The concept of the paradigm shift was introduced in the 1960’s and revised in 1970 by Thomas S. Kuhn in his influential book The Structure of Scientific Revolutions (Kuhn 1970). Kuhn was the first to publish this account of how knowledge in science advances: Scientists conduct experiments and carry out their observations to study unanswered questions arising from and within the dominant theory in the field. The dominant theory is the scientific community’s best collective understanding of how the part of the world they study works, and their experiments and observations provide evidence to either support or refute the theory. Alternatively, the results of scientists’ experiments and observations are explained in terms of dominant theory and the dominant theory is used to predict observations. Disparity between the current theory and the observations are common, and usually explained (sometimes satisfactorily, sometimes not).
Eventually, the discrepancies between theory and observation reach the point where observations no longer support the dominant theory, and the dominant theory no longer explains or predicts observations. When these disparities are no longer tenable, a paradigm shift occurs within the science. The theory that was dominant is replaced with a new theory, and the new theory frames and defines new research problems and new research methods. Some paradigm shifts occur quickly and are completed within a few years, and others occur slowly and are completed over decades. Once a new paradigm has become established, there are typically some individuals who continue to work according to the old paradigm, but those individuals find themselves increasingly marginalized. The difficulty in announcing a paradigm shift is that it is rarely characterized by clear boundaries, so it becomes clear that a paradigm shifted only after the shift.
Kuhn, Thomas. 1970. The Structure of Scientific Revolutions, 2nd ed. enlarged. Chicago: University of Chicago Press.