How much plot?

Agent Jonny Geller recently complained that The Independent's review of Jane Harris's new novel Gillespie and I contained too many spoilers. Having just  reviewed the book myself for the Financial Times, I was curious to see what he meant. On reading Carol Birch's review I thought Geller was being unfair, and challenged him to see if he could do any  better.

Reading a novel in advance of publication is like setting out on a long sea voyage in which you're not just ignorant of your destination, but even what sort of vessel you're travelling on. The captain may be a fraud and the unobtrusive fellow who brings your tea in the morning might turn out to be the one you should have kept an eye on all along. Certain tip-offs will help later passengers enjoy the trip more fully. Yet in alerting them to all the sights, you should never rob them of the thrill of surprise.

My rough rule of thumb is that the first quarter to a third of a novel generally constitutes the set up and is fair game. The reviewer can describe that set-up, and maybe drop a few broad hints as to the direction the action might go, but at some stage you have to say 'Now read on...' However, sometimes even an early twist is clearly designed to come as a shock. (In which case it'll probably be given away by the jacket blurb.)

When editing reviews, a literary editor all too often has not read the novel in question, so it can be difficult to ascertain how much is too much. I might send an entire review back with the terse note: 'Too plotty!' A phrase such as 'After his wife dies...' would have me trying to elicit where in the book this happened. If it fell foul of my golden rule, I'd amend it to something like 'After a family tragedy...' Sometimes even this is too much, if it's a 'lightning from a blue sky' development. Sensitivity to the writer's design is all.

It's amazing how wedded reviewers can be to revealing the plot: 'I can't discuss the book properly without revealing that she dies halfway through/the butler did it/it was all a dream.' Funnily enough, this defence does have some historical weight. Reading Virginia Woolf's early book reviews, I was surprised at how much of the plot she routinely gave away. Perhaps today we focus on plotting above all other stylistic issues; certainly her reviews were also full of insight about the craft of fiction. Most often, though, the 'this happens, then that happens, then finally...' school of reviewing is just lazy.

Gillespie and I does present specific problems for the critic. To use the maritime analogy above, it quickly becomes clear that deciding what sort of vessel you are on, be it comfortable cruise liner or leaky tub, is absolutely crucial. Harris's narrative is so deft at wrong-footing the reader that even to know at the outset what sort of story you're going to be told will take away from the purity of discovering/deciding that for yourself.

Sadly, Jonny Geller did not choose to rise to my playful challenge to write a review himself. At least he conceded that in this case it is a tricky undertaking. I recommend Harris's superbly slippery novel, but the less you know about it in advance, the better.