Thomas Kuhn, The Structure of Scientific Revolutions, chapter 2

‘Normal science’ is research that is based on one or more scientific achievements of past that is recognized as a base for future research. Nowadays these achievements are described in many science textbooks. These books present the central principles of these accepted theories and illustrate them. Before this kind of books emerged in the beginning of 19th century, books from science classics were often used (Aristoteles, Newton, Franklin, Lavoisier, Lyell).

Achievements that are innovative enough and yet do not answer all the relevant questions are called ‘paradigms’. This term is closely related to ‘normal science’. Paradigms offer models which form the basis for traditions: Newtons’ dynamics, Copernicus’s astronomy etc. People who work within same paradigms share same regulations and standards.

Different scientific subjects have different history of paradigms: electricity, light, genetics. Some fields developed their first generally accepted paradigm later than others.

Gathering evidence before a paradigm is established can be somewhat random because all collected facts seem equally important. Not theories but technology has (often) lead this process because many facts would be undiscovered without practical work – doctors, calendar makers, iron workers.

If paradigm shifts then some scientists will change their views but some will find themselves in isolation. Paradigms guide specialization – authors often start from where the theory books finish.