I have been following with much interest the discussion on "why teach the history of error?". I had intended to contribute by adding to this discussion no more than the following quotation from William James: "You can give humanistic values to almost anything by teaching it historically. Geology, economics, mechanics, are humanities when taught with reference to the successive achievement of the geniuses to which these sciences owe their being. Not taught thus, literature remains grammar, art a catalogue, history a list of dates, and science a sheet of formulas, and weights and measures." What I would have then said was that economics, when taught in the absence of its history, is a quite barren subject. Without some study of the history of economics, students are being short changed on what ought to be seen as a very important part of their education. What I was not going to say was what I felt more deeply. That is that there is a very large amount to be learned in even thinking about the ideas we have discarded. More importantly, I was not going to say that at least in some instances, there are ideas that have fallen by the wayside which may even be more valid and useful than the ones we find in our textbooks today. Mohammed Gani by citing my book in his criticism of Say's Law has brought me into this discussion to pursue, briefly, that last point. My own introduction to Say's Law may be instructive. It came about because I was reading John Stuart Mill's Principles (1848) and came across the best refutation I have ever read of Keynes's General Theory (1936). Not only did I find the arguments so compelling that I stopped seeing myself as "Keynesian" virtually from that moment, but I also realised that the arguments on both sides had a long history and had been debated for more than a century before Keynes wrote. Keynes overturned an older basis for economic thought with another ancient economic principle, but one which had been roundly rejected for more than a century at the time the General Theory was written. It also led me to the issue which was to be the subject of my thesis, published in 1998 as Say's Law and the Keynesian Revolution. The series of essays published in 2003 and cited by Mohammed Gani was intended to be an even-handed assessment of where Say's Law stood exactly 200 years after the publication of Say's Treatise. But if you want a full-scale defence of Say's Law, you should go to that earlier work. And there you will find the following points (amongst many others): Say's Law was the base framework for "macroeconomics" from the start of the nineteenth century through until the publication of the General Theory in 1936. Virtually every mainstream economist during that time accepted Say's Law, not passively, but actively, formally and explicitly. The list includes, but is by no means restricted to, all of the following: Say, James Mill, Ricardo, Senior, McCullough, John Stuart Mill, Marshall, Jevons, Walras, Wicksell, Haberler and Hawtrey. No one called it Say's Law, a phrase from the 1920s. The related propositions of Say's Law were, instead, the near-universally accepted basis for the theory of the cycle and were seen as essential to an understanding of unemployment. As a series of concepts, it was basic textbook theory and was discussed in just about every introductory text on economics published right up to the 1930s. Say's Law was thus not some hazy long-abandoned theory of the early nineteenth century. It was the absolute mainstream right up until the moment it was displaced by the economics of the General Theory. The phrase "supply creates its own demand" is not an adequate interpretation of what those who held Say's Law to be true had in mind. Basing criticisms of Say's Law on those words will ensure that even if the argument satisfies the author, it will have no relationship to the meaning of the Law of Markets held by those who thought it was valid. For those interested, the next issue of the History of Economics Review, which I believe will be released this month, contains an article of mine on this very question: "'Supply Creates its Own Demand': A Discussion of the Origins of the Phrase and of its Adequacy as an Interpretation of Say's Law of Markets". The development of Say's Law, moreover, had nothing to do with trying to understand barter economies. Since Lange and Keynes that is how Say's Law is frequently taught but it is not right. Say himself, in his own discussion, was dealing with the problem of a shortage of money and whether that would be an adequate explanation for slow sales. Discussions of the business cycle, built on Say's Law, never intimated that it was an abstraction without real world application to a money economy. Those who wanted to undermine the Law of Markets may have seen this as a useful line of attack, but it has nothing to do with its actual meaning as an economic principle. Let me end with this. For myself, if you want to study the history of error, you need go no further than any modern macroeconomics text based on C+I+G+X-M or IS-LM or AS-AD, which are all theoretical developments derived from Keynes's 1936 attack on Say's Law. It may be the mainstream but if you see things the way I do, descriptions of reality based on these models are one error piled on another. One day all of these may be added to the boneyard of discarded theories (is any theory immortal?). Right now, however, they are the theoretical basis for quite a good deal of economic policy, even though some people, and not just me, think they are invalid as a means to understand how economies in the real world actually work. Steve Kates