Not so long ago the historical novel was seen as a lifeless old dowager, an unintentionally comic figure dressed in old clothes and going through the same old motions in the dusty attic of the house of fiction. Then, round about 1970, something happened. In the past 30 years, there has been a renaissance of historical fiction, with novels such as Toni Morrison’s Beloved, Barry Unsworth’s Sacred Hunger, Charles Frazier’s Cold Mountain and Michael Ondaatje’s The English Patient (to mention the tip of a very large iceberg) raising the genre to new levels of sophistication and imaginative reach.

What these authors have realised is that “the recreation of the past” need not involve the donning of research and historical accuracy as a form of strait-jacket, stifling all imaginative freedom. That is the wrong way of looking at it. The historical record can be a gateway, a door into a different imaginative world. The past is, as L P Hartley wrote, another country. They do things differently there - which is why it is so fascinating to spend time with them. Liberated for a time from the self-impor tant tyranny of the present, from the babble of everything that poses as new, the writer and reader of historical fiction can delve more deeply into who we really might be.

That, at least, is the theory. But how does one actually go about writing an historical novel? From my own experience, as someone who has written novels set in the past, the present and the future, there are a number of do’s and don’ts.

Don’t use the words “doth”, “prithee” or “sirrah”. Unless, of course, you want to do so for the purposes of comedy or pastiche. (John Barth’s The Sot-Weed Factor is a rumbunctious 18th-century novel, written in the 1960s, while Peter Ackroyd has built a career out of literary pastiche.) These special cases aside, nothing has contributed more to giving historical fiction a bad name than pointless archaisms.

In my novel Blue Fruit (which, for strange reasons of its own, is set in the present) I tried to suggest an 18th-century prose style primarily through rhythm and quality of thought, rather than vocabulary. It creates, I hope, its own world, hovering somewhere between then and now. To get taken seriously by a modern audience, an historical novel must, somehow, acknowledge its own modernity. John Fowles’s influential The French Lieutenant’s Woman does so by bringing to the surface the bare bones of “research”. (It is interesting to see how Harold Pinter attempts to reproduce the same trick in the screen adaptation.) Every writer of historical fiction is striving for a difficult balance. You are writing a modern novel, for modern readers - but set in the past.

Enjoy your research. The process of discovering your historical context should not feel like donning a hairshirt. If the mechanisms of flintlock muskets bore you rigid, then ignore them. Find out what moves and excites you, and concentrate on that. (It goes without saying that unless you are fascinated by history, there is probably no point in trying to write an historical novel.)

Go to the places that you wish to describe. To be born in this country is a wonderful accident of birth for an historical novelist. You can’t step outside your door without tripping over some evidence of history. But beware of the “Heritage” prettificat ion of the past. The National Trust is your greatest friend, and your greatest enemy.

Don’t enjoy your research too much. There are few human activities more dangerously pleasurable than pottering around in libraries and secondhand bookshops. Research (a posh word for sitting in a chair reading a book) can easily become a form of degenerative disease. Remember that you are a writer, not an antiquarian. Unless you are content to become one of those ghost-like figures doomed to wander the corridors of the British Library for eternity, then the day cannot come soon enough for you to throw your research out of the window and begin to invent.

Plug the gaps as you go. It’s a good idea to mix research and writing. They can inspire each other. I am currently writing a novel set in the Forties which includes an important episode set in a prison. My general sense of that period comes from a mixture of sources whose eclecticism would defy conventional footnoting - books, old films, TV documentaries, conversations with elderly relatives. Having worked for two years as a teacher in a prison, I also have a general sense of what prisons are like. But what exactly were prisons like in the Forties? What did people wear? What were the routines? How long were their sentences, and for what crimes?

I put off finding out these things until I had reached that stage of the writing. By then, I knew exactly what kind of details I needed to make my particular story come alive. Reading accounts of prison life in the Forties suggested, in turn, new strands in the storyline. (My main discovery, incidentally, was how little prisons have changed compared with the rest of society since 1940 - or, for that matter, since 1840.)

Root your characters in the past. This is not to say that you must include “historical characters”. You don’t have to have walk-on parts for Christopher Marlowe or Napoleon. (The recreation of real events, as attempted in such “novelisations” of history as Gore Vidal’s Lincoln, is a distinct subgenre.) In fact, as a general rule, it’s probably better that you don’t. But at every turn you should ask yourself, “Could this story be set at any other place or time?” If the answer to this question is yes, then you are in danger of using history merely as a stagey backdrop.

Giuseppe Tomasi di Lampedusa’s masterpiece The Leopard is a classic portrait of a character who could only have lived at a particular historical moment. Sacred Hunger allows its plot to grow organically from the economics of the 18th-century slave trade.

Get things wrong. As a nervous undergraduate approaching finals, I worried aloud to my history tutor that I couldn’t keep all the dates in my head. They seemed to fly away like flocks of birds. He replied with an excellent piece of advice: “Don’t worry too much about the facts, they’ll only confuse you.” What matters is the idea, the mood, the feel. War and Peace, arguably the greatest historical novel, is notoriously littered with errors.

Don’t forget: there is no such thing as an historical novel, there are only novels. A few years ago I remember coming across, in the programme of events for a writer’s circle, an advertisement for a lecture on “How to Write a Novel about Smuggling Life”. There is something of the same absurdity in discussing how to write historical fiction. Historical novels are, above all, novels. The historical novelist, like any other novelist, must create a world; must create a language that embodies that world; must create people who change and are changed by that world. That is where it all starts. That is where the fun, and the agony, begins.

Adam Lively’s latest book is ‘Masks: Blackness, Race and the Imagination’, published by Chatto