Vocabulary Growth Study: An Example with theState of the UnionAddresses

This paper analyses the vocabulary growth over the State of the Union addresses from 1790 to 2014 (225 speeches delivered by 42 US presidents). Because the context and the content of these speeches are fixed, this corpus presents a snapshot of the situation of the country on an annual basis. Based on this corpus, this study evaluates the fitness of two models predicting the vocabulary growth. The main interest, however, lies in the variations between the predicted vocabulary length and the observed one. This unsupervised method is automatic and is able to deduce facts from a stream of texts. More precisely, this study demonstrates that US history can be subdivided into time periods in which the presidents generate more new words than expected, and other periods where the influx of new words are lower than predicted. In such cases, the presidents tend to favour more already used words or expressions, repeating the same arguments or explanations and proposing similar solutions. This is one characteristic of nearly all the US presidents from 1830 to 1930. On the other hand, when facing unexpected contexts or willing to suggest radically new solutions, the presidents will introduce more new words than expected. As examples of such presidencies, we can mention G. Washington, T. Roosevelt, F. D. Roosevelt, or R. Reagan.