Many years ago, I had the opportunity to interview Geoffrey Household. I travelled up to his home in Oxfordshire, and had a very pleasant afternoon. He and his wife were charming hosts, there was plenty of tea and cake and biscuits, all went well until I started the interview. No matter what I asked, he answered only ‘Yes’ or ‘No’. He wouldn’t expand on any point. No matter how I tried to rephrase questions so that he would have to say something else, I still got only ‘Yes’ or ‘No’. To cap it all, the local station was just a halt, with no shelter at all; it was winter and I had over an hour to wait for the next train. That was not a happy experience. I managed to get an article out of it, but that is certainly not one of the review I’m likely to reprint here any time soon.

Then at the other end of the spectrum you get Adam Roberts. I was asked to review Jack Glass by Interzone, and at the same time do an interview with Adam. My usual practice, doing an interview by email, is to send off half a dozen broad introductory questions. Then, when I get those answers, I go back with further, more detailed questions to open it all up. With Adam, the answers I got to my initial questions practically filled the word length I’d been given for both the interview and the review together. There was no point going back with supplementary questions, I simply wouldn’t have space for them. And I wasn’t sure I’d have room to review the book.

In the end, I combined the review and the interview, an experiment that I don’t think was a great success but it was interesting to try it. Then I edited the whole thing, edited it again, cut and cut and cut. I tried to make most of the cuts in my own words, though inevitably some of what Adam said had to go. In the end, I was still a couple of hundred words over my word limit, but Interzone, bless ’em, managed to fit it in. The whole appeared under the suitably punning title of ‘Adambrating’ in Interzone 243, November-December 2012. What I’ve included here is the original interview, before the necessary cuts and without the review interpolated.

Why a crime story?

You have chosen to use some of the oldest and most basic of crime story formats, the locked room mystery, the standard whodunit. What is about these structures that appeals to you?

I’ll take both those questions together.

The things I like about crime as a genre aren’t necessarily the things  that make it such a big selling modern publishing phenomenon.  I’m almost entirely uninterested in the cod-psychological-verisimilitude of alcoholic coppers and grimy criminals in Edinburgh or Malmö.  What I prize is twofold: the ingenuity—it’s also one of the things that draws me so strongly to SF—and the disruption.  This after all is the way a crime functions: it is a disruption that must be addressed.  Disruption is important.  It’s how I see my job as a writer.  Indeed, it’s how I see the job of any artist.  You take established forms and structures, the set generic constraints and readerly expectations, and you fuck them up a little, creatively.  Hopefully you don’t fuck them up so severely that they become broken; but you need to tangle imaginatively with your medium, or else you’ll just be extruding more Generic Commodified Product.  SFF needs less of that, I think.  So, in other words, crime foregrounds the principle of generic disruption that moves me on as a writer.  There’s a consonance, there.

One of the biggest problems facing SF at the moment is just the crushing immensity of its backlist.  It’s a splendid thing, of course: a person could spend their whole lives (as I intend to) reading through the history of SF and not exhaust it.  But it also puts severe strain on the writer who wants to write new SF.  Put simply: you believe you’ve thought of a brilliant new idea for your novel?  Nah.  Somebody, somewhere, will have done it before.  Where can you go?  I suppose a couple of strategies present themselves.  Writer A might simply ignore the backlist, and produce endless retreads of Neuromancer or Tolkien or Harry Potter or Twilight or whatever it is, for readers who don’t know or don’t care that they’re consuming derivative drivel.  That doesn’t appeal to me, I must say. Writer B might go meta, to write about the backlist rather than trying to overleap it.  One of the things I like about Harrison’s Light novels is the way in their imagined universe all the myriad, conflicting conventions  of classic SF ‘work’.  But here’s Writer C: she refuses to be cowed, and tries to plough straight through.  The genre’s possibilities are not exhausted yet, she says.  We can still be new.

In Jack Glass my starting point was a desire to smash together, CERN-style, the supermassive particles of Golden Age SF and Golden Age crime-whodunnits.  I particularly wanted to work with some of the hoariest, most thoroughly excavated conventions of both genres, to see if I could do anything new with them.  On the SF side, I got interesting in a theory that many of the generic and ideological assumptions of Golden Age space opera were Kiplingesque.  The first kernel of this novel was a poem, oddly enough.  It is called ‘The Mary Anna’, and it tells a traditional Golden Age SF in the form of a pastiche Kipling’s narrative poem.  Form and content married unusually well, I thought, and I started writing more in that imagined solar system.

Then there’s the crime portion.  One of the reasons the puzzle whodunit went out of fashion had to do with precisely this question of the clogging, stifling backlist.  Literally thousands of these sorts of novels were published in the 1930s, 40s and 50s.  Saturation point was reached. There’s only a number of ways you can play the three card trick—there’s a murder—there are a dozen (or so) suspects—one of them is the murderer—your detective investigates and reveals it was the one you didn’t expect! This sort of novel is built to deliver a particular ‘ahh!’ moment near its end.  The identity of the murderer can’t be too obvious, or the reader will go ‘pah!’ in disgust.  But by the same token the identity of the murder can’t be too random or abstruse, or the reader will go ‘well how was I supposed to guess that?  That’s rubbish!’  The whodunit is aiming at a very particular reaction: an ‘I didn’t see that coming, but now that it’s pointed out to me I see that I ought to have expected it!’  It’s a tricky thing to fine-tune. With the bog-standard whodunit (twelve suspects, one is the murderer) it’s terribly easy to provoke a readerly ‘meh’.  Personally I’m particularly interested in the way games writers have played to circumvent simply picking one of the twelve.  I’m talking about the more meta ways these writers have explored ways of wrongfooting the reader.  So, Agatha Christie tried: ‘twelve suspects—and they’re all guilty!’ and ‘twelve suspects,–but the narrator is guilty!’ and ‘twelve suspects—but the detective dunnit!’.  Michael Innes And Then There Came Both Ice And Snow constructs an ingenious twist: a man is shot with a pistol, there are twelve suspects—but nobody murderered him (and neither did he commit suicide)!  Margery Allingham’s Police at the Funeral details a whole string of consecutive murders, and the murderer turns out to be the first murder victim.  You see the kind of thing I mean.  So I sat down and tried to work out if all these more interesting alternatives to the bog-standard whodunit format had been done before.  I came to conclusion that there was one form that hadn’t: the reader is told the identity of the murderer at the start, such that she reads through the whodunit and it still surprised by the revelation of the murderer’s identity at the end.  So I thought I’d try writing one of those.

I noticed a specific reference to Agatha Christie. Were there particular crime writers whose work shaped your book?

One of the reasons I am so addicted to the reading of books is that I grew up in a house in which my mother read all the time.  She always had a novel in their hand, and it was almost always a crime novel.  The writers she read, I read: Allingham, Innes, Ngaio Marsh, Josephine Tey, Reginald Hill. But mostly I read SF. I devoured SF, and a few Asimov titles aside SF and Whodunnits seem generically immiscible.  The critic Linda Hutcheons famously distinguished between ‘epistemological’ fiction, like crime novels, and ‘ontological’ fiction, like SF; she suggested the two modes were orthogonal to one another.  I’m not sure I agree, mind.

To answer your question: in the run-up to writing Jack Glass I did read a bunch of puzzle whodunnits, to get my eye in, as it were: mostly actual Golden Age writers, although I did read a couple of more recent ones (the late Gilbert Adair’s Mysterious Affair of Style is a rather neat locked-room mystery, and a cleverly pomo pastiche to boot).

But then, this is part and parcel of the way you explore sf structures in your novels. So let’s expand that question: are you deliberately taking a structuralist approach to genre? Or, to put it another way: do you consider your works to be about science fiction as much as they are examples of science fiction?

‘Structuralist’? That’s fighting talk …

The short answer is: yes, I do.  It seems to me as desirable as it is inevitable.  The best novels, whatever else they do, explore the extent to which they are iterations of other books, make explicit the narrative and discursive codes out of which it is made.  We all tell ourselves stories all the time; it’s how we all get through our days.  It’s better be self-aware about that.  But I don’t see this as a structuralist enterprise; rather the reverse—it’s about deconstructing (it’s a demodé term, I know—what can I say? I’m an old-fashioned dude) the conventions and assumptions of genre.

This is beginning to take a rather academic turn, so let me ask straight out: how much does your day job influence your approach to genre?

Plenty.  Not, I think, in the sense that I read a piece of abstruse literary theory and then turn on my laptop thinking ‘and now I shall write a science fiction novel adumbrating this notion …’  That would be a catastrophic way of proceeding.  But one aspect of my job is: to read a whole lot of books, outside genre as well as inside it.  Everything I read informs everything I write; that’s bound to be true of any writer.  But I suppose my job, in its ‘writing literary criticism’ and ‘teaching literature’ puts me in a frame of mind where I tend to approach questions of narrative, character, style and structure in a self-conscious or self-reflective way.

The other thing that stands out in your books is how political they are. This is true of much sf, of course, but in Jack Glass, for instance, was the political overtone a conscious part of the work from the very beginning?

Jack Glass is, I’d say, less consciously ‘political’ than my last two novels, By Light Alone and New Model Army.  NMA in particular is all about the politics.  I saw it described in reviews and so on as a war novel, but I don’t think it’s that at all, or it’s only incidentally about that.  It’s actually a novel that tries to think through questions pertaining to democracy.  Jack Glass has no such ambitions. Its main ‘political’ idea is that sexual orientation, race and gender, whilst they matter, matter much less than wealth/poverty.

While we’re on the subject of politics, it occurs to me that with the exception of Jack himself, the strongest and most interesting characters in the book are all women. Was that deliberate?

Well, the first third is all men: that claustrophobic, all-male prison environment thing.  After that the novel needed to open the windows and allow the reek of testosterone to dissipate.  Apart from that: well, it would be nice to get to a situation where the Bechdel test was so internalised in writers’ operating procedures as no longer to be a thing the individual writer has consciously to bring to bear on their practice.