KEVIN CROSSLEY-HOLLAND

When I was a boy, I took over the shed at the bottom of the garden and displayed fossils and potsherds and coins in it and proudly called it my 'museum'. I charged people to come in, and my most prized possession was a Saracen shield dating from the Crusades.

Maybe if I ever come to write about my teens and adulthood - and I can't imagine I will - but if I do, then maybe I will want to say a bit more about the ways in which my parents' relationship with one another impacted on me in later years.

I see the role of the writer as creating a room with big windows and leaving the reader to imagine. It's a meeting on the page.

There are Arthurian legends in 14 or 15 medieval European languages. They are the product of no one time or place. On the contrary, in sum they represent a tremendous mine of human understanding, rather as the Bible does.

Before Arthur, I'd dismissed altogether writing fiction. You only have so many semi-sharp arrows in your quiver, I'd told myself, and I was not going to be able to write a novel.

Most writers, by the time they're 60, must have revisited their childhood a dozen times.

Think of 'Alice's Adventures in Wonderland.' It is equally intoxicating for children and adults. All this 'crossover' talk is something publishers are using as a selling device - a kind of post hoc rationalisation of what was happening already.

I am seriously interested in the psychology of childhood. And I've given a lot of my life to trying to see questions of personal development, as well as the great issues of the day, from a child's point of view.

For each detail I include, I throw dozens away. So I guess the first trick is to pick the right details, the most revealing details. Then I think one must simply write quick, clean, bright prose. For me, this means rewriting and rewriting: almost never adding, almost always cutting.

Like any decent researcher, I throw away 90% of my research.

Think of 'Alice's Adventures in Wonderland.' It is equally intoxicating for children and adults. All this 'crossover' talk is something publishers are using as a selling device - a kind of post hoc rationalisation of what was happening already.

Before Arthur, I'd dismissed altogether writing fiction. You only have so many semi-sharp arrows in your quiver, I'd told myself, and I was not going to be able to write a novel.

I see the role of the writer as creating a room with big windows and leaving the reader to imagine. It's a meeting on the page.

There are Arthurian legends in 14 or 15 medieval European languages. They are the product of no one time or place. On the contrary, in sum they represent a tremendous mine of human understanding, rather as the Bible does.

Maybe if I ever come to write about my teens and adulthood - and I can't imagine I will - but if I do, then maybe I will want to say a bit more about the ways in which my parents' relationship with one another impacted on me in later years.

For each detail I include, I throw dozens away. So I guess the first trick is to pick the right details, the most revealing details. Then I think one must simply write quick, clean, bright prose. For me, this means rewriting and rewriting: almost never adding, almost always cutting.

Most writers, by the time they're 60, must have revisited their childhood a dozen times.

When I was a boy, I took over the shed at the bottom of the garden and displayed fossils and potsherds and coins in it and proudly called it my 'museum'. I charged people to come in, and my most prized possession was a Saracen shield dating from the Crusades.

Like any decent researcher, I throw away 90% of my research.

I am seriously interested in the psychology of childhood. And I've given a lot of my life to trying to see questions of personal development, as well as the great issues of the day, from a child's point of view.