Here in the UK, the British Library has just opened a major exhibition on the subject of Fantasy literature. With input from the likes of Neil Gaiman, Aliette de Bodard and Roz Kaveney, it is delighting fantasy fans who visit or attend the various online events that support the exhibition. Naturally, as part of the curation process, the British Library team will have had to decide what counts as fantasy, and what does not. As anyone involved in academic discussion of genre categories knows, this is not a simple question to answer.
This article is not about defining genre boundaries. Huge amounts of ink and electrons have been spent over such debates already. And there are good reasons for such discussions. But they have problems. The first is that most good writers, given a definition of what they should be writing, will immediately look for ways to subvert those expectations. As Gary K Wolfe theorized in his book, Evaporating Genres, genres are continually evolving and leaking into each other. Also, genre boundaries are often used as an excuse for gatekeeping. I am fairly sure that there will be some crusty old supporters of the British Library who are shocked! Shocked, I tell you! That such a venerable and respectable institution should lower itself to running an exhibition about “mere fantasy”.
My contention here is not about genre boundaries; it is about exploding them.
If you were to ask the hypothetical person in the street what fantasy fiction is, there is a good chance that they will say it is stories about dragons and wizards. That’s not a bad first pass. After all, the two most famous fantasy worlds – George Martin’s Westeros and JRR Tolkien’s Middle Earth – do fit the bill. Westeros is famous for its dragons, and Middle Earth for a wizard. But there are forms of fantasy that do not fit this definition.
One of the more popular forms of fantasy being published at the moment is what you might call mythological fantasy. It consists of re-tellings of famous stories from myths, usually Greek and Roman tales, but other cultures are being mined as well. There are no dragons in Greek mythology, though there are many other supernatural monsters. There are no wizards either, because doing magic seems to be reserved for women. Interestingly, the authors of such books often do not come from the SF&F field. That doesn’t mean they look down on us. Natalie Haynes, for example, whose book Stone Blind, about Medusa, is doing very well, is better known as an historian. However, she also did a wonderful interview with Susan Cooper to kick off the British Library’s fantasy event series, and in that she clearly showed herself to be a fan of fantasy literature.
The bookstores, at least in the UK, tend to shelve books such as Stone Blind in General Fiction rather than in Fantasy & Science Fiction. Gatekeeping at work.
Another form of fantasy that often has neither dragons nor wizards is contemporary fantasy, particularly paranormal romance which tends to major on fairies, and on creatures also associated with horror fiction such as vampires and werewolves. I am, as her publisher, honour-bound to mention Juliet E McKenna whose Green Man series has done amazingly well for a small press. McKenna has upended the trope, using a male hero, a rural setting, and female monsters, and it has gone down a storm with readers.
Mention of horror also reminds me that large amounts of horror fiction are also fantasy. You can have horror that is all about serial killers, but standard tropes include vampires, werewolves, zombies and ghosts, all of which are undeniably fantastical. You can make an argument that horror is a subset of fantasy, and for many years the World Fantasy Convention was primarily about horror.
It seems, then, that the definition of fantasy can be much broader than books about dragons and wizards. I’m now going to go broader still, because in many ways the genesis of this essay was in a panel at the European Science Fiction Convention. In collaboration with others including Martha Wells and Jukka Halme, I was asked to opine on the subject of alternate history. In order to do so, we had to decide what alternate history was. And there I ran into a problem, because it turned out that most of my favourite alternate history books involved magic.
For example, there is The Dragon Waiting by John M Ford. This is a book about the Wars of the Roses, and the only dragon in it is Henry Tudor. However, one of the major characters is a young man called Hywel Peredur who has been to university in Byzantium to develop his sorcerous skills and now works as a court magician.
Then there is Ash: A Secret History by Mary Gentle. This massive book, published in four volumes but very much a single narrative, is set in mediaeval Europe. Our heroine, Ash, becomes a commander in the armies of the Duchy of Burgundy, a very real mediaeval state. However, the war that the Burgundians are fighting is against the Empire of Carthage which, through sorcerous means, has survived since Roman times.
Finally I have Anno Dracula by Kim Newman, in which Count Dracula has survived the events of Bram Stoker’s novel and has married Queen Victoria. He is now the ruler of the British Empire.
All of these books have excellent historical underpinnings, and yet they also all contain supernatural elements. I’m pleased to say that my fellow panelists were happy to accept them as alternate history, but the “official” definition of the genre disagrees.
Here I turn to a fine new book from Jack Dann. The Fiction Writer’s Guide to Alternate History does what it says on the tin, and does it very well. But the inevitable starting point of such a discussion is to define alternate history. Here Dann turns to a framework provided by the critic Tom Shippey. It goes as follows:
Alternate History should be based on a single divergence point from actual history which should be 1) plausible; 2) definite; 3) small in itself; and 4) massive in consequence.
That in itself does not preclude fantasy, but the word ‘plausible’ is something of a problem.
What Dann does really well in this book is to find a bunch of his peers, fellow writers of alternate history, and ask them for their views on the various definitions and strictures he has put forward. While some writers do embrace the Shippey definition, others do not, and everyone’s take on the subject is enlighteningly different.
Those writers – mostly men – who approve of the Shippey definition tend to take a very analytical approach to their fiction. Kim Stanley Robinson takes a science-fiction approach to the topic. He postulates a divergence point, and he develops history from there according to what he sees as the rules of the subject. He develops future histories in much the same way. This similarity of technique is probably why alternate history has long been deemed to be a subset of science fiction.
Dann and his contributors spend quite a bit of time discussing whether works are genuinely alternate history, or if they are instead historical fantasy. The dividing line here tends to lean heavily on the plausibility element of the Shippey definition, but there are other areas of disagreement too. John Buckingham states firmly that a divergence point is essential: “If you don’t have a divergence point […] then you don’t have alternate history. You have fantasy.” However, Janeen Webb suggests that a single divergence point is not necessary, and her argument is in part that real history doesn’t work like that. We’ll come back to that idea later.
Harry Turtledove dismisses the entire concept of genre rules and says that plot, characters and style are all that matter when writing a novel. Given that his The Guns of the South is one of the most famous alternate history novels yet written, you would think he knows what he is doing. But Terry Bisson says that any alternate history containing implausible science such as time travel is just fantasy. Turtledove’s novel famously depends on time travel.
Along the way, Jack Dann introduces a new subgenre that he calls “transcendent counterfactual fiction”. In this he puts Anno Dracula, and also the steampunk Cairo stories of P Djèlí Clark. It isn’t entirely clear to me how this subgenre is defined, but I have a sneaking suspicion that it is actually comprised of historical novels with supernatural elements that Dann thinks are sufficiently good in their historical research and presentation to escape the “mere fantasy” tag.
A major point of contention amongst Dann’s contributors is the story “Ike at the Mic”. Written by Howard Waldrop and first published in Omni magazine in 1982, it was a Hugo finalist in Short Story the following year, and has since been collected in many volumes. The basic plot involves Dwight D Eisenhower becoming a jazz musician rather than a soldier, and Elvis Presley entering politics, eventually becoming a senator.
Some authors clearly thought that Waldrop’s story was revolutionary, and in some ways freed the genre of alternate history from shackles that had been holding it back. Others pointed out that Eisenhower had no known musical talent, and Presley showed no interest in, or talent for, politics. In their view this made Waldrop’s story bad alternate history. John Kessel went so far as to describe the story as “completely bogus.”
My favourite contribution from a writer comes from Charlaine Harris. She says, “It was relatively easy to pick the divergence point for the Sookie Stackhouse novels: vampires have decided to make their existence public.”
Obviously if you believe that vampires are clever enough to hide their existence from us then this perfectly fits the Shippey definition. If you don’t, well maybe the Sookie Stackhouse books are just fantasy. Is it entirely subjective?
At this point we should probably turn to Tolkien’s famous essay, “On Fairy-stories”. This is the relevant part:
Children are capable, of course, of literary belief, when the story-maker's art is good enough to produce it. That state of mind has been called ‘willing suspension of disbelief.’ But this does not seem to me a good description of what happens. What really happens is that the story-maker proves a successful ‘sub-creator’.
What Tolkien is saying here is that children will willingly suspend their disbelief in order to enjoy the story, but when you are writing for adults you, the writer, must convince them of the reality of your invented world with the quality of your sub-creation.
What he doesn’t say is that this is inevitably a bit subjective. If you are reading a novel set in Roman Britain then your enjoyment of it will depend, not only on the skill with which the author has invoked the setting, but on your knowledge of that particular time period. If you are an expert on the Roman Empire, or your knowledge of Roman history has come largely from right-wing documentaries, you may be less inclined to believe the author’s hard work in trying to get the history right.
So is alternate history fantasy? Some of it clearly is, if it contains supernatural elements. Some of it claims not to be, because it is solidly grounded in real history, and knowledge of how history works. But is it? It is time to talk about how history actually works.
The basic concept of history, as embodied in the Shippey definition, can be summed up in a nursery rhyme that was popular when I was a child. It goes like this:
For want of a nail the shoe was lost.
For want of a shoe the horse was lost.
For want of a horse the rider was lost.
For want of a rider the message was lost.
For want of a message the battle was lost.
For want of a battle the kingdom was lost.
And all for the want of a horseshoe nail.
There we have a small, definite point of divergence that is entirely plausible and has massive consequences.
The Shippey definition is also reliant, to a certain extent, on Thomas Carlyle’s Great Man Theory of history. Writing in On Heroes, Hero-Worship and the Heroic in History, Carlyle states:
Universal History, the history of what man has accomplished in this world, is at bottom the History of the Great Men who have worked here.
While popular in Victorian Britain, this idea has fallen out of favour with historians, not least for the reason that many of them are women. The idea that persons of one’s gender can have no influence on history does not go down well. But in practice the idea that notable persons of any gender drive history forward has lost currency.
History is complex, messy, the product of the actions of many people. Marxist historians will champion the idea of Historical Materialism, meaning that history is driven by the conflict between the ruling classes and the proletariat. But even that is probably too simplistic.
“For the want of a nail” shows a clear and logical progression from our simple point of divergence to an inevitable conclusion. But how likely is it that such Holmes-like deduction is possible for any divergence point? We should bear in mind that complex systems are often inherently unpredictable. Weather systems, for example, are so complex that we have the Chaos Theory example of a butterfly flapping its wings causing a thunderstorm on the other side of the globe. If history is that complex, the idea that its progression can be predicted from a single divergence seems fanciful.
Of course that hasn’t stopped science fiction writers from trying. The most famous example is Isaac Asimov’s concept of Psychohistory, in his Foundation and later stories. Asimov was a chemist by training, and chemists are well aware of the unpredictability of the behaviour of individual molecules. Fortunately, statistics comes to their aid. Molecules are very small, and a test tube can contain an awful lot of them. The basic unit by which chemists measure quantities of molecules is not the individual, but the mole, which represents 6.02214076×10²³ molecules. When working with numbers that large, the behaviour of simple molecules becomes highly predictable, and the behaviour of complex ones can be understood in terms of the likely percentage yield of an experiment.
Asimov theorised that history on planet Earth was too unpredictable for science, but given a galactic civilization containing trillions upon trillions of humans, statistics would again come to the rescue. This is the basis of Harry Seldon’s theory of Psychohistory. Of course Asimov almost immediately punctured the theory by introducing a powerful rogue element, The Mule, but the basic idea remains.
Another possible theory of history is that of the unstoppable tide, and we find this used in Kim Stanley Robinson’s The Years of Rice and Salt. We know that, without any apparent intervention, our universe progressed from the Big Bang to produce stars, solar systems, planets, life, and eventually civilisation. What if the development of civilisation was equally pre-ordained? The Years of Rice and Salt postulates a world in which the Black Death killed off almost all of the population of mediaeval Europe; human progress continues pretty much exactly as it did in our world, except that a whole bunch of major developments occur in Asia, Africa and the Americas rather than in Europe. The descriptions of scientific breakthroughs are modelled exactly on experiments carried out in our world by Europeans.
So we have seen that history is probably way too complex for the sort of logical progression from a divergence point that the Shippey definition requires, and that even a massive change might have little effect on the course of history overall. This doesn’t look good for the definition of alternate history. In particular it casts doubt on alternate history’s claim to be an analytical process of rational development from the divergence point to an inevitable set of new outcomes. And if that claim can’t be substantiated, then all alternate history is just made up. Or, in the words of some of Jack Dann’s correspondents, it is fantasy.
The complexity of history strikes at the definition of alternate history in other ways too, because history is not static. Like science, it evolves as new information becomes available. This has direct implications for historical fiction. A favourite book of mine from my childhood was The Eagle of the Ninth by Rosemary Sutcliff. This books tells the story of the Roman Ninth Legion (Legio IX Hispaniola), which was based at Eboracum (York).
When Sutcliff wrote the book, it was a generally accepted historical fact that this legion disappeared mysteriously at some point during the history of Roman Britain. The most popular explanation was that it had gone on campaign in Scotland and been destroyed by the Picts, and Sutcliff wrote the book on that basis. These days Classicists are unconvinced by this argument. They note that the loss of a legion was a major event that would have been widely written about, and the commanders responsible shamed. No such records exist. It is therefore more likely that the Ninth was quietly re-assigned to another part of the empire where it was more needed, perhaps the Rhine frontier.
What does this mean for Sutcliff’s book? Is it now alternate history, because it describes events that historians believe did not happen?
A more dramatic example concerns the so-called Princes in the Tower, the young heirs to the Yorkist dynasty who were taken to the Tower of London by their uncle, Richard III, and supposedly never seen again. For centuries historians have believed that they were murdered there, either on the orders of Richard, or by agents of Henry Tudor following his victory over Richard at the Battle of Bosworth.
Finding out what actually happened has proved difficult because most of the administrative records that might have cast light on the subject were destroyed by agents of Henry VII. However, there are records that the King of England can’t touch. Recent work led by historian Philippa Langley, who was previously famous for leading the team that discovered the grave of Richard III, has focused on records in Europe. There they have found clear evidence that the two princes left the Tower of London and continued efforts to regain the throne long after Bosworth.
The consequence of this is that large amounts of historical fiction, including Richard III by William Shakespeare, have now become alternate history, because they rely in part on an event that did not happen.
The problem may even be wider, because writers of historical fiction can never be 100% certain that their interpretation of history is correct. Authors like Hilary Mantel and Philippa Gregory do amazing amounts of research for their books, and are lauded by historians for it. But was Mantel able to know the inner workings of the mind of Thomas Cromwell, or Gregory the minds of Anne and Mary Boleyn? Of course not. What they have done is decide how they think historical personages thought, and developed their plots on the basis of that. That’s pretty much what an alternate history writer does.
One writer who takes exception to this historical mind-reading is Guy Gavriel Kay. Indeed, he holds that it is immoral to write historical fiction about real people, because we cannot know how those people actually thought and felt. Instead Kay has created a parallel universe that he describes as a “quarter-turn” from our reality. In it, history progresses in pretty much exactly the same way as it did in our world. All of the identifying marks are still there. Only the names have changed.
Kay’s works, which span a wide range of history from Celtic Britain and the Byzantine Empire, through to Renaissance Italy, are impeccably researched and make for wonderful historical fiction. They do contain homeopathic quantities of magic, but in ways that make no obvious difference to the outcomes of the stories. And yet, because they are set in an alternate universe, they are generally viewed as fantasy, and shelved as such by bookstores.
Even some of the best historical fiction is, frankly, fantastical. Like many of my friends, I am a huge fan of Dorothy Dunnett’s Lymond Chronicles, beginning with The Game of Kings. Dunnett’s historical research is superb. But is the handsome, charming and multi-talented Francis Lymond really a plausible historical character? Or is he more of a Tudor version of James Bond? Given the number of women readers (myself included) who have a massive crush on him, I suspect the latter.
It is now time to turn our attention to science fiction. Earlier I noted that the processes of writing alternate history and science fiction are very similar. Both start with a proposed change in the world – either an historical divergence point or a new scientific discovery – and work forward from there to predict the consequences. But, as we have noted, such predictions are not really as logical and inevitable as writers might like to pretend, and if this holds for alternate history it must hold for science fiction as well.
Many science fiction writers make no pretense to following such strictures. We have already seen Terry Bisson complain that alternate history containing implausible science such as time travel is really fantasy, but much science fiction is based on such ideas. Indeed, there is an entire subgenre of science fiction that happily throws away any pretense to scientific plausibility. Space opera regularly features faster-than-light travel, actually intelligent artificial intelligences and so on. Writers such as Iain M Banks and Gareth L Powell have done very well with such works.
We have also noted that science evolves. What was once thought to be scientific fact is now known to be wrong. The Italian astronomer, Giovanni Schiaparelli, using the best telescope available to him, observed lines on the surface of the planet Mars. He suggested that these were evidence of canals built by an advanced civilization that lived on that planet. The idea caught on, and any science fiction written as a result would have presented the Martians and their canals as fact.
Science fiction, as a genre, has come to terms with the fact that many of its more popular ideas have long been proved false. This has led to the creation of Steampunk, a subgenre which takes Victorian-era ideas about science and writes science fiction based on them, as if they were true. Such books clearly exist in an alternate past, and may project forward to an alternate future. They are not based in our world. By the strict definitions of alternate history and science fiction, they are clearly fantasy.
One of the most obvious examples of the cross-over between science fiction and fantasy is the Star Wars universe. This was brought into stark relief recently with the release of the TV series Ahsoka. In this series, much of the action was set on a planet that seemed very reminiscent of Mordor from the Lord of the Rings movies. On that planet, people got around by riding giant wolves. Some of the major characters were a group of witches who, amongst other things, created zombie Imperial Stormtroopers. Miraculously, these were even less competent at firing blasters than their living counterparts, a feat that was previously deemed impossible by many observers of the franchise. Just to rub things in, the season finale of Ahsoka was titled, “The Jedi, the Witch and the Warlord”, a clear reference to the Narnia books of CS Lewis. And that episode featured images of giant statues carved into a mountain that once again evoked memories of the Lord of the Rings movies. A subset of fans was outraged. How dare Disney pollute their favourite science fiction universe with elements of fantasy? And yet these same fans are entirely comfortable with stories about a group of space wizards who fight duels with laser swords.
Similar examples of cross-over between science fiction and fantasy exist in literature. One of the most famous is the Dragonriders of Pern series by Anne McCaffrey. These books have always been shelved under fantasy in bookstores because they contain dragons. And yet, in writing the books, McCaffrey produced an entire science-fictional backstory for them. The planet, Pern, was settled by human space travellers. Finding themselves menaced by Thread, they bioengineered a local species of flying lizards to create the dragons that are a signature feature of the books.
The canals of Mars have been resurrected by Chaz Brenchley for his Crater School stories about a girls’ boarding school on Mars. Are these books science fiction as they are set on Mars, with actual space travel and Martians? Are they fantasy, because we know that neither Martians nor the canals ever existed? If the books were not set on Mars they would probably be classed in the subgenre of realist fiction comprising stories for girls. Categories are slippery.
Iain M Banks is most famous for his space opera series set in the galactic civilisation known as The Culture. However, he wrote one book, Inversions, that has been reviewed as fantasy. It is set in a mediaeval society with mostly mediaeval levels of technology. But not all is as it seems. One of the few governmental organisations in The Culture is Contact, a group of people (humans and mindships) charged with managing relations between The Culture and other civilisations. Within Contact there is a black ops division known as Special Circumstances, and members of this group regularly provide plots for Banks’ novels. Readers of Inversions who are familiar with The Culture will soon realise that two of the characters in the book are Special Circumstances operatives who have become stranded on a primitive world. The magical dagger that appears in the story is clearly a well-known Culture artefact known as a Knife Missile. It is all very reminiscent of Sir Arthur C Clarke’s observation that any sufficiently advanced technology is indistinguishable from magic.
Even without relying on Clarke’s Law, however, we have determined that much science fiction is not solidly based on plausible scientific extrapolation. And we have seen that the whole idea of such extrapolation is somewhat dubious given the complexity of how history actually works. It is not much of a stretch to say that all science fiction is actually a type of fantasy.
Before leaving genre fiction behind we should take note of the subgenre known as magic realism. The unkind definition of such books is that they are fantasies written by people from Latin America that have been re-classified in order to allow literary snobs to accept the likes of Borges and Marquez as “proper literature” rather than “mere fantasy”. A more useful definition is that they are books that contain magical happenings because the characters in the books believe that magic exists. No claim that magic is real is made; it is enough that the characters believe it is. Bookstores generally shelve such works under general fiction, not under fantasy. But if they contain magic, surely they are fantastical.
I note that, by that definition, the historical fiction of Guy Gavriel Kay is magic realism, not fantasy.
That brings us neatly to so-called mimetic or realist fiction, which is set in our world, and has no speculative or supernatural elements to it at all, honest.
Yeah, right.
Look at James Bond, for example. Granted Ian Fleming did base Bond on his experiences working in military intelligence during World War II. But these days Bond has drifted far beyond that. The movies have more in common with superhero franchises than anything else.
Someone who knows a lot about realistic crime fiction is Nick Harkaway, because his late father, who wrote under the name of John le Carré, was a master of the form. Harkaway, being of a younger generation, began his career as a science fiction writer, and I still recommend his debut, The Goneaway World, as having one of the best twists in any novel. But I don’t mention Harkaway because of his fiction, I do so because of a speech he made as Guest of Honour at a science fiction convention in Dublin some years back.
In this speech, Harkaway made the case that most contemporary, realist fiction is actually set in a parallel universe, because the characters in these novels do not have access to mobile phones, or even the internet. The plots of these books, he maintained, would fall apart if such technology was available.
I suspect that, in the years since Harkaway made that speech, writers of contemporary fiction have become better at incorporating modern technology into their books. In particular, young writers who have grown up with the internet will be much less technophobic in their outlook. It is probably also the case that authors have taken to setting their books in the recent past in order to avoid the complications that technology brings. However, for some contemporary fiction, Harkaway’s charge that it is set in an alternate reality holds true. And that makes it fantasy.
The important point here, I think, is that all writers of fiction make stuff up. If you check their bios on social media you will often find them saying things like, “I tell lies for money.” One of the regular users of lines like that is Neil Gaiman. His version is, “We who make stories know that we tell lies for a living.”
However, that is part of a much larger quote. Here’s the full version:
We who make stories know that we tell lies for a living. But they are good lies that say true things, and we owe it to our readers to build them as best we can. Because somewhere out there is someone who needs that story. Someone who will grow up with a different landscape, who without that story will be a different person. And who with that story may have hope, or wisdom, or kindness, or comfort. And that is why we write. – Neil Gaiman, The Graveyard Book
What Gaiman is saying is that all fiction writers make stuff up, and they all do so in order to tell stories to their readers. Those stories will not be true, but they may say true things. And that is exactly what fantasy does. A writer can make choices about how far from their real world the world of their story will stray, and how much effort they will put into convincing the reader of the reality of the story world. The reader also has choices to make as to how willing they are to accept the reality of the story world. But ultimately all stories are made up.
They are, to one degree or another, fantasy.
Cheryl Morgan is a writer, editor, and publisher. She is the winner of four Hugo Awards and is the owner of Wizard’s Tower Press. Her non-fiction has appeared in a variety of venues including Locus, the SFWA Bulletin, the Science Fiction Encyclopaedia, Vector and Strange Horizons. Her fiction has appeared in a number of small press magazines and anthologies. Cheryl was a Guest of Honour at the 2012 Eurocon in Zagreb and the 2019 Finncon in Jyväskylä. She was a keynote speaker at the Worlding SF academic conference at the University of Graz in 2018, and at When It Changed, an online conference on feminist science fiction organised by the University of Glasgow in December 2022.