Consolatory fantasy typically ends with the existing power structure back in charge, and they’re usually the good guys – no matter how unfair the society – so as a result I suppose that could be seen as optimistic. Of course, the bad guys are always much worse. Most space operas follow a similar set-up. If it’s not the barbarians at the gates, it’s the rot from within. Either way, the empire or republic is in for a kicking and the good guys have to put up the good fight to save it. If the empire does go down in flames, a new more powerful one will rise phoenix-like from its ashes. So far, so consolatory.

I will happily admit I deliberately set out to pastiche the consolatory fantasy template when I wrote A Prospect of War. Here’s the emperor – he’s under threat. So here’s a posse of good guys all set to fight the dark lord and defend the throne. And so the plot of the novel pretty much kicks off the conspiracy and sees the peasant hero gather his forces for the final battle.

However, part of the fun of writing the sequel, A Conflict of Orders, was then carefully upsetting that structure. The final battle takes place halfway through the book, rather than at the end of the trilogy. The villain is defeated (that can hardly be a spoiler) and the throne is once again safe… And then the tone of the story changes…

There is a plot hiding beneath the story of the An Age of Discord trilogy. Hints and clues to it appear in both A Prospect of War and A Conflict of Orders, and it was always my intention to bring that plot into the light and resolve it in the third and final book, A Want of Reason. But in the years since I finished writing A Conflict of Orders and now – when I have to write A Want of Reason from scratch to complete the trilogy – I’ve changed my mind about a lot of things. Not least what happens in A Want of Reason. Part of this is practical – I put together lots of notes for the third book back when I was writing the first two, but those notes now sit on a dead computer and are inaccessible. But it’s also true that my definition of what constitutes an optimistic ending, never mind an interesting story, has changed in the years since I completed A Conflict of Orders. Which is not to say that A Want of Reason will be a domestic novel – I’m not going to do a Tehanu (much as I would love to)…

But as A Want of Reason begins to take shape and settle into its story, I’m finding it a much darker novel than I had expected. The focus of the story too has altered, and now rests on a different selection of characters. Casimir Ormuz, the peasant hero, is still there, of course. But his journey to the resolution – never mind the resolution itself – is very different to the one I had originally envisaged.

I wrote each novella (and novel) of the Apollo Quartet to confound reader expectations. I see now that I’d been working to a similar principle – albeit considerably weaker – when I’d written A Prospect of War and A Conflict of Orders. But for A Want of Reason… I’m going all-out. The good guys will become bad guys, and the bad good, and the ending will neither reinforce the status quo nor raze the empire to the ground.

There’s not much room for innovation in space opera, given that everyone judges the subgenre by its bells and whistles. It’s either the world-building or – and this is a development of the past few years – its gingerbread prose which seeks to disguise common tropes beneath obfuscatory metaphors. The story templates haven’t changed, the tropes certainly haven’t changed. (There’s probably a Tough Guide to Space Opera, er, Space post somewhere in all this.) And those few space operas which have rung changes have generally caused very few waves. Has there, for example, been anything comparable to Nova published in the twenty-first century? (Having said that, are there any space opera authors as fiercely intelligent as Samuel R Delany currently being published?) There’s Ann Leckie’s trilogy of Ancillary Justice, Ancillary Sword and Ancillary Mercy, of course, which used an astonishing piece of sleight of hand in using female as a default personal gender to add a fresh new flavour to something Iain M Banks had been doing for three decades. And while Banks was certainly more innovative than pretty much every other writer of space opera – a consequence, I suspect, of having one foot in the literary fiction camp – even then he had a tendency to use tropes as they were set up rather than subvert or re-engineer them.

Sadly, Banks is gone and I suspect Leckie’s trilogy will prove a one-off blip. Space opera was already busy retrenching after the exciting times of the British New Space Opera of the eighties and nineties – not just Banks, but Take Back Plenty, Eternal Light, Light… But that movement introduced more of a hard sf sensibility to space opera (and some of the names attached to it, including McAuley and Reynolds, are probably better considered hard sf writers), without substantially changing its story patterns or its commonest tropes.

I’ve said before that space opera – if not science fiction itself – is an inherently right-wing genre (even if not all of its practitioners are right-wing). But more than that, I think space opera is inevitably drawn to the right. If someone writes a space opera which isn’t right-wing, it soon veers back to that side of the political spectrum. In part, it’s a function of the political systems which usually appear in space opera: emperors and empresses and empires and bloody great huge space navies. (I don’t, incidentally, hold with the argument that it’s the supposed tyranny of the laws of physics which lends science fiction, especially hard sf, its right-wing character.) However, I do think that science fiction has now, more than ever, reached a position where much of what qualifies as sf is little more than the rote deployment of sf tropes. There’s no insight, no consideration, attached. Put FTL into a story and no one so much as blinks. It’s just part of the furniture. Flat-pack science fiction.

And if you’re going to claim FTL is okay, it’s plausible, because there might be a Kuhnian paradigm shift which means it could happen… Which is, er, not my point at all. The tropes exist, they’re the building blocks of both space opera and science fiction. But I don’t think they should be used uncritically. I’d like to think I haven’t used them myself uncritically. Admittedly, a commercial space opera is likely not the best vehicle to deconstruct space opera tropes (but then I’d have said an commercial fantasy trilogy might not be the best place to deconstruct epic fantasy tropes, but Delany went and wrote his Nevèrÿon novels; but then, Delany…).

My area of interest in writing lies chiefly in the shape of stories, the narrative structures used to present a story in a particular way. I’m not interested in immersion – or rather, no more so than I need for a story’s world to be rigorous in my own mind. I’m not interested in literary techniques designed to make one reader response more likely than others… I jokingly mentioned in a recent conversation that I’d set a story on an exoplanet orbiting Gliese 876 but moved the setting to 61 Virginis because I didn’t think it plausible the story could have taken place given the original star’s distance, and likely travel times, from Earth. This is a science fiction story, of course, which posits a human civilisation across several star systems. No one would have noticed, but it was important to me.

If a science fiction story creates its own world , its version of Mars, Dubai, the Atlantic Ocean, etc, that doesn’t to me mean it does not demand the same level of rigour which pertains in the real world, in mimetic fiction. And at those points where the science fiction touches the real world… then the rigour applies just as much. This was a defining philosophy of the Apollo Quartet. The An Age of Discord space opera trilogy, however, does not touch the real world – at least not to any degree which might affect its setting. But its universe still needs to be internally rigorous. This may be why I find narrative structures and story templates preferable to be experimented upon – because they do not jeopardise rigour. (Yes, yes, you can make a point of ignoring rigour – surrealism, if you will – but that’s a different discussion.)

And so, in a more roundabout way than I am typical guilty of, it’s back to A Want of Reason and my total inability to wrap up what is supposed to be a commercial space opera trilogy in a nicely commercial way. The final chapter of A Conflict of Orders gives a flavour of the third and final book, and it wasn’t until I came back to that chapter a few months ago that I realised exactly what I’d set myself up for. Empress Flavia is on the Imperial Throne – and she’s kicked off a crack-down. When I first wrote it, it probably meant something in terms of my original plan for A Want of Reason. Now, it means: space opera fascism! And that’s what you’ll be getting: a space opera setting that moves ponderously to the right, in order to set up a climax that shifts everything irrevocably to the left. And, meanwhile, your favourite characters? I’ve either dialled them back so far in the narrative they no longer have any agency, or I’ve got them doing stuff villains normally do.

Because. Space opera.