Those soft-contoured, light-infused zones of weighty color, sometimes looming toward the viewer, sometimes moving away, offering access yet denying it - the traits of Mark Rothko's classic paintings are unmistakable.
They suggest a deep silence - the noiseless intensity of paint-resting-on-surface like contained condensation. Perhaps it is this quietness that, until recent years, has persuaded people that Rothko himself had somehow chosen to avoid words and explanations. But it has also taken years for his writings to surface. The publication of his book "The Artist's Reality: Philosophies of Art" only took place in 2004 - 34 years after he died and more like 64 years after he actually wrote it.
It is now plain that this American artist did not, after all, opt for monkish verbal abstinence, even when he started to produce his expressively mute classic paintings.
Writings on Art is the latest evidence of his revealingly eloquent ways with words.
This is not to say that he found it easy to convey in words the essence of paintings that he strongly felt were themselves entirely communicative. But they involved a new painterly language, and those who wrote about them, even appreciatively, resorted to art terminologies that inevitably missed the point.
The difficulties he himself had in finding word-equivalents for his paintings show up clearly in a series of elusive letters written to the critic and exhibition organizer Katharine Kuh in the mid-1950s. Her hope of publishing written interviews between the two of them never materialized, so this correspondence (though nothing she wrote to him is included) is a fascinating glimpse of why it didn't happen.
"From the moment I began to collect my ideas," he wrote her, "it became clear that here was not a problem of what ought to be said, but what it is that I can say."
It would seem that even if others often misunderstood his paintings, he still trusted their capacity to communicate directly with people better than words could. He confessed an "abhorrence for forewords and explanatory data. And if I must place my trust somewhere, I would invest it in the psyche of sensitive observers who are free of the conventions of understanding. I would have no apprehensions about the use they would make of these pictures for the needs of their own spirits. For if there is both need and spirit there is bound to be a real transaction."
Such an attitude intriguingly links this collection of his writings to the earliest of his essays, which are about art education. Rothko's "ideal teacher ... of creative art activity ... must possess the sensibility of an artist," he wrote in 1941. "Art must be to him a language of lucid speech inducing the understanding and exaltation which art properly inspires."
In an address to the Pratt Institute in 1958 he was still making the point: "I have never thought that painting a picture has anything to do with self-expression. It is a communication about the world to someone else. After the world is convinced about this communication it changes. The world was never the same after Picasso or Miro. Theirs was a view of the world which transformed our vision of things."
So, inarguably, did Rothko's.
Perhaps Rothko, even more than artists who have been much less articulate, was still, in the final analysis, caught between the impossibility of explaining on the one hand and the need to explain on the other.
He told the writer John Hurt Fischer: "A painting doesn't need anybody to explain what it is about. If it is any good, it speaks for itself...." But this remark was aimed at "presumptuous" art critics trying to interpret what can't be interpreted rather than at his own rather revealing attempts to explain. What he had to say, in fact, makes surprisingly good reading.
• Christopher Andreae has been writing for the Monitor since the 1960s. He lives in Glasgow, Scotland.