ABSTRACT
As a result of the fine work of Mark Rowe, Joachim Schulte, Gordon Baker and Peter Hacker, 1 it has now been evident for some time that there are deep affinities — affinities in style and textual organization, in conceptions of elucidatory explanation via comparisons, and in a sense of subjectivity housed within nature — between the Goethe of the Farbenlehre and the Wittgenstein of Philosophical Investigations. Among the very deepest of these affinities is their shared sense of the limits of metaphysical explanation. The identification of simple elements is always relative to purposes and circumstances, never ultimate. Hence there is no single kind of ultimate explanation running from the nature and behavior of ultimate simples to the nature and behavior of complexes composed out of them. There are often useful explanations to be found of how the behaviors of complexes are determined by the behaviors of their parts, but this kind of explanation is one among many. Comparative descriptions of complexes — whether of organisms, human practices, works of art, or chemical and physical structures — are not to be supplanted in favor of ultimate metaphysical explanation.
