Qualitative thinking in the age of modern computational chemistry - or what Lionel Salem knows

A personal assessment is given of the present state of computational chemistry, and in particular of the persistent need for qualitative orbital thinking at a time when exceedingly accurate calculations are possible. The need for the qualitative view arises, it is claimed, from the inherent difference between predictability and understanding, from certain potential impediments to chemical understanding intrinsic to human-computer interactions, from the peculiar yet productive way that experiment and theory interact in chemistry, and lastly from some special features of theory in science in general. (C) 1998 Elsevier Science B.V.