SOME EMPTY THEORETICAL BOXES
One of the most famous and fruitful -- yet in one respect futile -- controversies ever to appear in the economic journals was a dispute over the respective roles of theory and fact between two giants of the tradition of economics, A. C. Pigou and J. H. Clapham.1 The controversy began when the latter launched an attack on economic theory. He complained specifically about the contemporary version of diminishing return, constant return, and increasing return industries. He preferred to discard such concepts because, first, "the Laws of Returns have never been attached to specific industries... the boxes are, in fact, empty...we do not, for instance, at this moment know under what conditions of returns coal or boots are being produced."2 Even if filled with facts, such terms are not translatable into useful directives for public policy. And finally, any hope one might entertain actually to fill the boxes and thereby to establish empirical conclusions is "not very encouraging."3 The empty boxes of economic theory are, in sum, both unfilled and useless if filled; these weaknesses, however, are unimportant in the end because the boxes are virtually unfillable.
Pigou's counter-offensive was likewise multi-pronged. General categories have proved useful elsewhere, e.g., in mathematics. Indeed, to speak in general terms at all presupposes some kind of empty box; a term such as "commodities," and nothing less general, is necessary to analyse certain characteristics of hats and gold watches and onions at the same time. Since the time of Adam Smith theorists have employed such terms to "disentangle and analyse the causes by which the values of different things are determined."4
I shall discuss neither the particular concepts of diminishing, increasing, and constant returns, nor the question of theory's____________________