This and other nature studies are among Claude Lorrain: Drawings, a hundred-odd works on paper drawn from the stores of the British Museum and the Ashmolean Museum. The show is now in Oxford at the Ashmolean and goes to the BM in October. These things can't be displayed much. Their sepia lines and washes, sometimes delicate, sometimes bold, easily fade in the light. In many cases they have already. But even if you know them well, they still come as a surprise.
They surprise by contrast. Claude, a Frenchman who spent his working life in Rome, is Mr Classical Landscape himself. You don't imagine him out and about, on foot or with a donkey, looking for vantage points, setting up, and taking his observations down fast. His paintings - there's one in a neighbouring room at the Ashmolean - are so eminently studied, staged, balanced, composed.
And while they contain a wide knowledge of the look of the world, its lights and atmospheres, his observations are combined and refined into views of some impossible promised land - a country which no viewer could realistically imagine entering. Claude's is a poetry of longing, of hazy distances that pull the eye further than it can see, of long slow fades into invisibility. You are not there and never will be, never could be.
So that's a world, a vision or whatever. But when you turn back to the outdoor drawings of trees and woods and streams, what is clear is that they are a world, too. They are not merely raw material, observation fragments, ingredients for the mixing, a book of swift, rough nature notes. They are another vision, and a rival one in a way.
Claude's angle here is much more intimate, for one thing. Nature is close up, to hand, transient - these are trees you might be shaded under, streams you could cross, shadows that will pass. He appreciates that nature doesn't always offer a clear view or even a particularly interesting one. A Grove in Shadow depicts a prominent but quite unidentifiable something in the middle ground, just a brush smudge. View of Shrubbery with a Wall is pretty near a picture of nothing at all.
What we see at work, in fact, are completely different criteria for what qualifies as a picture. Images that Claude could never have used to fill a canvas - a clumped screen of trees, stretching horizontally across the paper, with no beguiling distances peeping behind them - can be enough. It is a much more modern aesthetic than that of the paintings, one that values spontaneous and visible marks, simplified and imbalanced motifs, a sense of personal witness and communion. Compared to the paintings, this is much more our kind of picture.
Still, Claude could tell the difference; that is striking, too. We sometimes think that old artists carried on the way they did because they just couldn't conceive things otherwise. But evidently Claude could. He could imagine quite different kinds of pictures from those he painted in oils.
He knew the difference. The trouble is, he probably put a very different value on it. His direct nature studies may seem to carry us straight back through time, setting us on the spot where the artist sat 350 years ago, but in another way they only stress what time-bound, historified creatures we are. For when it comes to drawings, our valuations and those of Claude's contemporaries are hardly in contact at all.
The drawings they prized and collected weren't these outdoor ones, but rather the drawings that were closest to the paintings: Claude's neat composition studies, and the drawn copies he made after his paintings. There are plenty of these studio drawings in the show - indeed, I think a point is being made of this. It is the old struggle between the scholar and the general viewer.
For it would have been possible to put together almost as large an exhibition with only the open air studies. But that would simply be to pander to modern taste. Better to show the full range of Claude's drawings, to demonstrate the many roles drawing played in his composition process. For to devote a show purely to pleasure, to admit that we just prefer the observational work to the finished studies - that would be historically irresponsible.
Maybe. And of course it would be highly fanciful to suppose that Claude's priorities were really our own, that he felt cramped by the rules of his time, that when he looked at his lovely outdoor work he said to himself: "Now that's what I call a real picture, unfortunately the clientele won't stand for it. I guess I'm just ahead of my time, but we can't help ourselves, can we?" In basic matters of taste, we are not responsible. We can't just bring ourselves to make 17th century judgements about 17th century artists, and I doubt whether - when it comes to the crunch - the most responsible scholar can either.
To force the issue melodramatically, imagine some destruction dilemma scenario in which, of two drawings in this show, one can be saved and the other must be lost; and imagine that they are Landscape with Ascanius and the Stag and The Tiber from Monte Mario. The first: a complete drawn study for a painting of the same name, the one in the Ashmolean, though with interesting differences. The second: a very free dissolving open- air view, done in pure wash, with no obvious relation to any of Claude's paintings.
Which to chose? The second, obviously. That's the picture we want more. The choice is historically conditioned. You can be pretty sure that if the choice had been Claude's, it would have gone the other way. And yet now no other choice could be made. We are here, not elsewhere.
`Claude Lorrain: Drawings' at the Ashmolean Museum, Oxford (01865 278000), until 13 September; then at the British Museum (0171 636 1555), from 9 OctoberReuse content