Constitutional Dialects: The Language of Transnational Legal Orders

Abstract

What influences shape the drafting of constitutions, and to what extent? One way of tackling these questions is to analyze constitutions at the linguistic level. Just as literary language is rich with indications of influence, genre, and authorship, constitutional language contains telltale signs of the mix of influences at work. Conventional empirical research methods that rely on the manual coding of documents are poorly suited to studying subtle linguistic patterns. However, new methods from computational linguistics excel at identifying such patterns and enable us to measure how much of a text is associated with a particular topic, influence, or genre.

This chapter from the forthcoming volume Constitution-Making and Transnational Legal Order employs a technique known as topic modeling to analyze over two centuries worth of national constitutions, together with a selection of international human rights treaties, for telltale signs of various influences on constitutional drafting. The results suggest that certain hegemonic forces have left a profound impact on constitutional texts at the linguistic level. To be more specific, topic modeling identifies linguistic manifestations of four highly prevalent TLOs at the constitutional level: British colonialism, French colonialism, Spanish colonialism, and socialism. Each of these transnational legal orders is associated with a distinctive genre, or dialect, of constitutional language. Changes over time in the popularity of these constitutional dialects are the linguistic artifacts of the rise and fall of competing empires.

Keywords: constitutions, transnational legal orders, computational linguistics, constitutional dialect, colonialism, british, french, spanish, constitutional genre, constitution-making, automated text analysis, automated content analysis, topic model, constitution-making