"Of course, one can go back to high school or, in my case, junior college and find roots; there are some indeed, but my professional beginnings were in the Berkeley that existed just before World War II and the scholarship I won to MIT. There I met the dazzling wunderkind Paul Samuelson. When I was browsing in the Berkeley library and came across early issues of Econometrica, Samuelson’s contributions caught my eye. When I got an opportunity to go to MIT, it was the possibility of working with Samuelson that confirmed all my choices. I was attached to him as a graduate assistant from the outset, and I tried to maximize my contact with him, picking up insights that he scattered on every encounter. Working with Samuelson, who was at the forefront of interpreting Keynesian theory for teaching and policy application, I was put immediately in the midst of two challenging contests—one to gain acceptance for a way of thinking about macroeconomics and another to gain acceptance for a methodology in economics, namely, the mathematical method. Later, both challenges were to be overcome, but for ten or twenty years opposition was fierce. Once Samuelson’s Economics became a widely used text in first courses in the subject, Keynesian economics was firmly embedded. There was no turning back from that achievement. The successive student generations turned more toward the mathematical approach in graduate school, and they taught or did research in this vein. That eventually established the mathematical method, first in the United States, then in Europe, Japan, India, and other centers. Much of the foundation was built in Europe, and many of the American masters at mathematical economics were immigrants, but Samuelson, Friedman, and others gave it a native-born American flavor, and the approach truly caught on in this country."