This year marks the 50th anniversary of the original publication of Thomas Kuhn’s famous book, The Structure of Scientific Revolutions. Kuhn, who taught at Berkeley, Princeton and MIT following studies in physics at Harvard, was a historian of science whose ideas have had a major impact on the philosophy of science.

Now in its third edition, Structure has had a lasting influence on our thinking about science. After 50 years, Kuhn’s ideas show signs of wear. But they continue to shape our “image of science”, to echo Kuhn’s own turn of phrase in the opening lines of Structure.

The argument of Structure

The core idea of Structure is that scientific research is based on underlying theoretical structures that provide a framework for research in a field for a sustained period of time. Kuhn’s name for these structures was paradigm. Indeed it was Kuhn’s use of the word that inserted “paradigm” into the popular lexicon.

His original use of the word was flexible. But he had two key points in mind. First, there’s a set of beliefs about a domain of study (including generalisations and a model of how the domain is constituted) that’s adopted as the basis for scientific practice in a scientific field at a time.

Second, there are a number of important examples of exemplary scientific research which later scientists look back to as guiding inspiration for their own research. Examples of this include Copernican heliocentric astronomy, Lavoisier’s oxygen-based chemistry and Darwin’s theory of evolution by natural selection. All of these constituted paradigms for scientists working in these areas for a significant period of time, both in the sense of providing an overarching set of beliefs about the world and in the sense of providing examples of exemplary research.

The controversy

Kuhn’s most controversial ideas relate to how paradigms change. In his terms, the replacement of one paradigm by another constitutes a scientific revolution. His use of the term revolution was deliberate. Like political revolution, scientific revolution involves a radical break with the past.

In what Kuhn termed “normal science”, scientists employ an accepted paradigm to solve “puzzles” that are thrown up in the attempt to apply the paradigm to nature. The solution of puzzles is governed by the norms and procedures of that paradigm.

But paradigms sometimes run into trouble. They face “anomalies” that resist solution within the paradigm. If anomalies proliferate, the community of scientists within a paradigm may enter a period of “crisis”.

The results of crises

In crisis, scientists behave in an unusual way. They propose and develop alternatives to the existing paradigm. An alternative to the paradigm may acquire a following as scientists “convert” from the accepted paradigm to the newcomer.

If the scientific community converts to the alternative, a revolution has occurred, and normal science recommences on the basis of the newly-appointed paradigm.

Kuhn points out that the choice between the reigning paradigm and the challenger is unable to be determined by the norms and procedures of normal science. Nor is there any set of fixed and all-encompassing rules of scientific method able to dictate the choice. The choice involves a shift of gestalt that Kuhn compares to religious conversion.

Irrational science?

For many readers of Kuhn, the take-home message was one of irrationalism and relativism, since choice between paradigms seemed unable to be made in a rational way on the basis of objective criteria. And this message of irrationalism may explain the popularity of Kuhn’s ideas. Certainly, it is one basis for the controversy that has surrounded Structure since its initial publication.

But, with hindsight, the message of the book can be understood differently. Kuhn’s critics argued that there are independent standards that may be employed in the choice of paradigm. Kuhn himself came to accept the point in later work.

Scientists, Kuhn thought, evaluate theories or paradigms using a set of values, such as accuracy, simplicity, consistency, breadth and fertility. The values are employed throughout the sciences, and may be employed to evaluate competing theories or paradigms.

A question of values

But there’s a hitch. They do not constitute a “neutral algorithm of theory choice”. They do not yield a mechanical decision procedure that will deliver a unique outcome acceptable to all parties.

The values may conflict with each other. They may be interpreted in different ways. Scientists who appeal to the same set of values may understand them differently, and reach conflicting decisions based on the same values. They may even reach the same decision on the basis of differing weightings and interpretations of the values.

From a more contemporary perspective, what this suggests is not that paradigm choice is irrational, but that rational choice between paradigms is a deliberative process in which scientists exercise a variety of judgements that may differ significantly. The choice is not a mechanical one governed by an algorithm. But it may be a rational one just the same.