Updated About content Print Article Share Article
views updated

neo-positivism A movement in early twentieth-century American sociology which blended together the three themes of quantification, behaviourism, and positivist epistemology. Its principal proponents were Franklin H. Giddings and George A. Lundberg, although the mathematical sociology of writers such as George K. Zipf (1902–50) can be seen as a development of neo-positivist theory.

In his Studies in the Theory of Human Society (1922), Giddings offered a defence of behaviourism, arguing that ‘psychology has become experimental and objective. It has discriminated between reflex and conditioning’. He also insisted that ‘sociology [is] a science statistical in method’ and that ‘a true and complete description of anything must include measurement of it’. Similarly, Lundberg maintained that sociology could be modelled on the natural sciences, and should observe the behaviour of human beings in social situations but without reference to concepts such as feelings, ends, motives, values, and will (which he described as ‘the phlogiston of the social sciences’). Like Giddings, Lundberg argued that science dealt in exact descriptions and generalization, both of which required ‘the quantitative statement’. He emphasized the importance of attitude scales in this context, and insisted (in common with earlier positivists) that science cannot formulate value statements, and that sociology must be a science in this mould.

In so far as neo-positivism had a lasting influence on the development of American sociology, this is perhaps best seen in later mathematical sociology, as for example in Richard M. Emerson's attempt to integrate mathematical theory and exchange theory (reported in J. Berger et al. ( eds.) , Sociological Theories in Progress, 1972)
. There are those (see, for example, J. Gibbs , Sociological Theory Construction, 1972
) who continue to insist that the most important criterion of a scientific theory is testability, and that only a mathematically formalized theory is empirically testable.