TL;DR
This paper investigates how political parties in Iceland change their language use over time, specifically focusing on "Stylistic Fronting" (SF) in their speeches. They found that parties in power tend to use more formal language (higher SF rates), and this linguistic shift is often a reaction to significant political events like economic crises.
What Problem Does This Paper Address?
While previous research has explored how individual politicians change their language use over their careers or in response to personal crises, and how different parties use language at a single point in time, there's a gap in understanding how entire political parties evolve their linguistic behavior over extended periods. This paper aims to fill that gap by examining the dynamic linguistic strategies of Icelandic political parties across multiple government periods, specifically looking at how they react to historic events and shifts in political power.
Methods
The researchers utilized the Icelandic Parliament Corpus, a parsed subcorpus of the Icelandic Gigaword Corpus, to analyze speeches from 1999 to 2021. They focused on a syntactic feature called "Stylistic Fronting" (SF), which involves moving a word or phrase to the subject gap position in relative clauses, indexing a more formal style. Their analysis involved processing over 180,000 instances of SF and its absence. They employed Python scripts for data processing and statistical analysis using R, including chi-square tests and mixed-effects regression models. The models considered factors like government period, party status (majority/minority), the role of the speaker (member/minister), sex, and verb type.
Datasets
The primary dataset used is the Icelandic Parliament Corpus, a parsed subcorpus of the Icelandic Gigaword Corpus. This corpus contains speeches from Icelandic MPs spanning the government periods from 1999 to 2021. No new benchmarks were introduced; the study leverages an existing, comprehensive corpus.
Key Insights and Ideas
- Party Status Matters: Political parties tend to use more formal language, indicated by higher Stylistic Fronting rates, when they hold political power. Conversely, they become less formal when in opposition.
- Linguistic Reactions to Crises: Historic events, particularly economic downturns, can trigger significant shifts in political power, which in turn lead to traceable linguistic reactions from entire political parties.
- SF as a Formal Marker: Stylistic Fronting in Icelandic is identified as a marker of formal speech, and its use by political parties can be strategically employed to convey a particular persona or stance.
- Ministerial Role and Formality: The role of a minister is associated with more formal speech, likely due to the preparation involved in their speeches. This effect indirectly contributes to the observed formality shifts related to party power.
- Divergent Strategies in Coalitions: In complex political situations, such as when parties with opposing ideologies are forced into a coalition, a party might deliberately decrease its SF use to differentiate itself from its "political opponents" within the same government, even while holding power.
Key Results
- The Independence Party and the Progressive Party showed stable SF use while in power from 1999 to 2009, but significantly decreased it upon losing power in 2009. They increased SF use again when returning to governing coalitions in 2013.
- The Social Democratic Alliance also increased SF use when in power, particularly during the economic crash period (2007-2013).
- Regression models confirmed that government period and party status were highly significant predictors of SF use, with significant differences observed between majority and minority party statuses.
- The role of minister was also a significant predictor, indicating more formal speech compared to members.
- The Left-Green Movement showed a surprising decrease in SF use when gaining political power in 2017, an exception attributed to the unique political circumstances of forming a coalition with opposing parties.
Limitations and Future Work
The authors acknowledge that their study did not delve deeply into the fine-grained stylistic conditioning of SF use, such as how the specific type of speech might affect SF rates. They lacked detailed data on the stylistic contexts of political speeches, relying indirectly on the MP's role (minister vs. member) as a proxy for formality. Future research could explore other linguistic markers, particularly lexical ones, that might indicate different tones. Additionally, the study is limited to the Icelandic language and political system, suggesting that similar analyses in different political contexts might yield different results due to variations in governmental structures.
Summary updated