Before 1962, paradigm was mostly a grammar term for verb tables
In 1900 Merriam-Webster defined paradigm only as a grammar term, the model for conjugating verbs. Then Thomas Kuhn's The Structure of Scientific Revolutions turned it into one of the most quoted ideas in science, describing how whole fields see the world, and why they sometimes flip, like a rabbit becoming a duck.
The Greek parádeigma meant pattern or example, from roots meaning to show alongside. Classical orators used a paradeigma as a comparable case that nudged listeners towards a conclusion without forcing it; Anaximenes described it as past actions similar or opposite to those under discussion. Plato's Timaeus uses the word for the model the divine craftsman may have copied when making the cosmos. Later, the linguist Ferdinand de Saussure used paradigm for a class of similar elements, in contrast to syntagma, elements linked by relationship.
Kuhn gave the term its modern sense. For him, landmark works such as Newton's Principia or John Dalton's chemistry of 1808 supply a framework of concepts, results and methods that guides later research. Inside it, normal science settles which questions are worth asking, what the leading theory predicts, how results should be read, and how experiments should be run. A paradigm is not a rigid rulebook; scientists can apply it creatively, and individuals interpret it differently.
Science, in Kuhn's picture, alternates between long stretches of puzzle-solving within a dominant model and sudden revolutions. A paradigm has two layers: a set of admired example experiments that people copy, and underneath, shared assumptions, some hidden and almost metaphysical, that shape how evidence is gathered. To insiders the paradigm feels like reality itself, so alternatives seem absurd and awkward data tend to be dismissed. Those unresolved anomalies pile up until the framework gives way.
Borrowing N. R. Hanson's idea that observation is shaped by theory, Kuhn called the change a paradigm shift and compared it to an ambiguous picture flipping: in the rabbit-duck illusion you can see either animal but never both at once. Dictionaries have since broadened the word to any theoretical framework, well beyond science.
Source: Paradigm
