Speak to most authors and they'll tell you the most difficult part of the strange, alchemical process that is novel-writing isn't working out an interesting plot. The problems arrive later, when fleshing out the characters that will inhabit the tales. Get the protagonist's voice wrong, and it doesn't matter if the scenario they find themselves in is intriguing, ground-breaking or emotionally resonant. Without believable characters, novels are nothing.
So it isn't particularly surprising that sometimes, authors take the somewhat safer option. They "borrow" characters from other writers' works - the more famous, the better - and place them in their own books. The most notable, and successful, example is Jean Rhys's 1966 novel Wide Sargasso Sea; essentially a prequel to Charlotte Bronte's Jane Eyre, we discover the previous, colonial life of Mrs Rochester, before she is shut away in the attic of Thornfield Hall, supposedly mad.
Wide Sargasso Sea works because Mrs Rochester is an intriguing character for whom imagining an interesting backstory is easy - and Rhys is a fantastic writer. But all too often, borrowing characters in this way is a dangerous game to play. The wrath of precious fans of the originals is easily incurred and the books will always be compared with their more illustrious predecessors. Still, it hasn't deterred Eric Rauchway. His new novel, Banana Republican, gives Tom Buchanan - the racist, snobbish, despicable excuse for a human being in F Scott Fitzgerald's The Great Gatsby - a second chance.
In a way, it's similar to the reimagining of Mrs Rochester; Rauchway is trying to rehabilitate or understand a much maligned character. So after Gatsby is dead, a rueful Buchanan finds himself in Nicaragua, where he is caught up in the archetypal Central American revolution. Banana Republican hasn't gone down well with the critics, however, not least because the titular Gatsby of the orginal is never even mentioned by Rauchway.
The New York Times called it a gimmick: "It's as if Rauchway wrote a generic farce about a long-forgotten revolution and then decided the book might get more attention if he recast the narrator as a refugee from The Great Gatsby," wrote Joe Queenan. So, is appropriating characters from other books a form of literary cheating? The author Jill Dawson, who has been twice nominated for the Orange Prize, taught creative writing for many years and is a board member of the Norwich New Writing Partnership - a highly regarded literature development agency that works with both readers and writers - doesn't think so.
"Not at all," she argues. "It's better to look at it as an homage or the continuation of a dialogue. The books I've loved are often the starting point for wanting to write fiction in the first place, so I can absolutely see the temptation and pleasure in exploring characters from other novels. "In fact, I've tinkered with Flaubert's Madame Bovary in some - unpublished - stories of mine, simply because I love that novel, have read it loads of times, and feel Emma Bovary still has resonance for me, 20 years after I first read about her. Also, of course, Emma is described as loving novels and wants to live like a character in one. So there would be layers inside layers, like Russian dolls, to explore."
Still, even Dawson admits the only truly successful example she can think of is Wide Sargasso Sea. So why do authors continue to use well-known characters? Is it a self-imposed challenge to carry on somebody else's iconic work, or just an easy way to make a quick buck? "Perhaps the temptation of doing it contains the challenge," says Dawson. "And that is, how can a new writer ever top what's been done brilliantly already? So, in those terms, as a creative endeavour, I do think it's fascinating."
Does that mean she would attempt it in the future? "Well, since I was 16 I've been puzzling over Thomas Hardy's Tess of the D'Urbervilles and wanting to write about her, although I haven't yet. You need to have an audacity, arrogance or confidence that I admire... but don't possess." Dawson is of course happy to extol the virtues of such an approach, perhaps because she works in a world where encouraging creativity is key. Sometimes, though, some of these character spin-offs are not-particularly well-hidden attempts at brand extensions for publishers. Scarlett: The Sequel to Margaret Mitchell's Gone With the Wind, Alexandra Ripley's authorised sequel, may have been a commercial success, craved by those who couldn't bear to imagine the story ended with "tomorrow's another day", but it wasn't a patch on the original.
Another sequel, Rhett Butler's People by Donald McCaig, is Gone With the Wind, but from the roguish anti-hero's eyes. This time, it was actually a good novel, but crossed a different creative line. As The New York Times said, "in reducing Rhett to a perplexed and worrying Everyman, McCaig reduces the power of Mitchell's original". So, can these updates, spin-offs, or acts of literary ventriloquism ever be as good as the original creation? For die-hard fans, that's unlikely. The author Jim Crace, like his peer Dawson, genuinely doesn't see why they shouldn't be, though. He's well-placed to judge as well. Perhaps his most famous novel, the Booker-shortlisted Quarantine - features a character from one of the most widely read books of all time: Jesus.
"Sometimes these works can actually be better than the originals," he argues, before once again citing Wide Sargasso Sea. "I love Foe by JM Coetzee, which takes place around the plot of Robinson Crusoe. For children, TH White's Mistress Masham's Repose is a fantastic sequel to Gulliver's Travels. And Robert Nye made a career out of writing resurrections of fictional characters: Falstaff, Faust, Merlin, Beowulf. These are all good books. Where's the problem?"
Perhaps, I suggest, the difficulty is that readers often feel authors are writing with somebody else's characters because they know they have a ready-made audience. That, well, they're being just a little lazy and unimaginative. "It's not lazy at all," says Crace. "It's no less lazy to set a book in a real location where you don't have to invent the streets or the culture. Historical novels are very popular, but you could say, by your rationale, that it's lazy to select a known historical setting rather than invent one of your own. For me, quite honestly, anything goes."
Overall, the short bibliography of work that actually includes other authors' characters and works - Wide Sargasso Sea, the well-loved Tom Stoppard play Rosencrantz and Guildenstern Are Dead, Margaret Atwood's The Penelopiad: The Myth of Penelope and Odysseus - would seem to suggest that such an approach is more miss than hit. So does Crace have any advice for anyone about to try? "Any book can succeed if it's well-written and relevant," he says.
True. Which is why, if American critics are anything to go by, Rauchway may have fallen at the first hurdle.