Personal Schedule
Change Preferences / Time Zone
Sign In
Search
Browse By Day
Browse By Time
Browse By Person
Browse By Mini-Conference
Browse By Division
Browse By Session or Event Type
Browse Sessions by Fields of Interest
Browse Papers by Fields of Interest
Search Tips
Location
About APSA
Bluesky
X (Twitter)
Scholars of judicial politics have long examined how courts react to political pressure, yet we know surprisingly little about how such pressures reshape the meaning of constitutional concepts over time. Existing approaches either focus on case outcomes or rely on close reading of selected judgments, which makes it hard to detect gradual and often subtle changes across entire bodies of case law. This paper proposes a methodological approach that addresses this problem by tracing informal constitutional change through the language used in judicial decisions.
The study draws on the complete corpus of Turkish Constitutional Court decisions from 1962 to 2025 and applies contextualized word embeddings to measure how the Court’s interpretation of key constitutional ideas evolves. It uses BERTurk-Legal to generate vector representations for concepts such as separation of powers, fundamental rights and secularism, and compares these vectors across different political periods. Cosine similarity scores and diachronic visualizations reveal semantic shifts that go unnoticed with more conventional qualitative or quantitative techniques.
Turkey is chosen as the main case because it provides unusually clear conditions for observing informal constitutional change. The country has experienced repeated episodes of constitutional crisis, executive–judicial conflict and political pressure on courts, yet these pressures often produced shifts in judicial interpretation rather than formal constitutional amendments. This combination of dense political variation and a stable constitutional text offers an ideal setting for testing methods that aim to detect subtle semantic change. In this sense, Turkey functions as a model case that allows the proposed framework to be evaluated under conditions where semantic drift is theoretically expected to occur.
The findings show that the Court adjusts its interpretive language during periods of heightened political tension, and that these adjustments accumulate over time in ways that amount to informal constitutional transformation. The methodological approach developed in this paper can be adapted to other jurisdictions and offers a new way to study constitutional evolution by analyzing judicial language at scale.