Analytical solutions without a theory
An empirical relationship is supported by confirmatory data irrespective of theoretical basis such as first principles. Sometimes theoretical explanations for what were initially empirical relationships are found, in which case the relationships are no longer considered empirical. An example was the Rydberg formula to predict the wavelengths of hydrogen spectral lines. Proposed in 1876, it perfectly predicted the wavelengths of the Lyman series, but lacked a theoretical basis until Niels Bohr produced his Bohr model of the atom in 1925.
Some empirical relationships are merely approximations, often equivalent to the first few terms of the Taylor series of the analytical solution describing the phenomenon. Other relationships only hold under certain specific conditions, reducing them to special cases of more general relationship. Some approximations, in particular phenomenological models, may even contradict theory; they are employed because they are more mathematically tractable than some theories, and are able to yield results.
- Hall, Carl W.; Hinman, George W. (1983), Dictionary of Energy, CRC Press, p. 84, ISBN 0824717937
- McMullin, Ernan (1968), “What Do Physical Models Tell Us?”, in B. van Rootselaar and J. F. Staal (eds.), Logic, Methodology and Science III. Amsterdam: North Holland, 385–396.
- Roman, Frigg; Hartmann, Stephan. "Models in Science". In Zalta, Edward N. The Stanford Encyclopedia of Philosophy (Fall 2012 ed.). Retrieved 24 July 2015.