At the end of February, I submitted my line edits for The Animals We Became, my next book — a queertrans retelling of the Fourth Branch of the Mabinogi (the story of Blodeuwedd) exploring gender, trauma, and compulsory heterosexuality through assorted nonconsensual animal transformations and all the other fascinating horrors of medieval Welsh literature.
Line edits are never a fast or easy process for me, but these took me five and a half months, which is considerably longer than I’ve ever spent on a similar edit — with my YA books, I don’t think I was ever given more than six weeks to turn around an edit, and sometimes it was closer to four. In hindsight, I don’t know how I ever managed that, save that I would put my entire life on hold while working on it; part of the problem this time was that due to health issues and academic deadlines, there were large chunks of time during that period when I was unable to work on my edits at all. When I did finally come to rewrite the book, it took me less than a month to do so… but it was getting there that was the hard part, as we’ll see.
I have an unconventional approach to editing, perhaps, and tend to tackle structural edits by opening a new document next to the old one and writing the book again, because to work within the same document feels stifling. I find myself too limited by what’s already there, unwilling to make radical changes, and the result is always worse. When I write in a new document, whole chunks of text will survive, but they’ll be retyped word by word, because if it’s not worth retyping then it’s not worth keeping. I’m sometimes quite aggressive in my edits, cutting whole chunks of the book and reworking major plot points, but eventually, I’ll wrangle things into shape. It’s labour intensive, but it works.
Still, by the time I get to line edits, I’ve usually set aside the separate document and embraced the world of tracked changes. My usual approach is to read the book aloud, checking for unintentional rhyme or repetition, refining the rhythm and style of each sentence, making sure that syllables fall where I want them to and alliteration is deployed where most effective or avoided where awkward. Line edits are a time for nitpicking and accuracy, so elements may get rewritten or reworked, but these are usually sentences rather than paragraphs, details rather than scenes.
I began like this in the autumn, but something was wrong. I could tweak and polish each sentence all I liked, but I wasn’t happy with The Animals We Became, and I couldn’t pin down why. It was a strange time to be working on it, with reviews coming in for The Wolf and His King before and after the UK publication in November, and I found myself experiencing a kind of Second Book Dread that I’d never experienced before. Although this will be my fifth published novel, it’s my first time having to write something that is both similar enough to a previous book to have some overlap of audience while also being very different in tone and not a direct follow-up, and the pressure of expectations and the knowledge that I would inevitably disappoint some of them started to get to me.
It was the positive reviews for The Wolf and His King that were troubling me the most, because many of the things that people seemed to like were things that were not in The Animals We Became. Some were never intended to be — this was never a romantic book, and I’d resigned myself to disappointing those who thought I was going to write romance from now on — but others nagged at me, because I didn’t quite like that they weren’t there, either. The depth of academic research and historical detail was far less visible in this book than in TWAHK, for example, and while that didn’t mean it hadn’t been part of my writing process, I began to feel increasingly less confident in the choices I’d made that were rendering it so much less obvious.
What I was missing, I realised eventually, was constraints to work within. Rules to push back against.
My books thrive on argument, little as I enjoy it at the time: I have problems with authority, and I need something to rebel against, a constraint to struggle within. With my YA books, these constraints were usually editorial: tight word limits, things I was told I couldn’t do in YA, requests to clarify more while somehow not adding any words, etc. My editors would give me a series of seemingly impossible and incompatible instructions and I would argue with the ones I disagreed with and then do what I could to reconcile the compromises with the text. Very often, suggestions I received weren’t directly actionable, but they revealed a problem that did need fixing somehow, and the compromise on how to do that would lead to a better result. The struggle to enact those changes within the word limits and while conforming to the expectations of category and genre would often produce scenes and details that were better than the version I created on my own.
With The Wolf and His King, the constraints were more self-imposed. One of the main constraints I imposed on myself was in the use of language, and my attempt to avoid anachronistic concepts in an attempt to immerse myself more fully in my characters’ worldviews and perspectives. In general, the historical setting provided most of the rules I found myself pushing back against: how could I make this work in the setting? What did this mean in the setting? What were the implications of this action, how would this emotion be expressed, what would characters fear or desire about this in the setting?
The rules weren’t so tight as to become stifling, because I knew that I could, at any point, escape. I’d promised myself that anachronisms were permissible for narrative reasons, and my author’s note at the beginning gave me a Get Out Of Jail Free card with regard to using them. If I did so, though, it would need to be with intention — I would need to genuinely think it was worthwhile, that it made the story better, rather than simply being easier. And I didn’t want to resort to them, because avoiding them was more useful.
A sidenote: my author’s notes always claim less fidelity to the rules than I actually aimed for while writing. I am actually much more precise and nitpicky than I pretend to be, but if I tell you that, you might nitpick me right back. But it’s also that the nitpicking at my end is, ultimately, a means to an end, not the end in itself. It’s not about whether the book is Historically Accurate For The Sake Of Being Historically Accurate; it’s about whether striving for historical accuracy helped me enter more fully into my characters’ minds and the world they lived in, in order to create something that rang true.
The real value of those constraints, I realised, was that they cut me off from the path of least resistance. They closed the door to cliché and to generic choices. They forced me to act with intention, rather than out of convention; they made me interrogate what I was doing, and decide what I valued about it. Arguing with editors or myself about whether to keep a detail that didn’t follow the Rules helped me realise if that detail mattered — or if there was one I could come up with that would satisfy me and follow the rules, potentially forcing me to do something more original and more creative in order to make it work.
These were not constraints that could be imposed at the start of writing. They worked because they happened at the editorial stage — and more, because they happened late in the editorial process, when tearing it all up and starting again from scratch wasn’t an option. The struggle to improve a scene became the effort to reconcile what was on the page with the rules I wanted to follow, and it was in the reconciliation that I would usually find a better version of the scene, one that felt fresher and more interesting than the more generic option that preceded it.
And the trouble with The Animals We Became was that I didn’t have any constraints.
Editorially, the book had had far less intervention than anything else I’d written. I drafted it in February 2023 and it was this first draft that my editor read, with no earlier beta readers, feedback, or redrafts before it entered the publishing process. Structural edits were reasonably light, at which point I sought out some beta readers, because I realised I needed more perspectives to wrestle with, more questions to answer, more misinterpretations to help me realise when I’d been unclear. I needed readers who didn’t know the story it was based on, too, so I could tell whether it made sense without context.
My first attempt at line edits, following this beta reader feedback, resulted in more than twenty thousand words being added to the book, so I asked for a second round. This I got, but I fear my editor likes my writing too much: she left 83 revisions on the document, promising me that it only needed some fine tuning. But something was wrong, and the more I tried to fine-tune it, the more I thought I might have strung it with completely the wrong strings.
The problem, I realised, was that there were no rules. I had added thirty thousand words to the book since the first draft, and it hadn’t been a problem; I wasn’t frantically trying to cut them. My editor’s 83 revisions did not provide a great deal of significant resistance to grapple with. I hadn’t set myself linguistic restrictions — I had deliberately chosen not to — so I didn’t have the same vocabulary constraints. And I’d set the story in a semi fantasy world (geographically Wales, chronologically non-specific, as in the original tale, and culturally fictional) which meant there were no rules there either: I didn’t have to worry about historical accuracy. I even had magic to help me escape any overly difficult questions about technology.
I could do whatever I wanted. And I had. But without anything to fight against, without resistance, with no specific paths along which to channel my creativity, it had began to slip along the well-worn tracks of easy answers.
Left to myself, I had created something generic. Not entirely generic, nor entirely without value, but something that lapsed more often than I liked into conventional patterns and ideas. Everything from architecture to household structure was just a little bit… uninterrogated, owing as much to modern pop culture ideas of The Past to any genuine world. And it was not, to my mind, anywhere near as Welsh as I’d wanted it to be, despite having known since I set out that this was a priority. (I’m not taking eight hours a week of Welsh lessons for no reason!)
I’ve been haunted throughout the writing process by the thought of my academic colleagues reading this book: several of them have been kind enough to read The Wolf and His King, but, well, if I’m honest with you, my main anxiety is that there’s considerably more sex in The Animals We Became and I have to go to conferences with these people. 💀 I realised, though, that my prudishness was fading beneath a greater and far more important concern about how this book would read as an interpretation of the text, and as a way of engaging with medieval Welsh literature, and as the latest instalment of English-language textual reception of the Mabinogi, etc.
And, okay, you can’t think about this too much. You’ll go mad. A novel is not an academic thesis; my narrative interpretations are not necessarily my academic ones; what makes a good modern fantasy novel is not necessarily what I think a medieval author intended; the themes I am using the Fourth Branch to explore are not necessarily ones I think are present in the same way in the original text, and so on. While I do find fiction a valuable way of exploring and understanding medieval literature, I am ultimately writing my story using an older narrative, not analysing the one that exists. Moreover, this is a queertrans book about gender and trauma and there are going to be those who are hostile to that because the field of Celtic Studies can, at times, be fairly conservative; these are not people it is worth trying to please.
But. It wasn’t that I wanted to please them. It was just that I wanted to be able to stand behind my own work. To say, yes, I did that on purpose. Yes, this was an interpretative choice I made. Yes, I had to decide how to handle that and this is what I decided to do. I found myself reaching for author’s notes, justifications, lists of Further Reading, as though this might prove I’d done my research. It was, I realised, insecurity, because I didn’t feel those things were visible in the text, and I wanted to prove I knew them anyway.
By the time I’d completed my first pass of these line edits, reading aloud and tweaking individual sentences — which took much longer than planned — I’d realised that I was academically unhappy with the setting and the choices I’d made. I didn’t want an author’s note to be the only place my expertise was visible! And I especially didn’t want the setting to be Generic Fantasy World Superimposed On North Wales — I’d never wanted it to be that, but I’d somehow failed to avoid it, because when I decided against certain constraints (a particular century in the real historical past, for example), I hadn’t imposed enough others in their place.
It was a perfectly good fantasy novel, I guess, but it wasn’t the story I was trying to tell, because it wasn’t engaging with the Fourth Branch on the level that I wanted it to.
Too late, I realised the constraints that were missing, the rules I should have been following, the research I needed to incorporate. Too late, I could see what the book needed to rescue it from sliding into a pit of Uninterrogated Genre Conventions. Perhaps, if the book had had years to stew (as I always prefer), I would have figured this out before I got this stage; perhaps if I’d had more beta readers; perhaps if I hadn’t been suffering from migraine-related brain fog that left me unable to maintain complex thought for months on end… But the circumstances were what they were, and there I was, with a book I suddenly knew how to fix but didn’t have time to.
In my second pass, I rewrote certain scenes and chapters, frantically burying myself in research materials. But it needed more than that, and having finally glimpsed what the book needed to be a decent retelling of the Fourth Branch, rather than merely a decent fantasy novel, I wasn’t prepared to let it go. So despite the knock-on effect this would have on academic deadlines, I found myself asking for an extension. My editor, though she reassured me that she thought the book was in good shape already, granted me the extra time, and then I rewrote the book properly, this time with a much better understanding of my own personal rules for how the narrative and its world worked.
When I finally gave it back at the end of February, there were more like 8,300 revisions in the document than 83, and it crashed my laptop every time you tried to interact with one of those tracked changes. (I’d rewritten it in a separate document, and used “compared changes” to produce a version tracking them — it was the only way it was viable.)
It’s… better. I’m academically much happier with it. I think it’s doing something more interesting; I think it’s engaging more meaningfully with its setting; I think engaging with the setting is more than just interesting background noise but actively contributes to the main themes; I think pulling it apart like this enabled me to see certain other flaws and fix them. There are choices that some people will disagree with, but which I made because they suited my themes, or because they were interesting to me, or because there was simply no Right Answer to a particularly thorny question of interpretation and I chose this one as being the one I could live with. I think I can stand behind them now, because I have thought about why I did them, and I’ve done them on purpose.
It would, however, have been nice if I’d realised two years earlier that this was what I needed to do — or perhaps two months earlier.
Because, as I’m realising, it’s the process of trying to wrestle existing material into shape that often makes imposing rules on one’s world and narrative valuable, far more than trying to write something that fits within them in the first place. It’s the push and pull of having to make it work within an existing structure, when you can’t simply extend the scene indefinitely or completely alter the setup. It’s the way it forces you to question what you value enough to keep, and what can easily be overwritten with something new.
The book isn’t finished yet. We have copyedits yet to go, and I’ve asked that the copyeditor errs on the side of too many comments rather than too few, because I’ve realised I need to be put in a position of defending my narrative choices if I’m going to stand by them. And while I am quite good at challenging myself — I hold myself to high standards, and always have done — I almost failed this time. Not because I couldn’t see that there was a problem, but because both figuring out the problem and how to fix it turned out to be almost too much for me, and I didn’t know what rules I was trying to follow until it was nearly too late.
Too few rules, it turns out, makes it too easy to take the path of least resistance and write the less interesting version of your book. You need something to wrestle with to force creativity, whether that’s a tight word limit or an editorial argument or a strict conformity to historical accuracy or never using the letter E or making every paragraph start with a certain letter of the alphabet or a rhyme scheme or metrical incorporating a particular series of medieval laws or whatever set of constraints you ultimately impose. Yes, in some ways, it’s about making life harder for yourself — but that’s because cutting off the “easy” options is what cuts off clichés and generic choices.
Perhaps, now that I’ve articulated this to myself, it will be easier in future to realise that I need to identify my constraints earlier in the line edits process. But perhaps they’re something that can only be determined through wrestling with the book in the first place, trying to understand why it isn’t clicking.
I don’t know. All I know is that eventually, I hope, I got there. It has been a difficult five and a half months, and I’m quite burned out, for all that I feel I have little to show for it, but while some days produced no words or visible progress, maybe they too were part of the process of wrestling with the book.
Or maybe I was simply wrestling with my brainfog, and should cut myself some slack, because it’s hard enough writing books and doing a PhD while chronically ill as it is.
The Animals We Became will be published by Gollancz on 26 November 2026, and you can pre-order it now (more links to follow as they filter through to other retailers).