It is argued that technological progress is marked by a series of discrete “barriers” and “breakthroughs,” which create new fields of technological opportunity. These are not random events, although their exact timing is undoubtedly very difficult to predict. They occur when and where they do because of discontinuities in the laws of nature. The technological life cycle can be defined as the period from a major breakthrough which opens up a new territory for exploitation to the next major barrier. It is characterized by a rapid increase in marginal productivity of R&D to a peak, followed by a more-or-less continuous decline thereafter, as the territory is gradually exhausted. This model is qualitatively consistent with the well-known “S-shaped curve” phenomenon, describing measures of technological performance over time. The model is also qualitatively consistent with Schumpeter's explanation of the so-called Kondratieff or “long” wave.