Abstract
Social science is viewed as requiring a paradigm fundamentally different from that currently in use which reflects essentially the habits of classical physics. Its basic characteristic is the prevalence of minor and major events which amplify sequentially in such a way as to converge toward or diverge from expected outcomes. Other distinguishing characteristics of social science are discussed in the context of amplification and include the historicity of its subject matter, the everpresent tension between stability and change, the frequent primacy of phenotypic specificity over a genotypic generality, the importance of context and meaning in the constitution of data, and the altering effects of theory on reality. Recognition of these characteristics requires a change in the role of the investigator so as to include him more intimately in the research process. It also suggests the usefulness of substitution of an orientative sample for the usual representative sample to take into account the researcher's new role, especially in the context of amplification. The orientative sample permits an accretionary N ≥ 1 initially, as the investigator is sensitized and instructed in the course of his research.