How many characters in a novel?

When starting a new book, a vital question is: how many main characters will take part in the story? This is closely linked with the question of how many murder suspects to include, although the two issues are not quite identical. I often give this a lot of thought, because it is vital that readers care about the characters, even if they don't like some of them. My latest reflections on this topic have coincided with reading a book by Ruth Rendell which highlights the dilemmas that an author can face.

I mentioned the other day that I pay attention to and try to learn from critical, as well as positive, comments in reviews, as long as they seem sensible, and supportively meant. In relation to a couple of books I've written in the past, a question has arisen about whether the early part of the book was challenging because of the introduction of a significant number of people and plot strands. (On this topic, Kate Ellis and I agreed at Blackpool last week that our tendency in this direction had been influenced by those wonderful early series of Taggart by Glenn Chandler.)

In The Frozen Shroud, I tried to address this by reducing the number of potential suspects. This made me worry that the solution to the mystery might be too obvious, but I'm reassured by the reviews that are in to date. With my new book, though, I'm going to tackle the issue in a different way.

Ruth Rendell's The Saint Zita Society boasts an unusual way of helping the reader disentangle what was going on. Rather than have a 'cast of characters' in the manner familiar from some trarditional mysteries (those fine writers Ngaio Marsh and Christianna Brand were keen on this device), the publisher included a diagram of Hexam Place on the endpapers of the book. This showed the houses and listed their occupants. A neat concept though I have to say that, even so, I  found myself overwhelmed with the number of people who cropped up in the first fifty pages, not something I've ever experienced when reading with this gifted writer. Clearly, she too is continuing to experiment with her approach, and that is what keeps authors fresh..

Forgotten Book - The Murderer of Sleep

My intermittent campaign to revive interest in Milward Kennedy continues! Today's Forgotten Book is one he published in 1935, probably at about the height of his fame (such as it was) in the Golden Age. It's called The Murderer of Sleep and boasts, in typical Kennedy fashion, two maps of distinct crime scenes: one is a hotel, where a theft and murder occur before the main action starts, and the other is the sleepy village of...erm... Sleep. The Macbeth inspired title clearly amused Kennedy.

It's an entertaining book which maintains a sprightly pace more or less throughout. I say 'more or less' because arguably the revelation of the main culprit's identity is delayed just a little too long. To my mind, it becomes pretty clear who the police are looking for, and there were moments towards the end when they just needed to get on with it all a bit quicker. This is, however, a relatively minor quibble. I did enjoy reading the book a good deal, even though I'm not sure I agree with those usually severe critics Barzun and Taylor who describe it as Kennedy's 'masterpiece'.

This novel illustrates a difficulty that all writers of whodunits face - it's one I'm conscious of myself, certainly. The question is: how many suspects should one have in the story? Too many, and the result is clutter, confusion and inadequate characterisation. Too few, and the puzzle is too easy to solve. Getting the right balance is one of the keys to success. This is what Agatha Christie did so well, so often.

Kennedy, I feel, errs on the side of having too few suspects here, and he could have done a bit more to probe the culprit's motivation. I think his difficulty arose from the rather unusual structure that he adopted in telling the story. This makes it a pleasing and fairly original read - and these are the strengths that appealed to me. But it does make a constant shifting of suspicion between candidates for the murderer rather difficult. As so often, Kennedy's ambition outstripped his achievement. But what I like about his was that he was an ambitious writer. Keen to vary the formula.

I have struggled in vain to find out any more about Kennedy's life than can be found in the usual places. If anyone knows any more about this interesting man - whose day job was as an international diplomat - I'd be keen to hear it.