The frequency Distribution of English Syntagms

People use language principally as a means of communication, each language community determining the scope of the particular language it uses. Within the language, various sets and subsets may exist depending upon regional and social differences. Most studies of such subsets have been concerned with lexical variations. In this paper, a methodology is outlined by which the frequency usage of syntactic structures can be measured for subsets of a language, and for a language as a whole. In order to establish homogeneity features within a subset, reliance is made on calculations of chisquare on batches of sentences within the subset. Before considering whether or not a more sensitive test should be applied to detect significant differences within and among subsets of a language, it seemed advisable to determine the frequency distribution of syntagms throughout particular genres. An extension of the methodology to accomplish this objective is described and the results for one type of syntactic structure within one genre are tabulated. These results indicate that the frequency distribution of syntagms within these limitations approximates to a Poisson distribution.