A glass-fronted skyscraper at night, every single window lit, and revealing a series of miniature, meticulously sharp tableaux of the interior.
A bright white wall of alpine mountainside, evenly speckled all over with dark trees and rocks.
A hugely spacious hotel atrium-lobby, whose architecture is viewed with miraculously exact symmetry, and whose colours are incredibly pure and flat. And vast May Day crowds, packed trading floors, offices with zooming perspectives of desks, sweeping nocturnal city-scapes.
His pictures are, for photographic prints, very large. Two and a half metres is a common width. Whatever the subject, his shots look highly rehearsed, set up, as far as can be from snaps. They seem to hold more than the eye can see. They have a great depth of focus, far and near, all sharp. Their angles of vision can be extremely wide (and with no apparent distortions). They are packed with very small but very solid, high-resolution detail (often these details are ant-like humans). They have regularities that appear too neat to be true. The images fall perfectly into flat, abstract designs. They have within them unreal patterns of repeated elements.
When people interpret Gursky, they tend to talk about the way his pictures stress the depersonalisation, regimentation, massification, standardisation, hygienisation of modern society and modern nature. If this is indeed their vision - and, as stated, it seems considerably less fascinating than the pictures themselves - then it is not a vision that the artist just discovers in the world he points his camera at. He deliberately puts it there. To be specific: this is a body of work that has perhaps in every case been subjected to digital manipulation.
Though seamless to the eye, the process is sometimes almost visible to common sense. Surely the horizon above the river couldn't be that dead straight. Surely that rising vista of the Bundestag, which keeps on going up, involves more than one point of view. Surely the lighting of that crowd of figures on the Chicago trading floor is too bitty and collage- like. Surely that hotel lobby is beyond belief. And can you spot literal image-repeats in the bits of dark rock and tree dotted around the snow? Maybe one doesn't spot the half of it. Maybe one would be staggered to discover how little unaltered actuality remained. I don't know. But I must admit, I never feel quite happy about all this.
Another visitor to the gallery didn't feel happy at all, saying: "He doesn't know whether he wants to be a photographer or a painter." That visitor, I'm sure, was a photographer. But the question is, with image technology moving where it's moving, is Gursky or anyone else really obliged to decide between the two professions? Can't you be both, or something in between? Well, I think there are points to argue here.
People tend to say bluffly: images are all just images, however they're produced - and besides, traditional photography is hardly innocent of monkey-business. That's true, of course. Photographers can make very free with lenses, cropping, printing, the over- or under-exposing of details. (Gursky does a lot of this, too). And you may say this is no different in kind from the stuff that can be done with computers.
I think there is a difference, though, and it's as much ethical as technical. Traditional photography does generally draw a line where manipulated photography doesn't. It won't change the shape or position of something in an image. It won't introduce bits from other images, or splice two images together. No doubt it won't do these things partly because it can't do them, or not as seamlessly as computers can. All the same, the line itself is not entirely arbitrary. It's a matter of respecting some basic matters of fact.
But if a rational line can be drawn, why insist on it? Here the case for traditional photos seems weaker. Yes, some photos are about fact and evidence. But lots aren't, and then they have to defend their photographic status with something like an angler's boast: that, through luck, alertness, cunning, patience, they actually saw and caught this image. So? It's as if the picture, whatever its virtues, can only be worth anything if it's founded on a find.
It can be hard to shed this feeling. I was shocked, going round, to realise how much I missed that sense of trouvaille, and felt cheated by its absence - what, he just makes it up? But I take the point. To insist on the pure find is often to impose a redundant virtuosity.
However, the argument is likely to turn on the image-manipulators too. For you could equally ask them: so why this insistence on making images that still look just like photographs? Why do collage, but then remove all the visible cuts and jumps to create a smoothly unified view? Why stay loyal to the distinctive photographic look - that purely tonal style of image-making that comes out of the camera?
For surely, the only claim this particular look has on us is precisely its long association with found truth. But if this association is now utterly broken, why hold on to the style, which is now just one style among many? If you have your vision of the modern world, why not paint it or collage it or whatever? Why stick to making fake photos?
There can be answers to these questions, and answers which don't involve fakery. For instance, you may say that the photo-look has become simply the lingua franca of the modern image. Once it had the authority of truth, but now it's just normal, the style everyone understands, so one had better use it. Or on the other hand, you might specifically want a truth effect - the idea could be to make people imagine the manipulated picture's fiction as actuality; or maybe to stage a clash between the photo-look and an impossibly unreal image. Or a photo-look might be employed to make some point about photography itself.
And now this argument-excursion has got us back to Gursky's work and what it's up to, namely: several of the tricks mentioned above, especially the last. His pictures aren't trying to pass themselves off as normal photos. They're trying to be super-photos, pictures that, through artifice, are excessively good at being the things a normal photo might wish to be. (No surprise that a photographer should resent them.) They are glutted with the photographic. They're like a parody - a parody not of individual artists, but of a whole art and its whole take on the world - or of a world which has been so over-photographed that now it turns itself into a photo before our eyes.
But to see this, of course, one needs to see the pictures as not quite true. Whether, as image-technology advances, and we get used to it, and the photo-look is more and more dissociated from any truth value, we'll then still be able to catch Gursky's large and beautiful ironies, is a point that can at least occur to us.
'Andreas Gursky - Photographs 1994-98': Serpentine Gallery, Kensington Gardens, London W2; daily; admission freeReuse content