To all such criticism as this (and I notice it only because it expresses the feelings into which many sensible and thoughtful minds have been fashioned by infection) the answer is simple and straightforward. It is just as impossible to generalize granite and slate, as it is to generalize a man and a cow. An animal must be either one animal or another animal; it cannot be a general animal, or it is no animal; and so a rock must be either one rock or another rock; it cannot be a general rock, or it is no rock. If there were a creature in the foreground of a picture, of which he could not decide whether it were a pony or a pig, the Athenæum critic would perhaps affirm it to be a generalization of pony and pig, and consequently a high example of "harmonious union and simple effect." But _I_ should call it simple bad drawing. And so when there are things in the foreground of Salvator of which I cannot pronounce whether they be granite or slate, or tufa, I affirm that there is in them neither harmonious union nor simple effect, but simple monstrosity. There is no grandeur, no beauty of any sort or kind; nothing but destruction, disorganization, and ruin, to be obtained by the violation of natural distinctions. The elements of brutes can only mix in corruption, the elements of inorganic nature only in annihilation. We may, if we choose, put together centaur monsters; but they must still be half man, half horse; they cannot be both man and horse, nor either man or horse. And so, if landscape painters choose, they may give us rocks which shall be half granite and half slate; but they cannot give us rocks which shall be either granite or slate, nor which shall be both granite and slate. Every attempt to produce that which shall be _any_ rock, ends in the production of that which is _no_ rock.
It is true that the distinctions of rocks and plants and clouds are less conspicuous, and less constantly subjects of observation than those of the animal creation; but the difficulty of observing them proves not the merit of overlooking them. It only accounts for the singular fact, that the world has never yet seen anything like a perfect school of landscape. For just as the highest historical painting is based on perfect knowledge of the workings of the human form, and human mind, so must the highest landscape painting be based on perfect cognizance of the form, functions, and system of every organic or definitely structured existence which it has to represent. This proposition is self-evident to every thinking mind; and every principle which appears to contradict it is either misstated or misunderstood. For instance, the Athenæum critic calls the right statement of generic difference "_Denner_-like portraiture." If he can find anything like Denner in what I have advanced as the utmost perfection of landscape art--the recent works of Turner--he is welcome to his discovery and his theory. No; Denner-like portraiture would be the endeavor to paint the separate crystals of quartz and felspar in the granite, and the separate flakes of mica in the mica slate,--an attempt just as far removed from what I assert to be great art, (the bold rendering of the generic characters of form in both rocks,) as modern sculpture of lace and button-holes is from the Elgin marbles. Martin has attempted this Denner-like portraiture of sea-foam with the assistance of an acre of canvas--with what success, I believe the critics of his last year's Canute had, for once, sense enough to decide.