I talk below about George R. R. Martin’s Song of Ice and Fire series, including the fourth and fifth novels. Where possible spoilers will be in white text (highlight to reveal) or otherwise signposted.
I’m a big fan of HBO’s Game of Thrones. I prefer to follow the story through the TV show (because it’s better than the novels), but recently caved and read all of the books in the series so far - too much time on my hands I guess. The bad news is that the fourth instalment is a stinkeroo of truly epic proportions. The good news is that it gives me an opportunity to hold forth pompously about the fantasy genre. The aim is a carefully argued slag-fest, though there might be some backsliding on the ‘carefully argued’ bit.
For anyone not familiar with the story - it is set in a fictional world, roughly modelled on medieval and early-modern Europe, in which powerful families compete for supremacy. As well as historical sources, it has clear debts to Tolkien (as most works in the fantasy genre do) but markedly improves on his model - Westeros, Martin’s fictional world, knows realpolitik, religion, and ideology (all those orcs and bad guys that follow Sauron – why do they follow him? What beliefs motivate them?). Most remarkably, seasons last for years, with winters long, unpredictable, and awful. At the fringes of this world and of the narrative are strange and disturbing stirrings of long-forgotten supernatural elements.
The marginal role of the supernatural is a virtue - it makes the books feel a little less Dungeons and Dragons, which is a virtue if you’re a snob like me - and this reticence is central to the series’ narrative ambitions, as I understand them. The first novel opens with an encounter with the ‘Others’, very unpleasant inhuman ice-dwellers who can reanimate the dead – the coming of winter is therefore ominous. This first glimpse of Westeros gives the reader a fuller insight into this world than its inhabitants enjoy, almost all of whom, we learn, believe the Others to be either long-gone from the world or possibly merely mythical.
This is promising, quite sophisticated stuff.
1. In minimizing the presence of the fantastical, Martin shifts emphasis from the existence of these monsters to characters’ belief in their existence. What is proposed as interesting about this world is not the implausible fact of the existence of monsters, but the fact that it contains different kinds of belief, an array of epistemological perspectives on what sort of world this is that we are reading about. This is sophisticated stuff – any controversial fictional truth must trigger reactions not just in the reader but also in the characters within the fiction. So this world is in some ways as complex as our own; although the novels are an exploration of a world resembling our own mythical past (it features dragons, e.g.) they do not, as much fantasy does, offer a wholesale abandonment or wishing away of our complex present for the lost simplicities of the mythical past (which is a problem with Tolkien). We stand in relation to Westeros’s ancient myths in much the same way as we stand in relation to our own – which fits M.’s ambition to create a historically verisimilar, rather than a purely mythical, world: this is not crude escapism, as scepticism still accompanies us. Martin therefore sets himself up for a more ambitious development, conceivable only in a fictional world – what happens to his characters’ belief and scepticism when the myths are proven true?
2. In foregrounding characters’ beliefs and knowledge about what exists in this world, it presupposes, or at least provisionally removes from the frame, or de-controversializes the question of their existence. We are not presented with a bald assertion that snow zombies (also called ‘White Walkers’) exist in this world, but with the second-order fictional reality of what is believed about them – ‘leaving aside whether or not you think White Walkers exist, what do you think of Tyrion’s opinion that they don’t?’ In other words, it bounces us into accepting an implausible fiction by a kind of misdirection. I’m a bit wary of labelling this ‘misdirection’ as there is no real trick –Martin is not really lying to us by claiming that snow-zombies exist in his notional world, or trying to make us presuppose as true a false proposition. He doesn’t need to persuade us of anything, as his fictional world is whatever he asserts of it. What he does need to do, however, is create a sense of the integrity of his world and the fittingness of what inhabits it – if he puts speaking humans in it (as of course he does) then we have no problem in believing that they belong there (it is non-controversial that notional worlds contain people like us, and does not need to be asserted as true in a given fictional world). Necromantic snow zombies, however, are created by authorial diktat and their presence in this world is caused by pure assertion rather than their belonging there. If not handled well, the Others would feel like an artificial construct intruded upon M.’s world as a sort of hypothetical fiction – ‘if I proposed a zombie-containing-world, what would you think?’
So M. avoids making a bald proposition that would make plain that his fictional world is a paper-thin assertoric construct that expands permissively with each arbitrary assertion he makes. The existence of his monsters is embedded in the beliefs of the characters that populate his world, thus preserving an illusion of its integrity. This sense of Westeros’s integrity contributes to the immersive, engrossing depth of his storytelling – rather than focus on our own relation of belief from without (do I believe in a world containing snow-zombies?) it focuses on the internal belief relations within Westeros. I will return later to the matters of narrator’s say-so and immersiveness
(at great and painful length).
3. The stage is set, then, for the introduction of the supernatural as a strange and unsettling development – as indeed would be the case were we to discover tomorrow that dragons or ghosts or honourable politicians actually exist. This is where we see that the realist elements of Martin’s style are key to his (hear me hesitate as I type) literary ambitions. In an example of what classicists call aemulatio, M. makes a tendentious response to his fellow fantasy authors in which he tests how their creations would fare transplanted into and re-run in a more realistic world. It is a counterfactual ‘what if’ that tests and exposes the under-explored inferential possibilities or under-imagined hypotheses of the fictional worlds of others. My exemplar for this sort of tendentious re-examining is Alan Moore’s Watchmen, in which he asks what it would be like if a superhero (of the Superman rather than Batman type) existed in a world similar to our own. Terrifying, uncanny, destabilizing, downright dangerous, is the answer.
[I’m not especially well-versed in the fantasy genre, so much of what follows might be junk – but I’ll have a pop nevertheless] Martin seems to go in for this sort of competitive re-imagining. Jaime Lannister is the best example – a Prince Charming in golden armour, handsome, knightly (sort of), famed in song and story. Except, as M. points out, Prince Charming were he to exist, would almost certainly be a violent nihilist. Of course he would – he would be a man who spends his days merrily hacking people into lumps of bloody meat, all the while caring only, but intensely, about his looks and his hair. It must follow that a man like this would see little meaning in life (or do I just associate Jaime with Prince Charming because he looks so much like the prince from Shrek 2?). There are other signs of an emulous (i.e. imitative but competitive) relationship with Tolkien – markedly Tolkienesque terms like ‘wight’, ‘warg’ are borrowed, though it’s noticeable he steers clear of elves and dwarfs. In fact his making Tyrion a dwarf (in sense of ‘little person’ rather than ‘member of bearded race adept at mining’) is perhaps another corrective. The narrative structure of parallel journeys across the fictional world, with occasional crossing of paths, partings, splits, etc. is clearly based on the Tolkien model, though I’m not sure M. does much to improve on it. Sensibly he by and large renounces Tolkien’s creation of whole languages and instead focuses (successfully) on creating idiom – epithets (‘the Mountain who rides’, ‘the Beggar King’, ‘knealers’ for civilized folk living under kings), mottoes and catechisms (‘Winter is Coming’, ‘a Lannister always pays his debts’, ‘the night is dark and full of terrors’), proverbs and truisms (less successful – ‘words are wind’, ‘dark wings, dark words’), and phrases fixed by usage (‘take the black’, ‘pay the iron price’). His greatest success in creating a universal portrait of a fictional people – language, culture, belief – is by far the Mongolian-style ‘horse lords’ of the Dothraki (it is known).
4. But I digress. The series begins, to borrow a term from the narratologists, with a prolepsis. The political and quasi-historical realism of the story is imminently to be turned on its head with the irruption of the supernatural – snow-creatures and dragons (ice and fire) are waiting. To use one of M.’s most effective and pregnant mottoes, ‘Winter is Coming’ and with it an ancient horror.
Except, at this rate, winter is going to be a long time coming. Martin is a chronically slow, inveterately waylaid storyteller. I wonder sometimes if his fondness for the pregnantness of foreshadowing and the weightiness of foreboding words exceeds his narrative skill to plot his way to their fruition. ‘Winter is coming’ (the motto of the Starks, a family from the wintery north) is ominous, but he needs to make good on it and [spoiler] doesn’t until the end of A Dance with Dragons, which is far too late in the day if he plans to wrap things up in seven books. [end spoiler] He enjoys creating prophecies, and riddling allusions to what is to come (‘the dragon has three heads’, Cersei and the ‘valonqar’, Arya and the dwarf-woman seer / Melisandre in the TV series) but there is a feeling that it might be slightly gratuitous – that this isn’t obeying the structure of a narrative, perhaps, but attachment to foreshadowing as a mere stylistic adornment, with no pay-off in sight.
A major problem is that Martin’s chosen method of narration inevitably holds up the progress of the story. Each chapter is told from the third-person perspective of a character, usually one of the major ones, so that although the action is spread over a large world, it unfolds locally according to how the variously located witness characters experience it and reflect on it. Curiously, M. simply names chapters after the character from whose perspective it is narrated – so any one novel might have three or four chapters with the same name (though he seems to be gradually moving away from this a little in the most recent novel). It reminds me of what a friend (Patrick are you reading?) told me of sticking titles on to essays, chapters, theses, etc. – reluctance to name something is a sign you don’t love it. I can understand this – a discomfort or ambivalence with what you have made, such that you don’t want to make it a thing in its own right, with its own name, ready to stand up on its own two legs out there in the wider world. Given how painfully slowly M. writes, it seems possible that behind the chapter titles lies a tortured difficulty in being happy with what he has written.
But I digress – how was it ever thought a good idea to tell such a large, sprawling story (‘epic’, if you’re feeling generous) through limited character perspective, and not only that but through the perspectives of only a few chosen characters? Is anyone editing this stuff? The more characters he introduces the more characters there are that have to be given chapters [spoilers](the tedious Dorne chapters in A Feast for Crows). [end spoiler] Moreover he has to stick to relatively consistent chapter length, so as not to favour or neglect individual characters and thread (different readers favour different storylines); and each chapter needs to feel like a significant development of the character’s arc, and has to give the reader time to settle themselves into the new viewpoint – so M. can’t flit, and just ‘check in’ with a character quickly, or skim across the different perspectives or easily contrast events in different parts of the narrative (unlike, say, the Gormenghast novels). So things are happening all across Westeros that at any one point we mostly cannot know about. He has put his characters in charge of the narrative, not a narrator, and he and his readers pay dearly for this abdication. As a result he gives us an ever-expanding world with a proportionally ever-diminishing perspective from which the reader views it, and as he introduces more and more threads (and boy does he), the action as it narrated is diluted, an ever smaller drop in a growing ocean. It is interesting that the credits to the TV series feature an impressive panoramic sweep of a map of Westeros that expands as the story develops – reading the novels we feel ever more distanced from this bird’s-eye view of the world created because any perception of its totality recedes ever further from us. We are restricted at all times to a small patch of narrative turf in a giant world, and the more that world expands the more we feel at any given point that we are missing out – it is like a cruel peep show.
Why hasn’t an editor pointed out that this strictly episodic, inflexible narrative style is completely unsuited to such an ambitious enterprise in fictional world-building?
But what is really extraordinary is that effective, or suitable narration, is sacrificed for so very little! Characters’ internal dialogue all too often just re-emphasizes their central conceit or obsession or some basic shaping attribute (‘I wish I had both hands’ says Jaime; ‘I’ve decided not to be Arya any more’ says Arya; ‘I wish I wasn’t craven’ says Samwell; ‘I am Cersei and I am a lion’): it does very little to portray the life of the mind in the characters, to deepen them and make them psychologically interesting. Worse, all too often the characters’ internal voice isn’t really a separate voice at all, just a counterpoint to the narrator’s voice – it continues on from or very basically complements the main narrative voice (summarizes pithily or ironically e.g.), in a way that feels more like Martin’s voice than the characters. Below I follow M. in italicizing the ‘internal voice’:
was as amiable as he was clever, but too lowborn to threaten any of the great lords, with no swords of his own. The perfect Hand.
Baelor the Blessed once had visions too. Especially when he was fasting. “How long has it been since you’ve eaten?”
How is the character able to interject into a train of thought begun by the author (esp. evident in the final example)? He or she shouldn’t know what the narrator is saying. (Incidentally I probably don’t need to obscure characters’ names here, as the Riverlands scenes are so lifeless I’m not sure what I’d actually be spoiling). But I am being harsh here, especially as when Martin likes a character he portrays him very well (and it’s always a him – problematic again), Tyrion being the best example. That said, the most immediate and obvious criticism one can level against A Feast for Crows is that [spoiler] he abandons all of his most likeable and indeed strongest characters for the entirety of the novel, to the extent he has to apologize for this in a postscript. [end spoiler]
I wonder if M.’s obvious attachment to Tyrion, which I imagine most readers share, brings us a little closer to what lies behind his struggle to get going and deliver the plot that is promised – that he is too attached emotionally to the world he has created to turn it upside down, as must happen come winter.
But this is part of a larger argument I want to make, and I will return to it. There are other interesting reasons (well I find them interesting) why Martin might holds off from plunging into supernatural free-for-all this is promised.
First, his hedging with the supernatural only makes its full introduction all the more difficult. Having set himself up as positing a more realistic fantasy world, in which the claims of fantasy can be better, more sceptically interrogated, how does he then discard his scepticism and pull back the curtain, without seeming to revert to the kind of fictional world he is trying to resist or improve on, in which narratorial diktat runs riot?
Once M. really gets to depicting a world that contains fantastical necromantic snow-zombies he will find his narrative in an anything-goes fantasy world, in which what is possible is just whatever the story asserts to be possible as it goes along – and this is quite other (or Other) to the more nuanced style he has cultivated so far, in which his world allows the reader new insight into real-world, shared realities rather than into merely fabricated, ad hoc fictions. So, for instance, he gives us the chance to know better the typical figures, the types, of fantasy fiction (which exist as literary and literary-critical realities) by instantiating them in more sophisticated, complex form – I said before that Jaime Lannister is interesting because he shows the new Prince Charming-attributes that are inferable within this world (sadism, nihilism, etc.). The same can be said of the real-world historical elements that are incorporated: we get to know (in a hypothetical, what-if sort of a way) what it might be like to experience a Wars of the Roses-type situation, experience of the real thing being now impossible.
But how can we get to know the Others better when, being fictional, any description of them constitutes part or all of what can be known of them? They are created as things, things we can know about and quasi-experience, by the very act of describing them. Yes, Martin can make us know them, but as this knowledge is tautologous, it is rather cheap – the information he conveys as narrator isn’t about anything, except itself, and so isn’t really worth knowing. Because White Walkers do not exist as things in the world, unlike the Prince Charming archetype, it is otiose to attempt to make further inferences of them, to interrogate them – it is an arbitrary investigation. There is nothing wrong with this, at all – it is central to how fiction works. But for Martin it marks a big shift in the sort of world and the sort of knowable reality that he takes as his theme – and he hesitates in making this shift because, I think, he hasn’t figured out how to do so without radically altering his project and writing, essentially, a different kind of fiction, one epistemologically thinner and weaker.
After all, it is of greater interest to write a story (as Alan Moore did) proposing that Moriarty once employed Allan Quatermain and Dr Jekyll, than to propose that Carl Sparrowhawk once had his portrait painted by Helmut Schnarbel (Sparrowhawk was a famous surgeon and Schnarbel a celebrated artist, in the fictional world I created just now of which the previous statement was true). Fantasy is less sophisticated than literary fiction because the latter deals with people, things, and ideas and in doing so all of the propositions that are made of them. It is a contribution to a discourse. Tolkien invented languages in which no one had thought, and wrote about a world pristine and untouched by ideas. This is why Tolkien's work is juvenilia.
Incidentally, this doesn’t just affect M.’s portrayal of the supernatural. The novels’ appendices contain long, painfully long, family trees of the various families of Westeros. Now maybe these family trees underpin the whole thing – maybe he devised them first as the quasi-factual bedrock of the world that he then described, as a sort of maquette (just like Graham Greene wrote the novella The Third Man purely to provide a disposable model for the screenplay of the same name). Maybe too some readers find them indispensable to their understanding and enjoyment of the plot. Fine. I wasn’t one of them. I found them almost improvisational, as if the story was being made up there, on the spot, right before my eyes, with no illusion whatsoever of its pre-existence. It felt utterly flimsy and erasable, as if I could say ‘John begat James who begat Joseph who begat Jeremy’ and hey presto! I just told a story. It reminds me of that trope in TV and film in which a story is narrated while it is simultaneously being sketched, with the animation being added to and erased as the narrative proceeds. There’s probably a name for this but I’m too tired to go searching TV Tropes.
So Martin can’t get on with his plot because its inevitable destination might diminish his novels’ thematic sophistication. I wonder if he is also held back by other fictive problems – the return of the supernatural threatens ominously to turn his world upside down, but on the other hand any extension of a fictional premise is (for the reader) banal, because the only previous assumptions or beliefs undone were themselves similarly unreal. Nothing is overturned or undone or lost in redefining the reality of a world that was only ever notional – it is the same expansion, the same licence as that which created that world in the first place. If, shock! horror! it is found that snow-zombies and dragons do exist in the fictional world of Westeros, then we simply accept it as the perfectly admissible redefinition of a flexible premise, because we agreed to begin with that Westeros was a made-up world. So, in fact, it is difficult to create irruptive elements, anomalous presences in fantasy worlds because their borders are already so accommodating. Compare, for instance, the very underwhelming revelation in The Matrix Reloaded that the hero Neo can now fly – so what, we think, he’s a computer-generated fantasy in a world ruled by robots. Consider my mind unblown.
So perhaps M.’s plotting blunders in not having everyone in Westeros go batshit crazy after the battle on the Fist of the First Men (‘what do you mean they exist? Are you kidding me? This is terrifying news.’), and in having certain bits of news travel unaccountably more slowly across his world than others (why don’t more people know about the dragons and White Walkers yet?), are explicable – a fear of bathos. What should be big news in his fictional world might seem to the reader just another otiose expansion of the fictional premise.