Something monumental has occured: I have started to organise my Goodreads books by genre.
I tried doing this once before, when I joined GR… but I found the ad hoc categories I’d picked deeply inadequate, and rather than slowly reforming them I just scrapped them all in a fit of pique.
So now I’ve created a different set of ad hoc categories without adequate forethought, and I’ve no doubt it’ll all be different this time.
Cynicism aside, I have learnt a little from doing this once before: namely, that normal conceptions of genre (or at least subgenre) are frankly inadequate for the purpose.
Now, I should defend that claim a little. I am not somebody who obsesses over genre classifications. In a genre demarcation debate, I’m the one saying “look, does it really matter?” and “hey, linguistic demarcations are always vague and fluid around the edges” and so on. I’m totally happy just lumping everything in The Genre into the category “genre” and leaving it there.
The problem is, almost everything I read is ‘genre’. And some of it is quite different from some other bits. And I’d like to be able to express those differences. So I started trying to divide things up. And then once you start, or at least once I start, it doesn’t feel right to, say, create one shelf for “TSR stuff and stuff that’s like TSR stuff you know what I mean” and one for “classic science fiction” and then another for “absolutely everything else”.
And it’s not illegitimate, I don’t think, to think about questions of genre. Sure, we need to bear in mind that these questions are not very important questions – but so long as we keep a sense of perspective, there are almost no truly bad questions in the world. The mere fact of questioning is itself a good thing.
It’s also worth saying upfront that genre classifications depend on the purpose of classification. This is mostly done for the purposes of marketing new books, which is why the labels are so broad and vague – you want to give a little bit of an idea, but you don’t want to turn people off by seeming too niche either. And also, modern marketers don’t have to worry about changes in genre over time. “Epic fantasy” is easy to say, but there is a hell of a lot of difference between, say, A Song of Ice and Fire, The Wheel of Time, and The Lord of the Rings. [In my scheme below, these are three different genres].
On the other hand, a true historian of the genre would have other motivations – time period, who influenced whom, what was thought of at the time as related to what, and so forth. So, for example, I’m not including a ‘New Wave’ category, because to me this is more a historical issue than an issue of content per se. And the difference that I feel are relevant to how I experience a book aren’t necessarily the ones that matter to other people.
Penultimately, a conceptual note: ‘genre’ can describe characteristics across many possible dimensions of a work – style, length, setting, themes, and so forth. Theoretically all these dimensions can be combined in any permutation. In practice, I think books tend to cluster into vague but discernable genres combining multiple features – or at least my experiences of them tend to so cluster.
And finally: because this is just my Goodreads shelving system, it does not include many potential genres that I just don’t have representatives of on my shelves (or at least if I do I haven’t sorted them yet). Similarly, ‘genres’ that I only have one book of are likely to have been folded into another, similar genre, whereas genres I have lots of books from are likely to have been broken down more finely. This is just how I am currently shelving the things I have currently shelved; it does not pretend to be a definitive classification (though I hope it may be interesting to others along the way). Appropriate classification depends on what is being classified…
That all said, here’s how I’m currently dividing things up:
Top-level category: Fantastika
A term to distinguish from mundane fiction. It’s intentionally broad and theoretical, and somewhat arbitrary. It would be more normal these days to call it ‘speculative fiction’, but that’s a moutful and feels artificial, and perhaps over-specifies (not all of this is really ‘speculation’). ‘Fantastika’ seems like a good general word for it: stuff that has stuff in it that clearly isn’t realistic. [It could be seen as part of a three-fold distinction in narratives: narrative non-fiction (stuff that really happened), mundane fiction (stuff that didn’t really happen but believably might have happened or might yet happen or could be happening) and fantastika or fantastic fiction (stuff that really obviously didn’t happen, at least not in a literal sense).
I’m using ‘Fantastika’ here to include only narrative works. I’ve separated out, at least for the moment, ‘poetry’ and ‘secondary fiction’, although I might re-integrate them later. By ‘secondary fiction’ I mean material that is broadly fictional, but presented as a factual depiction of the fictional facts of a narrative work (and which might pass as a non-fiction compendium of facts about that work in some cases). So this includes bestiaries, atlases, ‘world of’ companion books and so on.
Within fantastika, I’ve conceptualised four second-level categories (not reflected on my GR shelf names for reasons of space and clarity): protofantastika, perifantastika, SF, fantasy, and liminal fantastika. What do I mean?
It only really, to me, makes sense to speak of ‘fantastika’ (‘spec fic’, ‘the fantastic’, etc) against a background assumption of realism. Therefore, anything that would be fantastika if published today, or that has a lot in common with fantastika, but that predates the establishment of a clear mundane-dominant/fantastika-deviant dichotomy, I’ve put in ‘protofantastika’. This probably means before about 1800, though one might quibble over the dates. So far, this has meant things like Swift’s Gulliver’s Travels, and Shakespeare’s The Tempest.
Yes, this is a stupid term, but it’s the best I’ve got. This comprises three categories of works that are clearly non-mundane, but that, like protofantastika, seem to sort of try to edge around the mundane/fantastika distinction, in this case by claiming to not be the sort of work that that distinction applies to. Children’s stories are stories at least in part aimed at children, which happily ignore questions of reality and possibility for the sake of the story. Fables and fairy tales share many tropes of children’s stories, and may in many cases be told to children, but also try to remain relevant for adults, and have distinct styles and themes of their own. They borrow the guise of (or genuinely are representatives of) specific pre-fantastika genres in order to claim respectability. [This category ‘ought’ to be split in two, and children’s stories could be split into many, but since I don’t have many on my shelves right now it doesn’t seem worth it]
I’m also conceptually including dystopia here. Dystopia refers to a vision of a possible world that is not very pleasant, with parallels drawn to the real world to serve as a warning. These are unusually on the surface science fiction, but they don’t have to be – they don’t have to involve anything impossible happening (so they might be mundane) and sometimes they might be presented through a realistic frame story in any case; in addition, the emphasis is often more on the message than on the narrative, and so they tend to highlight the fictionality of the world. So I think they fit best as perifantastika. Other potential genres that could be put in this category are utopia (opposite of dystopia), and tall tale, in which a seemingly fantastic story is given an intentionally questionable frame, leaving the reader unsure of whether the events of the story are presented as real or not – in other words it is a realistic story about a non-realist story being told. This genre was commonplace up until about the middle of the 20th century, but has since largely vanished (I’m sure isolated examples remain, particularly in short story format).
Since I’m not primarily an SF reader, I’ve only divided this into four, and one of them is a fake category anyway. General SF is just where I’ve put some short story collections that may cross subgenre boundaries, plus the novel Cloud Atlas (essentially a collection of novellas mimicking the tropes of different genres). The three major categories are therefore science fiction proper, social science fiction and scientific romance.
Social science fiction, the bulk of my shelved SF, examines the social and anthropological ramifications of a particular proposed technological environment. The emphasis is on exploring the world and the characters within it. So from my shelves, Riddley Walker, Dreamsnake and A Canticle for Leibowitz all look at how people act in different kinds of post-apocalyptic scenario, while Cat’s Cradle looks at how people react to an imminent technological apocalypse. The Caves of Steel, on the other hand, is about a future hyper-populated Earth, and about attitudes toward technology, exploration, and in particular toward artificial intelligences. China Mountain Zhang meanwhile posits a world not so different from our own, but with significant changes in patterns of economics, politics, technology, entertainment, and so forth. The exact technology producing the environment in question is often left vague, and may well be flagrantly impossible.
Science fiction proper, meanwhile, is less interested in the ramifications of technology than in the technology itself. The lines can of course be hard to draw between these – the difference is in emphasis. Science fiction proper explores the possibilities and limitations of a technology, largely leaving the social context for the imagination. On my shelves, so far this category just means some of Asimov’s stories. But that can provide a good example. The stories in I, Robot examine the concept of robotics – what robots can do, what they should do, what they could end up doing. Many of the stories are essentially set up as puzzles – given the facts of the matter and the stated nature of the technology, what is really happening? Asimov’s The Caves of Steel, on the other hand, leaves the questions of what R. Daneel can or cannot do (and why) largely to one side, and instead examines how Elijah Bailey (and the rest of humanity) interacts with him, and so is more accurately social science fiction. A further common (but not universal) tell-tale sign is novelty: in science fiction proper, we are often given some new technology, and see only how people initially react to it, whereas in social science fiction the technological environment is usually more fixed, and we see the settled forms of life that have developed around it. So, for example, a story where a robot is invented that is so realistic that a woman falls in love with it is a science fiction proper story (and is an Asimov story iirc), whereas a story about a society where people frequently fall in love with lifelike robots, and that investigates the consequences of this (loss of human interaction, possible love-laws to prevent such pairings, people growing up with robotic parents etc) would be a social science fiction story.
Finally, scientific romance shares with science fiction proper the relative lack of interest in social consequences, but also shares with much social science fiction the relative lack of interest in plausibility. Typically, this term is used for a narrow range of pre-WWII pulp stories, but I think it should/can apply much more widely than that. The core of scientific romance is a story about individuals – usually an adventure or other heroic escapade – that is draped in the clothing of science and technology. But it’s more than just a story about individuals – it has grandeur and portent, and claims to be about something much bigger. It often hints at wide vistas and vast time-scales and exploration and the fate of mankind. I’ve shelved three examples so far: Bester’s The Stars My Destination happens to have spaceships and teleportation, but is otherwise a retelling of The Count of Monte Cristo, with the addition of an overt narrative of pivotal moments and the destiny of all mankind; Asimov’s later novels, Foundation’s Edge and Foundation and Earth involve intrepid explorers, decisions about the fate of mankind, and the ruins of worlds tens of thousands of years old. None of them are particularly concerned with either the technology involved (Bester’s teleportation technology is just that somebody discovers you can teleport by thinking about it really hard) or the social consequences of that technology (Bester throws in some teleportation-related window-dressing, but mostly gives a fairly unexplored neofeudalism).
Let’s begin with the big stuff. There’s a general marketing genre of ‘epic fantasy’, but the more I think about it the less it really seems to make sense to me. Because this label is used for what seem to me to be at least three completely different sorts of story.
‘Neoromantic’ works have a number of distinguishing features. They tend to involve protagonists who are ordinary people (sometimes even people drawn from ‘the real world’ for added ordinariness), and they tend to confront those people with vast moral imperatives, often but not always centering on the end of the world or the conquest of the world by an evil power. To answer these imperatives, the protagonists must embark on a heroic quest, in the course of which they must confront their own weaknesses, mental and physical… but most importantly of all, moral. The dimensions of the moral conflict are central to the work, and as a result there are often worlds painted with bright whites and dark shadowy blacks and bursts of vivid green and red – world with high contrast. The conflicts of the modern world are transformed and writ large – the joys more joyful, the sorrows more sorrowful. Although overt inconsistency is avoided, there is little concern for ‘realism’ per se, except in the psychology of the protagonists and some close associates. Indeed, the gap between the real world and the fantastic world is often intentionally heightened by the use of archaic or unusual language, and the trappings of a familiar yet alien society – often mediaeval in nature, but there is little interest in creating a ‘realistic’ or ‘historically accurate’ mediaevalism. Rather, historical trappings are borrowed for their symbolic effect. In neoromantic fantasy, a great deal is symbolic. Many of these stories are also overtly religious: the external quest is only a symbol of the internal quest for the protagonist’s soul. This, indeed, is more important than the overt quest structure, which may be done away with if there is sufficient alternative moral challenge and growth.
In this category, I have included The Lord of the Rings (although the symbolism is more subtle than in some works), Donaldson’s Chronicles of Thomas Covenant, and White’s more pessimistic and non-theist The Once and Future King (although most of the series lacks an overt ‘quest’ structure). I’ve also included The Silmarillion here, although its tone is quite different, and after some debate I’ve also placed Richard Adams’ Shardik here.
What I consider ‘true’ epic fantasy evolved out of neoromantic fantasy by retaining its scale, concerns, and in most cases tropes, while exchanging symbolism for realism and toning down the language into something more approachable. Because of the loss of the dominant moral dimension, the overt quest structure becomes more central – but at the same time, because a symbolic moral victory is no longer convincing, the quest may often develop into or be in part replaced by full-fledged war. Because the symbolism is less important, there is less interest in moral nuance: epic fantasies resolve more easily into good-vs-evil conflicts (neoromantic fantasy is often constructed along these lines in theory, but concerns over human iniquity and faith in the fundamental unreliability of evil often make the conflict more multisided in practice), which typically end with strong clear victories for the good guys (neoromantic fantasy is often pervaded by loss even in victory, as the world remains sinful and impure). In format, epic fantasy is even more likely than neoromantic fantasy to expand into long series and to feature sprawling casts.
Two classic examples of (early and late) epic fantasy are the original Dragonlance Chronicles trilogy and the monumental Wheel of Time series, as well as Feist’s original Riftwar trilogy. I’ve also included Eddings’ Elenium and Tamuli here; I could have included his earlier Belgariad and Mallorean as well, except that they seem to be even better described by another category, as explained later below.
In ye olden days, a ‘saga’ was a long, very long story often dealing with the affairs of kings and warring families. More recently, stories are often described as ‘sagas’ when of great length and complexity, again often centering around families (most famously the ‘Forsyte Saga’ and related novels).
So I’m using the term fantasy saga to describe a genre that grew out of epic fantasy. Fantasy saga keeps the scale of epic fantasy or even expands it further, but pushes the existential war against evil out to the periphery of the story, or even dispenses with it altogether. Instead, much of the actual content of the story revolves around conflicts between families (or equivalent institutions), or within families, and the political environment around them. The trend toward realism continues, and there is emphasis on psychological development of the characters; in general, ‘good’ and ‘evil’ are replaced by competing factions of flawed individuals. There are often multiple generations involved, and long periods of time can be covered.
The most famous example of this genre is Martin’s A Song of Ice and Fire; another great writer in this vein is Robin Hobb (in particular, her Liveship Traders trilogy feels like a 19th-century family saga transposed into a fantasy world), although her original Farseer Trilogy may actually be closer to traditional epic fantasy when taken in isolation. Feist and Wurts’ Empire trilogy is an earlier example of the form: epic events do take place, but the emphasis is very much on their significance for political and military manoeuvreings of the Acoma family and its enemies (indeed, many of the epic events are actually ‘borrowed’ from another series altogether, and happen off-page).
Sword and Sorcery
Sometimes, however, fantasy likes to deal with small things rather than large. Not everything can be a grand decades-long struggle. The antithesis of this approach is the traditional genre of sword and sorcery. Works in this genre – which tend to be brief novels or even short stories – see events from a limited, personal perspective. In many ways, however, they are at least superficially similar to neoromantic fantasies: they often feature unusual language use, for instance, and tend to deal in extremes rather than nuances. However, the moral dimension is reduced – sometimes entirely, leaving the story of a protagonist just trying to survive in a brutal world, but other times only partially, leaving us with clear moral dimensions but on a very small and personal scale. Not “how will we be able to prevent the conquest of all mankind?” but “how will we rescue those scantily-clad women before the dragon eats them?”
Worldbuilding in this genre is piecemeal; some authors may try to build continuity over time, but it’s a secondary concern to the story at hand. Conflicts tend to be external rather than internal – the characters tend to have already passed or failed the tests that neoromantic fantasy sets. The stereotypical conflict of the genre is between the heroic, sword-wielding warrior (ideally barbarian and unsophisticated) and the dastardly, book-reading sorceror or wizard and his evil henchmen; although of course a great deal of variety exists. In form, the genre tended toward serialised chains of small adventures – there are 32 ‘Fafhrd and the Grey Mouser’ stories plus a novel, for instance. Often, however, even internal chronology was not maintained – Fafhrd stories and Elric stories, for example, were both originally published in a different one from their logical chronological order. The characters were established, but details of their lives did not have to be witnessed in order – in part because protagonists of these stories tended to be perpetually on the move.
The genre sometimes seemed to desire to be taken seriously as literature, but was always close to the edge of ridiculousness, something many of its most succesful works were fully aware of – Leiber’s stories of Fafhrd, for example, were an intentional deflation of the seriousness of the stories of Conan.
I don’t have any ‘core’ S&S on my shelves yet, unfortunately, so the representatives are all debateable. I’ve considered Gene Wolfe’s Book of the New Sun and Urth of the New Sun as S&S, albeit very late, literary and abnormal S&S. The dense religious symbolism and puzzle-like structure are not representative of the genre… but the ornate language, strange far-future world and episodic picaresque structure all seem to hail from this genre. Meanwhile I’ve also included two of Terry Pratchett’s early Discworld novels, Guards! Guards! and Sourcery. Both of these present parodies of the genre (as do many of his other early works), but are also in my opinion close enough to the genre to belong within them.
Neopulp is not the direct descendent of sword and sorcery – there were a few decades in-between – but it is in some ways a spiritual descendent (which is why I’ve given it this name). Like the sword and sorcery stories of the pulp magazines, neopulp is concerned primarily with small-scale conflicts. However, rather than existing in an antagonistic relationship with epic fantasy (as blood-soaked sword and sorcery often was to idealistic neoromantic fantasy), neopulp enjoyed a symbiosis with its better-selling cousin. The best example of this may be the Dragonlance setting, anchored by the epic (though small by modern standards) Dragonlance Chronicles and a number of sequel series. These series inhabited a world, and readers of the main series would grow curious about these worlds; thus, books grew up to meet that curiosity. For instance, at the beginning of the Chronicles, a group of friends meet up after five years apart; accordingly, we get the six-volume Meetings series detailing how the different characters first met, and the six(?)-volume Preludes series detailing what happened in those five years. The Villains series gives stories of some of the antagonists; Warriors gives stories of some one-line Boba Fett characters the heroes meet along the way. The heroes encounter two different Elven civilisations, so naturally Elven Histories gives us a trilogy exploring the life of Qualinesti founder-hero Kith Kanaan, his reasons for leading his people away from their Silvanesti cousins, and his time among the Kagonesti wild elves. The Kagonesti are also seen in one volume of the Lost Histories series, giving stories from the perspective of a bunch of different species that don’t really feature much if at all in the main books. Some random woman has adventures in an undersea city – nothing to do with Chronicles, really. Some civilisation collapses tens of thousands of years before Chronicles. Etc etc. Books before the main series, books after the main series, books during the main series. Lots of books. A hell of a lot of books. There are about 300 books in the Dragonlance world if I recall correctly. And another 300-odd in the related Forgotten Realms world. Neopulp had in common with S&S not only the small scale of the stories, but also the large scale of the catalogues: as with the endlessly serialised, spun-off and imitated stories of the twenties and thirties, much of the appeal of neopulp was from the reader’s ability to always get more of the same. Originality was not highly prized, and in particular authorial identity was not prized. Not only did most succesful neopulp writers write within ‘shared world’ settings, but on some occasions (as with the pulp writers of an earlier generation) they did not even publish under their own names. FR’s Avatar trilogy, for example, was written by Richard Awlinson – except that in reality Awlinson was a composite of two different authors (Scott Ciencin, who wrote the first two books, and Troy Denning, who wrote the third) and an active editor (James Lowder, who himself wrote a sequel some years later).
Quality was also not of the essence, and varied wildly between installments; continuity between books was often strained, and in particular the prose style tried to be approachable, and usually unpretentious (though sometimes a trace of Victorian prolixity remained). The emphasis was on story. Neopulp differed from sword and sorcery in being even less ambitious (none of Moorcock’s Byronic echoes here!), and in having a much more developed sense of place. These were not (always) random ogre-threatened villages in the wilds of nowhere, they were established settlements with people and (broadly-drawn and often anachronistically American) culture, that were part of a broader world, and consequently the stories often connected implicitly or explicitly to broader narratives. Nonetheless, because the draw was the world, there was a surprising amount of experimentation possible with the plot – while many were of course formulaic, others broke away in new directions. To mention the two Dragonlance novels I’ve re-read since I started this blog: Dark Heart is the story of a frustrated teenage girl struggling with gender roles and parental disability in a one-parent family in a conservative small town (plus there’s a werejaguar), while Lord Toede is an autoparodic philosophical novel about an evil man trying to do good (due to a bet between devils) but unable to evade his own nature, with a significant dose of Groundhog Day into the mix (and a sentient siege engine, ogre erotica, a frog-dragon and a necromancer called Bob). They aren’t necessarily good novels, but they’re interesting! And then there’s the entire Planescape setting…
The giants of neopulp were the D&D worlds of TSR (later Wizards of the Coast), and the novels attached to the Star Wars films, but there were many less famous settings, and of course there were also individual authors trying to make a living by themselves. Among my Goodreads books I’ve included FASA’s Shadowrun novels, as well as Weiss and Perrin’s Deathgate Cycle (could be termed epic fantasy, but since most books are essentially standalone adventures that are only united at the end, I’m sticking with neopulp for now).
As my examples show, I’m using neopulp for both fantasy and SF; after all, the fun-is-all-that-matters approach of this subgenre rather undermines the idea of strict genre demarcation. When TSR decided they wanted stories with spaceships in their dragon-oriented fantasy, demarcation issues didn’t stop them (mostly to be found in Spelljammer, but also a couple of Dragonlance short stories, iirc). Similarly, the entire premise of FASA’s Shadowrun setting is ‘cyberpunk but with magic and dragons and elves and shit’.
Similar to neopulp in some respects is the genre of social fantasy. This parallels the SF genre of social science fiction, in that it explores social relations within a fantastic setting. Like neopulp it often tells small stories, of individuals caught up in or simply tangential to important events; but unlike neopulp it places its emphasis on the lives of those characters and the nature of the social environment they inhabit, rather than on the ‘adventures’ they undergo.
On my shelves, this is mostly represented by Terry Pratchett, who gives us the lives of ordinary people in strange situations. Maskerade, for example, is a story of a young woman dealing with self-esteem issues when she moves from a small town to the big city and becomes an opera singer. The magic only exists to heighten the scenario. Alongside Pratchett I’ve placed (most of) the Pern novels, which are concerned with depicting a particular fantasy world, and the impact of various innovations and events (these may traditionally be called ‘science fiction’, due to the SF explanations given in some later books, but the distinctive features of the society being explored are not (at least at first) technological). I’ve also put here Volsky’s Illusion, a fantasy retelling of the French Revolution, and two novels by Margo Lanagan, which each give very ‘realistic’ portrayals of societies living with a hint of fairy tale about the edges.
The bildungsroman category is for novels that are primarily stories about young people growing up, but who just happen to be doing it in a world with magic or the supernatural in it. This doesn’t include all stories with young people – the emphasis has to be on the young person’s coming of age. The Wheel of Time, for instance, I’ve classed as epic fantasy, because the story is about Mat, Rand and Perrin saving the world. Seraphina, on the other hand, I’ve put down as bildungsroman, as the story seems to me to be about Seraphina’s self-esteem, social status and love life, with that saving-the-world stuff firmly in the background. The genre is also distinct from social fantasy, in that social fantasy, as well as tending to feature more adult characters, has a broader focus within the world it’s describing. So for example I included Pratchett’s Maskerade as a social fantasy but I put Equal Rites in bildungsroman: because Equal Rites is all about Esk growing up, leaving her home, going to the big city, liking a boy, etc, while the world around exists to support that story, whereas Maskerade may primarily focus on Agnes, but uses her plot only as one part of its broader commentary on society and human nature (most obviously, the POV of Granny and Nanny is much more central to the story than in Equal Rites). The bildungsroman, in focusing on the young character, will also probably feature arcs of growth and maturation and instructive lesson-learning. The bildungsroman is often oriented toward young readers, though this is not essential.
This is also where I’ve put the first two David Eddings series (though I suppose you could argue only the first really belongs here?), the Pernese Harper’s Hall trilogy, and Pratchett’s Mort, as well as Pratchett’s novels about Susan (in which Susan grows up and learns Important Lessons).
Neoromantic Fantasy has its moral or symbolic or psychological purpose; Epic Fantasy has its clash of good and evil; Fantasy Saga has its houses and its kingdoms. Neopulp has its fleshed-out (if not always well-edited) settings to give it substance; social fantasy looks at its setting in even more detail. Bildungsroman doesn’t dare become too weird, lest we lose empathy for its protagonist. Sword and Sorcery has a little more freedom, but typically limits itself to a relatively understandable, constant world.
Unbounded Fantasy is what happens when the author intentionally does not limit themselves to a single, well-defined style or setting. Sometimes this involves an intentionally inconsistent world; other times, it just means a world that is much more complicated and genre-bending than first seems to be the case. The key characteristic of the genre is that the reader is constantly being confronted with situations where, so far as they can see, anything could happen. Twists occur that would have seemed entirely impossible only a short time before.
In my shelves so far, this genre comprises several of Pratchett’s Rincewind novels (The Colour of Magic is a prime example of the style), as well as Adams’ The Hitchhiker’s Guide to the Galaxy, and the books of the Abrams webcomic Sluggy Freelance (which makes Pratchett and Adams seem dour and conventional by comparison). As these show, unbounded fantasy is often used for comic effect and ‘non-serious’ books. However, I also include Vonnegut’s more literary, tragicomic The Sirens of Titan, as well as Marshall Smith’s Only Forward – both claim to be science fiction, but in truth they’re wild tours of the imagination marked out by their unpredictability. Finally, there’s Gentle’s Ash, which plays with the distinctions between alternative history, fantasy and science fiction and is, again, constantly subverting expectations.
And then there’s general fantasy, a catch-all for a few cross-genre short story collections.
It may be noted that I haven’t created a shelf for ‘grimdark’ – I don’t have to, because I don’t have much yet that could be given that name, so I’ve not had to decide yet whether it needs its own shelf, or whether it might fall under fantasy saga, neopulp, social fantasy or sword and sorcery. At present the most grimdarky thing I’ve shelved is Hurley’s God’s War, which I’ve put down as neopulp. It may be different in its origin, and may be better than a lot of neopulp, but it seems to me that its tone is very similar to a lot of neopulp, and particularly to the Shadowrun novels.
With Fantasy and SF out of the way, what’s left? Well, a bunch of books that sit on the edge of fantastika. These are not books that try to weasal out of being fantastika on a technicality, the way fairy tales or dystopias do, but instead they simply limit the amount of non-mundane content they possess, or blur the lines of what is and is not mundane. I’ve separated out a number of genres here: parallel fantasy, surreal, magic realism, and liminalia proper.
To put it bluntly: if ‘fantasy’ is set in another world, and ‘science fiction’ is set in our world but in the future, ‘parallel fantasy’ is fiction set somewhere that is both our present (or past) and another world. It’s our world, but with something big changed, so that it’s fundamentally different… but apart from those central changes, everything works more or less as it does here.
Parallel Fantasy is a big genre, which conceptually includes three other, each very important, subgenres: alternative history (our past or present but with history having taken a different course), secret history (our actual past or present, or something very close to it, but with dimensions (often involving fantasy creatures) unknown to the general public) and superhero fiction (our world, but some people have supernatural powers and typically fight crime in some way). These are three very large marketing categories, but conceptually they are very close together – how secret does the academy of wizards have to be, how rare must vampires be, to make it secret history rather than (fantastical) alternative history? In any case, these aren’t areas I’ve read that extensively in, so I’m just bundling them all together in a single genre.
The surreal is rooted in reality, but blurs the lines between reality and dream: dreamlike things start to happen in ‘reality’, or people have very ‘realistic’ dreams. The word derives from the surrealist movement, but I don’t intend to limit the term that finely – I use it for any dreamy fiction, whether after or indeed before the surrealists proper. In particular, one key example I take is Chesterton’s The Man Who Was Thursday: A Nightmare (note the proto-surrealist subtitle!). Kafka mostly falls under this category too.
This doesn’t apply to any old story where there is a dream or where dreams become reality: what matters is not whether the author says that something is a dream, but whether it follows dream-logic. This is typically associated with unsettling and unpredictable content, where nothing, even the ontological foundation of the story, is sure to remain in place (although this need not be bizarre or absurd – the dream elements may in fact be quite small), and where strange occurances require no explanation. The first line of Kafka’s famous story, for instance: “Gregor Samsa awoke one morning to discover himself transformed in his bed into a gigantic insect.” It’s the sort of thing that happens in dreams, but not in reality, and Kafka makes no effort to explain why this happens. A SF writer might explain that his neighbour is experimenting with a mutation-ray, a fantasy writer might suggest that gods or malicious elves are at work, a parallel fantasy writer might not know why exactly it happened but might posit an entire world where this sort of thing happens on a regular basis to people and poor Gregor has to find some guide to show around his new insectoid-subculture. Kafka just tells you that it happens, and that’s an end to it.
Typically, however, the things that happen do (as in dreams) have some potential symbolic significance, even if they do not make any sense taken literally.
I’ve included here The Man Who Was Thursday, and also (more controversially perhaps) The Master and Margarita. Delany’s Dhalgren is here, as is, incongruously, Sophie’s World – a book about the history of philosophy on the surface, but it later introduces elements of Berkelian Idealism and undermines the ‘reality’ of its own setting. I’ve also put Borges’ Fictions here, although I’ve also got it in another category (I couldn’t be bother to make a ‘general liminal fantastika’ shelf just for this book).
A genre closely allied to surreal, and some forms of parallel fantasy, in depicting impossible events occuring in what appears to be an ordinary, normal reality. The distinction I’m drawing lies in the nature of the underlying blurring of boundaries: in surreal, the line is blurred between reality and dream, and between possible and impossible. In magic realism, by contrast, the line is blurred between truth and metaphor. Events are described in such a way as to leave ambiguous in many cases whether the events take place literally as described, or whether the description is fanciful or symbolic. For example, a woman in One Hundred Years of Solitude is described as being bodily raised into heaven: is this literal, or is it a flowery way of saying that she was ‘too good for this world’ and died? The reader is not invited to decide on an answer. Likewise, when a man is shot in a closed room, his blood forms a trail leading back to his mother: is this a ‘real’ event, or is it a way of saying that the blame for his murder/suicide was seen by all to be on his family’s hands? Or just a way of emphasising the distance he has come from his birth, with his blood retracing his journey from his mother’s womb? And if some of the magical events may be read metaphors for possible things, some of the possible things may be read as impossibilities: when everybody in town ‘forgets’ that a certain thing has happened, is it that they are pretending to forget, for political reasons, or is there really a supernatural amnesia at play here? The book does not insist on drawing a line between these two interpretations: perhaps it is one, perhaps both – the question is not really important. The metaphorical, the theory goes, is just as much part of what is really experienced as the literal.
This difference in underlying ambiguity also leads to a difference in style: where surreal fiction uses its impossibilities (and improbabilities) to alienate, shock, and destabilise, magic realism aims to makes its impossibilities seem safe, believable, in-keeping.
As my name suggests, I’m using this for fiction that really straddles the boundaries of mundane and fantastic. If surreal introduces disturbing elements of dreams, and magic realism introduces metaphorical realities, and parallel fantasy sharply demarcates possible and impossible but simply moves a few things from one category to the other, liminalia does its best to avoid setting its own limits at all. In this respect, it’s close to unbounded fantasy, but where that genre defines itself through its refusal to set boundaries, liminalia is not even willing to define which boundaries it does not have.
These stories may be set in ill-defined or non-specific places, or otherwise avoid explaining (in some cases completely, in others until a revelation near the end) ‘where’ they are happening and what rules apply; alternatively, they can be fixed in ‘reality’, yet tease the possibility of something supernatural or impossible, often in the form of a single inexplicable thing. The Amazing Adventures of Kavalier and Clay, for example, has many events that are wildly improbable, yet that cannot be completely ruled out as impossible, with the exception of one single element that appears definitively unreal but possibly metaphorical or symbolic in some way. The book could be read as a magic realist novel, or maybe as surreal, but the reader is reluctant to do so because the element is so singular and marginal that it seems it cannot define the entire book. Likewise, The Prestige is entirely mundane, with the exception of one element. That element is important, and treated consistently as though this were a parallel fantasy; yet it is not explained or exploited enough to make the entire novel, which would otherwise be mundane, suddenly enter this fantastic genre. On the other hand, the same author’s The Affirmation may be read as entirely mundane, as entirely fantastic, or as two different stories belonging to two different genres. Many of the short stories of Du Maurier and Margo Lanagan likewise exploit this ambiguity between real and unreal. All these things I am therefore calling liminalia; for want of a better explanation, we might say simply that these are books that cannot quite, or cannot confidently, be called ‘mundane’ and non-fantastic, but in which the non-mundane elements are so small or so ambiguous that it also does not seem fitting to definitively assign them to fantastika.
Finally, we have macabre. Now strictly speaking we might be drawing all sorts of distinctions here between the gothic, the macabre, dark fantasy and horror… but since I don’t read much of this and have shelved even less, and currently this entire category for me consistes only of a single book (Dracula), I’m happy just using a single shelf for now.