On Writing: The Danger of Metawriting, OR, How I Wrote A Book That Did Exactly What I Wanted And Screwed It Up Anyway


The Arenaverse novels – Grand Central Arena, Spheres of Influence, and now Challenges of the Deeps – are inarguably my most successful solo novels. Grand Central Arena itself has continued to bring in significant amounts of money for me even now, seven years after release.

I'm also very proud of the Arenaverse itself; developing it was, and continues to be, a huge challenge. One of the Arenaverse's key aspects from a writer's and reader's point of view, of course, is that it is not merely a story but a salute to, and occasional commentary on, the Golden Age (and later) speculative fiction that it draws from. This "meta-level" of the writing is both overt and covert in multiple aspects of the books, ranging from the direct references/commentary inherent in the existence of the Hyperion project to those found in the genre conventions that are used, abused, and sometimes broken and rebuilt.

When this kind of thing is done well, it can be a lot of fun, both for the author and the reader, as they get to enjoy something that manages to partake of both the old and familiar and the new and surprising in ways that a purer, more independent-of-metalevel novel can't manage. However, there's some major challenges in this kind of thing which can, as they say, come back to bite you on the nether regions.

I'm going to describe how this happened to me with Challenges of the Deeps.

Originally, the book that was to be Challenges was supposed to be focused on, really, one thing: the journey of Ariane and her friends into the Deeps of the Arena with Orphan. But as I was beginning work on Challenges, I learned that (barring a miracle with one of my remaining contracted works) Baen would not be taking any more works from me.

What that meant was that I had to try to find a way to make Challenges a … well, if not completely satisfactory conclusion, at least an acceptable stopping point for the series, where it could sit for a while without being completely frustrating for readers. That meant that I had to try to resolve a number of important issues all at once, some of which, like the Molothos War, I'd intended to be multi-book ongoing problems. But that particular Chekhov's Gun had been sitting on the mantelpiece since GCA, so I couldn't just leave it there; the same was true of Ariane's powers, the mystery of Hyperion, and a few other things. I discussed some of those issues in my Big Idea article on John Scalzi's site, here: http://whatever.scalzi.com/2017/03/16/the-big-idea-ryk-e-spoor-7/

As can be seen from that article, I was pretty satisfied with the solution I'd come up with. Naturally, no book ever gets perfect reviews, and I in fact expect some negative reviews (and worry a bit if I don't see any, because I know that when *I* see a book with just "this book is awesome" comments, I am dubious about the value of the rating).

But for Challenges, I noticed a pattern in the negative reviews – from people who had liked the first two books, clearly had anticipated the release of the third, and felt that I'd somehow dropped the ball, "phoned it in" in the words of at least one review. Seeing the same kind of comments – positive or negative – on a book is a sign that there really is something there, something that caused this reaction. In this case, I had obviously done something – or failed to do something – that seriously ruined the experience for people who had clearly enjoyed the first two.

What was it?

(SPOILERS for the Arenaverse, and Challenges of the Deeps in particular, will be seen in the following section; if you don't want to be spoiled, you probably shouldn't read farther until you've finished the book!)



Some of the objections had to do with "superpowers" and characters not being challenged. This did, however, puzzle me, as I had clearly been foreshadowing both Ariane's power (with the Shadeweavers and Faith as partial templates) and the possibility of the Hyperions having something beyond the ordinary pretty much all along. Multiple significant scenes, including Wu Kung's full power being awakened, DuQuesne's use of psionic as well as physical power, the great race between Wu and Tunuvun, and so on, were ones I had known I was going to write for years, many of them from before I finished Grand Central Arena itself in 2009.

I'd worked hard, in fact, to insert the foreshadowing of these events and capabilities of our characters. And certainly I didn't think anything in there showed Ariane, DuQuesne, Wu, or Simon having an easy time of it ("not being challenged"). And obviously for many people it worked – they saw what I was doing, and it made sense, and if it was a bit more compressed in time than I might have liked, well, at least that meant Challenges really was jam-packed with action.

So … where had I gone wrong?

After carefully looking over the books, and the reactions, I came to the conclusion that – as I might have guessed – the problem came down to Marc C. DuQuesne.

DuQuesne is of course the most meta element in the entire series. He is, in name, physical appearance, and general conversational style taken (with the blessing, thankfully, of the Smith heirs) from the character of the same name in E. E. "Doc" Smith's Skylark series. In-universe, he was also designed after the works of Doc Smith, as a product of the ill-fated Hyperion project – a "DuQuesne" who combined aspects of that villain and the heroes of both Skylark and Doc's most important series, Lensman.

While writing with DuQuesne, I tried to insert some information about that fictional background – the background that made him who he is, and that referred to Doc Smith's work (as a salute to the man who had most inspired me as a writer), so that the readers could get some idea of what he was supposed to be like. But at the same time, I deliberately tried to avoid getting into too many details; I didn't want to step TOO hard on the toes of the Smith estate, who were at the time I wrote Grand Central Arena also negotiating for a possible movie production of Lensman. It wasn't necessary, I felt, to go into the details of what a Lensman was, what their powers were, and similarly to reference all the things that DuQuesne and his opposite number Seaton encountered or learned to do in Skylark.

I think I was right to do so; being given a non-binding blessing is nice, but doesn't give me the right to trample too blatantly through the work of my inspiration.

But that was also my mistake, and I couldn't see it for a couple of reasons. First was that – of course – I was far too close to the problem. I wasn't going to be able to see clearly what vital information was missing from the point of view of someone who wasn't a Smith fan (or who might not have read them for four decades).

Second was that I normally depend on my beta-readers to catch and backstop me on fumbles like this. But during the process of writing Grand Central Arena there had of course been many discussions on the background – involving the beta readers. Most if not all of them had also read and even commented on the seven or eight chapters I have written of Hyperion Origin, which is DuQuesne's story. This meant that they, too, had internalized knowledge of the character – of this specific version of DuQuesne – which simply wasn't available to the casual reader.

And there, I think, was the crux of the problem. DuQuesne does several key things in Challenges of the Deeps – builds a super-science communication gadget as well as duplicating Simon's improvised weapon from Spheres of Influence, demonstrates telepathic capabilities, and later in combat with Vindatri shows immense psionic powers that allow him to keep up with Ariane and Wu. These are actually completely in line with what one might expect from a protagonist from a combined Skylark-Lensverse, and in description I sprinkled references to those powers (such as the rainbow sparkling near DuQuesne's wrist – I wasn't going to explicitly describe a Smithian Lens, but hint at it, yes).

But unlike Ariane, who had gained her powers from something we already knew in-universe and was even being explicitly trained, there was no clear precedent for DuQuesne's abilities except for those self-referential, and pretty much metaknowledge-based, hints. Similarly, Sun Wu Kung is a literal figure of legend, whose powers – in any of his fictional versions – are fairly straightforward and, as a god or battler of gods, expected to be pretty damn awesome.

The reader could be expected to anticipate those characters becoming something superhuman, especially with what happened with Wu and Ariane in Spheres, but there was – looked at from the point of view of someone who didn't have all the metaknowledge – a lot more of a big fat question mark on top of Marc C. DuQuesne, and thus a lot more of an issue with him just yanking out a big can of whupass without a lot more detailed buildup.

I suspect if I had been able to write this set of events as I originally planned it, there would have been less of a problem. I would have had more time to subtly but repeatedly bring up hints about DuQuesne's origin, the powers he'd had in his Hyperion world, and so on, enough so that once it became obvious that the Hyperions were being granted powers equivalent to those they had in their "native" universes the readers would have a decent idea of what to expect from DuQuesne.

But without that, and with the accelerated pace of events making all three of those characters (plus, elsewhere, Simon) go from zero-to-sixty in only one book, the lack of clear expectations for DuQuesne affected EVERYTHING around him, including the perception of the capabilities of his allies. If the reader feels that the author's not playing fair with them on any significant point, that feeling can affect their experience of the entire book. Thus, not only did they feel that DuQuesne was getting deus-ex-machina superpowers to address any problem needed, that sense of suspicion pervaded the other key elements of that confrontation and the one following, so that the most climactic moments of the book were diminished by that single problem.

That is what I think happened with that group of readers, and it is, truly, entirely my fault for failing to see what I had left out.


To those readers – my apologies.



  1. skyrocket says:

    Hi, You make a good point about DuQuesne, but I think you are too hard on your self. You’ve always hinted that DuQuesne has superpowers and hinted at where they originate. As an Engineer I enjoyed the times he has used his powers to make new things. (The heart of engineering). We still do not know a lot about DuQuesne full powers. I assume you will develop them when he has a suitable villain to oppose. We have watched the Monkey King unleashed and I liked the a number of the scenes. The one of him flying back to the ship with his arena friends was very good. I am looking forward to seeing DuQuesne use his full powers to make a similar stunning come back from defeat.

    In the last book you used one of my favorite themes. (Pick one. The good guy that is rotten, or Bad guy that is good). For instance, the Psi cop Bester on Babylon 5 is my favorite bad guy. There is one episode “Dust” where Bester is a good cop trying to control a drug problem on Babylon. He makes a very good impression doing all the right things to help control the outbreak. At the very end you find that dust was invented by the Psi corp to enhance Psi powers. Bester isn’t interested in helping mundanes control a drug problem. He just wants to keep dust for Psis. With one comment at the very end you see a pleasant smiling friendly face rotten to the core.

    By the way I am one of your beta readers. I lurk, but almost never comment. I have very little expertise on writing. (I can not help with that). The things I can help with (Engineering) ring true. So I have had nothing to say so far.

Your comments or questions welcomed!