← Back to Blog

How to write historical fiction

BY Jack Hadley
3rd Jan 2018

Want to know how to write historical fiction? Or perhaps you're writing something based around a historical figure? If so, questions about how much or how little a writer should rely on research, and how much they should invent, come up regularly. History offers compelling and dramatic stories ready made for use in fiction, along with a cast of larger-than-life characters. But it can pose problems too. Bestselling historical novelist Janet Ellis and tutors on our online writing courses Suzannah Dunn and Catherine Johnson share their advice.

‘When I first wrote a historical novel, it was such a relief to have the ending decided for me! Before that, I’d found endings difficult. Well, actually, I’d found plots in general difficult – but now, courtesy of the historical record, here they are, ready-made,' says Suzannah.

‘The downside – as you might imagine – is feeling roped into a sequence of events that (at least in some form) have to be depicted, have to be made to happen on the page. So, it can be enabling, but at the same time it’s constraining.’

Author and tutor on our children's writing courses Catherine Johnson agrees that research can sometimes intimidate: ‘I rely on street maps an awful lot, where my characters are going and the size and shape of the world is always vital. I have to convince myself of the reality of the world before I convince readers. But for me it can be unhelpful to get bogged down in research. I find it helps to find out what I need as I go along, obviously I need to check things and not be inaccurate, but otherwise I'll just end up reading too much.’

Janet Ellis, author of The Butcher’s Hook and a former student on our London writing courses, and her literary agent Gordon Wise, share their insights on the process behind researching for a novel, and their thoughts on the tricky question of dialogue in historical fiction.

Janet Ellis, author of The Butcher’s Hook
‘For me, research is like a mattress. Once you've got it right, it supports you. You can bounce on it, too. But be careful not to let it topple on top of you – mattresses are heavy.

‘I didn't have a pattern to doing my research. Sometimes I needed a specific fact – one of my characters was a butcher and I knew nothing about eighteenth-century butchery (or contemporary meat matters, come to that) – so I wrote to the Worshipful Company of Butchers and asked to use their library. But the really interesting thing about nosing around, I mean; investigating their archives, was that I didn't always find a specific answer to a certain question, but lots of other wonderful, vivid, unexpected things. Even if I never used those facts (and sometimes just in a throwaway remark from one of my characters), they were like flavour in a stew.

‘To be perfectly honest, I sometimes made things up. Or at least, if I wasn't entirely sure about accuracy, then I borrowed something similar about something else and kind of bent it to fit. I believe that if you're sufficiently immersed in what your people are doing and the time that they're doing it in, then you can create their world pretty well.

‘Which almost answers the second question. As a reader, I don't want reams of research regurgitated on the page. Just as in modern fiction, I don't want a fulsome description of what someone's shoes are made of or who designed the chair they're sitting on, so in historical fiction you can be very bogged down by an avalanche of facts which get in the way of the scene or story. But as the writer, you have to know what everything looks or sounds or smells like.

‘You need to feel the way the stone floor meets a soft shoe or what the food tastes of. You have to fully imagine the rooms your characters occupy or how they'd feel about death, say, or geography. You just don't need to tell your reader everything. I think it's like being in a room full of closed cupboards – you have to know what's inside but you don't need to open them to prove it.

‘You do need to provide enough information for your reader to trust you, though. If you get something badly wrong, or leave out something significant that your characters would have encountered, for example, then your reader will stop trusting you. Any novel is a trust exercise and you have to prove your reader is safe and won't stub their toe on a tricky detail as you hurl along hand in hand with them. How much you research you do is a matter of personal preference. And you'll definitely stumble on yet another wonderful nugget of information just as you wave goodbye to your final draft.

‘It's important that dialogue reflects its time. It's joyful to rediscover phrases that have fallen out of use or the way words were employed differently in other eras. Getting it right is a bit like singing – you want to hit the right notes, but you're free to interpret the tune. For beginning to hear those voices, I went to ephemera: contemporary notes, shopping lists and diaries were a mine of information about the vernacular and for the way people actually spoke when they weren't being formal. Reading playscripts of the period was very helpful, too, in terms of rhythm and tone. Everyone will tell you to say it aloud. Everyone is right.'

Gordon Wise, literary agent representing Janet Ellis (and Plague Landauthor SD Sykes
Someone who tackles research/setting/environment first, plot and characters second, is a real turn-off for me. I want to read about real people who think and breathe, whenever or wherever they live. Only put something in an historical setting if it has to be there – if that adds uniquely to the plot.

‘For SD Sykes [former Curtis Brown Creative student and acclaimed author of two historical novels], she wanted to explore a world where there was a thin veil between the living and the dead, and the upheaval of the apocalyptic 14th-century Black Death afforded that. Janet Ellis wanted to find a time when a woman’s lot was such that there was some rationale to her wayward heroine’s actions.

‘Don’t use lots of forsooths and egad, sirs. They’re just corny and you don’t know if anyone actually spoke like that. Dialogue should never get in the way of understanding things, and if it’s larded with cod historicisms it completely undermines what you are trying to convey. I’ll stop reading an historical on the first page if the tone is just trying too hard.

‘You want your characters to be talking to people now, so just get them talking. You’ll be more authentic if you let them sound like real people, rather than working up something that may never have been how they spoke at the time. But for vocabulary, if there was drama in the period in which you wrote, it’s worth researching that, as that’s likely to be the closest record to actual spoken word. However, remember that they wouldn’t have actually spoken in rhyming couplets or blank verse – that was just a stage convention.

‘Focus on story, plot and character first. Ermine capes and marble halls should come into play when needed to make a point, but don’t let your obsession with them lead the story or you’ll ruin it, and lose the reader.'