Scientific discovery is a complex matter. Its occurrence is often serendipitous and it may take time for scientists to recognize that an anomalous phenomenon has been discovered. Then the scientists must rework their expectations, conduct experiments and observations to determine if they can accommodate the new facts.
A rich variety of conceptions about knowledge generation and proper scientific reasoning were expounded in the early modern period to explain and bolster particular theories about living and nonliving nature. Often these accounts were not explicitly labeled as ‘methods of discovery’ but they clearly captured aspects of the scientific practice of binding together facts and showing them in a new light, namely inducting from specific to general conclusions.
For example, Francis Bacon insisted that scientific discovery must proceed by observing and measuring natural phenomena, combining and analysing them and then deducing from these facts what the nature of the universe is like. His colleague, Sir Isaac Newton, also advocated a scientific method of making observations and experiments, systematically expounding the definitions and axioms of knowledge, and drawing conclusions from them by induction.
Some philosophers have argued that these early modern accounts of scientific discovery cannot be used to support a logically grounded theory of discovery, because they involve a kind of intuitive process that is not amenable to rational examination. Others have taken a more pragmatic line, holding that discovery can be characterized as an extended, reasoned process. Norwood Hanson argues that this extended, reasoned process of articulating and developing ideas is governed by a special logic that differs from inductive and hypothetico-deductive logic.