modern technology is a revealing, but not a bringing-forth in the sense of poiesis. Everything is ordered so that it can stand by as standing-reserve, not as an object. He goes on to say that modern physics itself sets nature up as calculable forces, that man may come to understand himself as standing reserve. ⁹

We can leave aside Heidegger's meditations on the Rhine or the "saving power" of Hölderlin. Regardless, we have here a pretty good descriptive analysis of the activity of modern science, and it is not so difficult to see the connection with technology and science fiction. We can say that modern science fiction posits the world as being identical to or equivalent to the world as constituted by the sciences. These sciences can be natural sciences or social sciences. The world-given-as-scientific is freely distorted and subject to variation, and lends itself to the activity of the construction and fabrication of new worlds. The fictional world itself does not need to form a totality. It just has to refer back to the totality of the real world that is already given and already posited in this particular fashion. The variations made from the real world (as known to the author, at least) are significant only insofar as they are implicitly juxtaposed with the real world as scientifically determined. Some texts go so far as to assume a figure-ground inversion in which the relative positions of the implicit real world and the fictional variations are swapped.

It is not that science fiction is wholly innocent of scientism, of course. It is inarguable that science fiction inherits a particular positivist spirit; I don't mean Carnap, Mach or Schlick but the world-view that owes much to Comte (even without direct inheritance). I do not simply mean Comte's philosophy of science, but his theory of history with the outmoding of religious and metaphysical modes, his hierarchic model of the sciences, a faith in large-scale engineering and the belief in scientific shaping and manipulation of society. ¹⁰ Comte's late "religion of humanity" today reminds us most of those religions in science fiction novels, and those real-life attempts based on them. Closer to the actual practice of science fiction were cybernetics, Dianetics and General Semantics, which are not so alien to a post-Comtean vision. There has been a recurring sympathy for scientism, a Saint-Simonian belief in the rule of engineers, and a rejection of metaphysics in numerous authors, beginning with Heinlein and others.

More interesting than the enthusiasm for Comtean social engineering is the commonplace "scientific attitude" toward everyday life. If I were allowed a Jerry-Seinfeld-tier observation, I'd point out that the prototypical "ideology" of the average writer is a kind of humanism, a positive attitude toward the individual point of view. In authors like Greg Egan and Peter Watts we see instead a distinct and unusual physicalist point of view; even before the modern formulations of this in the work of the Churchlands and Thomas Metzinger, it has been a long supposition in science fiction that consciousness is a feature of complex material systems, and that the brain can be treated like a machine. Science fiction is unusual for having prominent members who accept an ontology which considers consciousness, etc. epiphenomenal, nonexistent or emergent from physical matter, and reducible to material causes, and for demonstrating this in a literal way in fiction. I would wager that similar attitudes are rare or not as prominent (to the extent of playing a significant role in writing) among literary writers.

Soft science fiction, in the sense of science fiction that is inaccurate about science, is science fiction (in my and Delany's sense) insofar as it acknowledges and incorporates an understanding that it diverges from known science. It retains a scientific character in the absence of scientific content as the fictional world might itself be determined by various fictional or imaginary sciences that nevertheless retain the form of scientificity. Nevertheless, my reason for taking the hard-science fiction thread as important here is that there are a long traditions of fantastika, surrealism in many kinds of fiction, but (with few exceptions) only Anglophone Campbellian and post-Campbellian science fiction has explicitly foregrounded, to a high degree, a strong materialism. ¹¹ Literary fiction on the other hand seems to be associated more with humanism or an anti-humanistic ironism. The very extreme of this is Peter Watts' Blindsight, which attacks the very category of consciousness and its products (including literature, I would argue). ¹²

But it might be that focusing on scientism and physicalism is a red herring. A clearer target here is the difference between the "manifest" and "scientific" images of the world, according to Wilfrid Sellars. The manifest image is how the world appears to us in everyday experience, while the scientific image is how the world is presented in our best theories of physical science. ¹³ Science fiction is unusual since it has attempted to write from the point of view of the scientific image itself and to dramatize it for a human subject. There is the paradox that literature refers back to human experience and senses, while sometimes science fiction (even soft science fiction) wishes to present material that is impossible to experience (for instance, periods of millions of years, black holes, distant stars, realistic non-human intelligence). Much of the effect of hard SF comes from the dramatization, for the "manifest image," of an experience that is real in terms of the "scientific image" but alien to the common Lebenswelt. The vertigo caused by this is an experience of the sublime. Common literary approaches assuming a "naive" immediate life-experience therefore have difficulty making sense of these features of science fiction, especially hard science fiction.

This makes it unusual that hard SF itself has not been theorized much. A more substantial example of it is a special 1993 issue of Science Fiction Studies, but many of the ideas presented have not been developed to a high degree. Kathryn Cramer has commented that hard SF “enjoys greater popularity with readers and writers than with critics”; less criticism is written on it than on many other aspects of SF. ¹⁴ Such a reading of hard SF should ideally reinforce the connection with modernist science fiction, and Delany is one of the few who have done so. While it is true that hard science fiction is not privileged qua science fiction, I wonder if the relative neglect of it comes from the lesser training humanists have in science (hard SF texts, however, should be catnip for historians of science, given their cross-training).

I would consider the first wave of "literary" science fiction as being genuinely science fiction. Proponents and detractors of the New Wave have claimed that it was not really science fiction (I refer to the work of Samuel Delany, Thomas Disch, Robert Silverberg, J. G. Ballard, even James Blish, who was not in the New Wave but an early modernist experimenter), because of their attempts to incorporate highbrow references to the Western canon and modernist literary techniques to a given SF topos. They're modernist as opposed to being postmodernist as they were not taking them as tropes to be parodied or used ironically. Irony is more apparent with the contemporary New Weird and with postmodern literature; my impression is that postmodern work is more likely to mine popular culture for expressions and to use them in their fictionality, instead of "playing it straight," which Silverberg and even Ballard are capable of.

Delany's reading of Disch's short story "Angouleme" was to show that it was science fiction (even in the classical sense) in the positing of an objective external world and how it was constructed in a literary fashion to be read through "SF reading protocols." What science fiction does is the displacement of metaphor into literalization; something is made literal when it loses its inherent meaning-material and becomes a determinate object just like any other. A good pair of texts to show the distinction is Silverberg's Dying Inside and Roth's Portnoy's Complaint (or Bellow's Herzog!), since both have near-identical topics but Silverberg literalizes the alienation in the latter, creating an almost-similar but distinct literary experience.

It is now that I can apologize for overweighting science fiction for too long and for leaving fantasy until now. Given my identification of science fiction with scientific-image literalism, what do we make of fantasy? Consider how I linked the language of science fiction to metaphysics: the literal and the determinate-physical. Delany's conception of science fiction is that metaphorical expressions like "her world exploded" or "she turned on her left side" are turned literal. There is no reason that literalism has no place in fantasy: both science fiction and fantasy can include sentences of the form "The red sun hung high in the sky."

I've used the term "literalism" often in an under-theorized fashion, but we can take a look at the second essay in Northrop Frye's Anatomy of Criticism. ¹⁵ Frye comes up with a theory of symbols, forming a hierarchy: we pass from the paired literal and descriptive phases, and then to the formal, mythical and finally the anagogic phase, where the symbol is treated as a monad (his favourite example is how Dante's Beatrice represents the Catholic Church as the Bride of Christ). This is the background for the third essay, on myth, since he believes that all literature has its source in mythology, and so he works out a Great Chain of Being ascending from Water to the Divine.

Frye himself has this to say about science fiction. While this is perceptive (and the relationship between science fiction and myth needs further study, most famously with Kubrick/Clarke's 2001), the problems come from Frye's own Romantic vision of literature. Even if we were to take Dave Bowman as the mythic hero of 2001, film and book portray Bowman first of all as a professional astronaut and embedded in a technological civilization with other professionals. This is why it is just as possible to argue that science fiction (at least in Frye's conception) can often be written in a high-mimetic or even low-mimetic mode (the latter especially in the New Wave).

Even popular literature appears to be slowly shifting its center of gravity from murder stories to science fiction—or at any rate a rapid growth of science fiction is certainly a fact about contemporary popular literature. Science fiction frequently tries to imagine what life would be like on a plane as far above us as we are above savagery; its setting is often of a kind that appears to us as technologically miraculous. It is thus a mode of romance with a strong inherent tendency to myth. ¹⁶

Frye is trying to assimilate science fiction to a kind of technological poesy; and it's not wrong that technological poesy is an aspect of science fiction as written. Frye's method as I understand it is an attempt to generate the imagery associated with each stratum of literature (myth, romance, high mimesis, low mimesis, irony) through the descent of a primal unity through all these levels. This is why the monad is the summit of the symbol in literature. The closest that he discusses a science fiction author (or to be precise, an author of scientific romance) is Wells, and it is interesting what he says: "The destroyer which appears at the end of H. G. Wells’s Tono-Bungay is notable as coming from a low mimetic writer not much given to introducing hieratic symbols." ¹⁷ It is possible for symbols to exist within low-mimetic works, and for these symbols to have genuine mythic resonance, but can we say that Wells subordinates them to a kind of literalness?

This is where (drawing on Frye's own Romanticism) I suggest that fantasy has a distinct relationship with language that is dare I say more in line with Romanticism. Metaphors are sometimes rendered literally in fantasy. But the sense in which they are rendered literally is not simply that the "higher levels" of the symbol are not subordinated (as they are in Wells' Tono-Bungay) but that these levels are themselves made literal. The question here is whether attributes are univocal and literal, or whether we accept the literalness of a mythic structure. The difference is that it is not only that gods and heroes are treated literally, but that the symbolic universe they are embedded in, the mythic resonances, these aspects themselves are given reality. In fantasy, metaphor can be incorporated into an analogical conception of being. Human and cosmic attributes can share the same ground or stand in relation and so the world is poetic, or rather posey in the older sense.

This is where my definition of fantasy begins: Fantasy is a genre that makes literal a Romantic metaphysics. Where fantasy diverges from Romantic literature is that what Romantic literature places on the side of the subject, fantasy places on the side of the object. There is not a specific ontology that comes free with fantasy, and one can number the possible metaphysics: the Great Chain of Being, participatory ontology, microcosm-macrocosm metaphysics, neoplatonism, animism, forms of monism. I am not saying that all fantasy requires an analogical metaphysics, or that all these are the same. One issue is how non-Western metaphysics is involved here: the scientific world-view is unusual. There have been sciences and scientific approaches dating back to the ancient Near East and Mesopotamia, and so obviously I'm not suggesting that it spontaneously appeared with Descartes. ¹⁸

Fantasy in general presupposes a metaphysics that has a hierarchal mode of being, or which includes analogical and archetypical correspondences in reality. In general we can say that fantasy presupposes that meanings or normative statuses (to use Robert Brandom's term) are inherent to reality itself and not a result of social mediation. These particular meanings and values which are inherent to reality need not be anthropocentric or theocentric, though both are common. While my prototype here is Western philosophy and esotericism, I see no reason it cannot also assimilate non-Western metaphysics, for instance a metaphysics like that of Daoism. However given that non-specialists have limited knowledge of them, it might be better to consider this on a case by case basis.

Even the variations on this metaphysics, for instance the "satanic" or "nihilistic" variations, are meaningful only insofar as they negate or vary this metaphysics. What separates fantasy from pre-modern folklore and mythology is that the scientific worldview is known and so the premodern world is not automatically considered the correct one, or at least seen as being unrivalled. This is why fantasy can also have evil gods, nihilistic magical systems (it is for the same reason that there was no modern, coherent, openly self-identified Satanism until Christianity was de-legitimized in Western society and no longer the obvious answer). We can call some of these examples "fantastical Satanism," even though the Devil scarcely appears: for instance by inverting the order of being, by placing blindness and evil on the top, or making analogies and correspondences paradoxical, inconclusive or meaningless. These inversions are however only meaningful given the prior meaning of a given Romantic metaphysics. ¹⁹

Ironic and parodic takes on such a metaphysics belong to fantasy insofar as they presuppose this, unless fantastical tropes are themselves reconstructed in a disenchanted-realist world, with the humour coming from the contrast. Most works that are "mixes of science fiction and fantasy" are usually really fantasy even though they embed signifiers from both. ²⁰ I say this since the scientific signifiers are often used in an ironic mode, and are embedded in a broader world which still runs on analogical correspondence and suchlike. (A good comparison is between Star Trek and Spelljammer, even with the former's cheerful lack of interest in scientific accuracy).

Works of science fantasy where the fantastical elements are themselves treated non-ironically as possessing scientificity (even if of an imaginary science) would then be science fiction. The same goes for "hard fantasy" like certain works of Brandon Sanderson, which can be considered science fiction in this sense. It explains Sanderson's popularity among "rational fiction" and similar online communities. My point, however, is not really to propose a scientific taxonomy that can be falsified with enough strange hybrids. Rather, it is important to look at the metaphysical investments of works themselves, especially hybrids, instead of simply letting them be. This is the reason I'm not proposing elaborate genealogies or invariant characteristics but rather the ontology that is linked to particular language, or protocol of reading.

Works that dramatize the failure or incompleteness of scientific cognition or ironize it from within are second-order as they are meaningful only in relation to such a scientifically-structured reality, and are therefore not necessarily fantastical (unless this failure is explicitly grounded in a strong Romantic metaphysics), and so can be counted within science fiction. ²¹ It is not that some metaphysics is real and some is unreal, since I use the term participatory ontology from Radical Orthodoxy, which includes figures like John Milbank. There are indeed Catholics and Anglo-Catholics and Orthodox Christians who go around today believing in such an enchanted universe.

Even if in reality the universe was analogical (in a Christian or non-Christian way) I would consider a work drawing on those categories to be fantasy. ²² David Bentley Hart is an Orthodox theologian who pronounced Vedantic and Neoplatonic affinities, and has written a criminally underrated fantasy novel, Kenogaia: A Gnostic Tale, inspired by the Gnostic Hymn of the Pearl. Hart obviously does not believe that the events of the novel happened, but believes in such a metaphysics. His novel is fantasy even if his particular ontology were to be correct. The novel is readable as fantasy and can be enjoyed as such by someone unfamiliar with Orthodoxy or the Hymn of the Pearl. The idiom of the novel presupposes such a metaphysics that is participatory and attributes such a meaningful structure to reality. This definition is agnostic to whether the universe really is at bottom disenchanted or not.

It is here that I have to finish off what I have sta rted. It may be that non-Western SFF does not work according to this schema and that requires more investigation. Another future direction of investigation would be looking at the social condition of "scientificity" and the social meaning of Romanticism more broadly in a "disenchanted" world today.