I began in 1976, with small abstract paintings that allowed me to do what I had never let myself do: put something down at random. And then, of course, I realized that it never can be random. It was all a way of opening a door for me. If I don't know what's coming - that is, if I have no hard-and-fast image, as I have with a photographic original - then arbitrary choice and chance play an important part.

My method is related to an attempt to do something that might be understood by today's world, or that could at least provide understanding. In other words, doing something I understand and that everyone understands. This natural desire for communication is also found in other domains, like reading and discourse, etc. I also hate repeating myself; it gives me no pleasure whatsoever. Once I've understood something, I need to start off on new ground.

Turning to the colour-classification methodology: The starting point are the four pure colours red, yellow, green and blue; their in-between shades and scales of brightness result in colour schemes containing 16, 64, 256 and 1,024 shades. More colours would be pointless because it wouldn't be possible to distinguish between them clearly.

To believe, one must have lost God. To paint, one must have lost art.

Throwaway snapshots come closest to achieving the state of pure picture.

I've never found anything to be lacking in a blurry canvas. Quite the contrary: you can see many more things in it than in a sharply focused image. A landscape painted with exactness forces you to see a determined number of clearly differentiated trees, while in a blurry canvas you can perceive as many trees as you want. The painting is more open.

But it is also untrue that I have nothing specific in mind. As with my landscapes: I see countless landscapes, photograph barely 1 in 100,000, and paint barely 1 in 100 of those that I photograph. I am therefore seeking something quite specific; from this I conclude that I know what I want.