Semiology is a science of forms, since it studies significations apart from their content. I should like to say one word about the necessity and the limits of such a formal science. The necessity is that which applies in the case of any exact language. Zhdanov made fun of Alexandrov the philosopher, who spoke of "the spherical structure of our planet." "It was thought until now," Zhdanov said, "that form alone could be spherical." Zhdanov was right: one cannot speak about structures in terms of forms, and vice versa. It may well be that on the plane of "life" there is but a totality where structure and forms cannot be separated. But science has no use for the ineffable: it must speak about "life if it wanted to transform it. Against a certain quixotism of synthesis, quite platonic incidentally, all criticism must consent to the ascesis, to the artifice of analysis; and in analysis, it must match method and language.