In the most recent issue of the American Journal of Education, I sought to show that extended teacher-preparation programs were risky and promising because of the following: a) absent significant changes in the benefits of being a teacher, increasing the costs of becoming a teacher (by about $20,000) would result in a decline in the quality and quantity of teacher applicants. b) taxpayers' costs for the additional year of college and subsidies for graduate study of sufficient size to minimize the shrinkage of the candidate pool, would amount to billions of dollars. c) despite the fact that extended programs have been tried for years, the only evidence related to their efficiency is negative. d) extended programs, in themselves, did not address the fundamental weaknesses in teacher-preparation programs. e) a cognitive developmental theory of teacher learning would lead one to a different model of teacher preparation that focused on collegebased preprofessional education, school-based development of practical competencies and the opportunity for those who are committed to a professional career in education to undertake graduate study on completion of three to four years of classroom teaching. I concluded from these observations that we should be encouraging innovations of various types and systematically assessing the results before the nation is asked to accept a single model for preservice teacher preparation. Three of my colleagues were good enough to comment on the article. Gary Rhoades seemed to find the overall argument persuasive but argued that I did not calculate the costs of and obstacles to the alternative I proposed. He is right, and I will return to this later. Bernard Gifford and Hendrik Gideonse, however, argued that my case was fatally flawed. Let me respond briefly to each of the main criticisms made by these estimable leaders of the movement for extended teacher preparation. It should be noted that the critiques of my article were based on a next-to-last draft, and there are substantial differences between the draft and the published article. While the basic criticisms leveled at