focus) helps to deepen and expand the explanatory power of the theory of plate tectonics.

A century ago the atomic substructure of matter was a theory, which became better established as new evidence and inferences based on this evidence deepened the complexity and explanatory power of the theory. Today, atoms are an established component of matter due to the modern capability of imaging individual atoms in matter with such tools as scanning-tunneling microscopes. This kind of progression from theoretical construct to observed property leads to some confusion in the minds of many people about the nature of theory and the distinctions among theory, evidence, claims, and facts. The history of science further reveals that theories progress from hypotheses or tentative ideas to core explanations.

Thus, another source of confusion for the public understanding of science is the use of the term “theory” to represent promising ideas as well as core explanatory theories. Core explanatory theories are those that are firmly established through accumulation of a substantial body of supporting evidence and have no competitors (e.g., cell theory, periodic law, theory of evolution, theory of plate tectonics). For much of science, theories are broad conceptual frameworks that can be invalidated by contradictions with data but can never be wholly validated.

To give a specific example: it is an observed property that things fall down when dropped near the surface of the earth. Repeated observations give the rate of acceleration in this event, both its global average and local variations from that average. Newton’s law of universal gravitation and Einstein’s general theory of relativity are two successive theories that incorporate this observation and give quantitative predictions for the size of the gravitational effects in any situation, not just on earth. These theories describe but do not actually explain gravitation in the conventional sense of that word; they invoke no underlying mechanism due to substructure and subsystems. The general theory of relativity includes Newton’s law of gravitation as a special limited case (an approximation or idealization, valid to high accuracy under certain conditions), but it is a more general theory that makes predictions for cases not covered by Newton’s law (e.g., the bending of light paths by the sun or other stars).

In this example, drawn from physics, the theories are expressed in mathematical form and their predictions are thus both precise and specific. They lend themselves readily to computer modeling and simulation. In other areas of science, theories can take more linguistic forms and involve other types of models. What is general is that scientific theories are valued when they (a) incorporate a significant body of evidence in a single conceptual framework and (b) offer predictive suggestions about future directions for study that are specific enough that one can test the theory’s validity and



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement