I don't think it's naive to assume a photograph can provide evidence of certain important aspects of the external world. Just because it can be easily faked or misread, it doesn't mean the documentary potential has no value. A fossil of a feather preserves some characteristics of it size, shape and texture, if lacking in many other ways. Likewise with a black and white photograph of a feather with a ruler placed next to it. Photographs usually need a backstory to support the evidence presented. To a scientist, that may mean tentatively trusting the source of the photograph and how it was taken. In law, it may mean the photographer might have to be called in as a witness.
And I can think of few circumstances in art where the backstory is not important. In the field of painting in the age of photography, it suddenly became important to many to know if a painting was made from life or photographs; in landscape photography in the digital age, the question about manipulation is inevitable. It's not so much of a legal issue, it's simply a way for the viewing public to know something of the artist's intentions, which in turn helps them process what they're looking at. We're human. It's not always just the cold arrangement of things on a surface that's important, but the circumstances of how it was made. Especially when you're serious about buying "fine art".
By invoking "artistic license" and claiming there are no rules, Alain undermines one of the medium's most powerful attributes. He concedes that photography can be used as evidence, but ignores the potential that has for art. The problem is, he works in the vernacular of western landscape photography, right down to the use of titles of the specific locales they were taken. This kind of work likely attracts a public who are suspect of invented realities. But because Alain has moved a few trees around, stretched the height of a mountain, used false color, and scrubbed tributaries from waterways, he's got the wrong backstory for that public. It also puts him in the position to have to explain that some of the miraculous features in his shots are as he found them, even if he rearranged some trees here and there.
I think most people concerned about manipulation would come to understand that a photographer might amplify atmospheric effects with global moves, exaggerate the effects of light with contrast or "dodge and burn" methods, or that the color of the scene is not the same spectral value as it was the moment the picture was taken. But it's the cut and paste; the moving and/or distorting of major topological features that signals that the artist is not interested in preserving the original fingerprint of the optical projection. This destabilizes one of the chief advantage photography can claim over other arts, however tenuous its connection to reality. The slippery slope of all this leads to photographic confections like "Sugarloaf Rock", by Peter Eastway, the making of which is profiled on this site. It's fine to do this sort of photography, it's just that you have to think about it in a very different context. The growing popularity of this approach is probably why so many people are asking the question about manipulation. But as kencameron suggested at the top of page 3 of this debate, photographers should be careful about complaining. I would add that it's perfectly reasonable to want to know if that breaching whale leaping out of the Antartic ocean (also seen on this site) is connected to an actual event and not a cut and paste job, dropped in from a stock photo found on the internets.