Turkish Constitution Lemmatizer, 2025

This project adapts the text of the Turkish Constitution (1982) into an interactive semantic installation. Using a Turkey-specific natural language processing lemmatizer (Zeyrek/Zemberek), the system parses the formal legal text, extracts each word's morphological root (lemma), and constructs a client-side linguistic index.

By browsing through articles, chapters, and parts, the viewer interacts with the law not as static legal directives, but as a complex linguistic web. Hovering over a word dynamically maps and connects identical roots within the article using visual links. Clicking a word activates a global semantic index, exposing all other sections of the constitution containing the same root and providing non-linear pathways of navigation through the state's foundational document.