Can we detect conditioned variation in political speech? two kinds of discussion and types of conversation

Abstract

Previous work has demonstrated that certain speech patterns vary systematically between sociodemographic groups, so that in some cases the way a person speaks is a valid cue to group membership. Our work addresses whether or not participants use these linguistic cues when assessing a speaker’s likely political identity. We use a database of speeches by U.S. Congressional representatives to isolate words that are statistically diagnostic of a speaker’s party identity. In a series of four studies, we demonstrate that participants’ judgments track variation in word usage between the two parties more often than chance, and that this effect persists even when potentially interfering cues such as the meaning of the word are controlled for. Our results are consistent with a body of literature suggesting that humans’ language-related judgments reflect the statistical distributions of our environment.

How to measure political polarization in text-as-data? A scoping review of computational social science approaches

“…The qualitative analysis of the results points out to five general topics: the type of environment (real life or online), the online data contamination, the type of platform, the timeframes commonly selected, and the language dependence. Delving into the type of environment, the 11 papers selected for this scoping review can be split between in-real-life (IRL) speech (Mathew Gentzkow & Shapiro, 2010;Matthew Gentzkow et al, 2019;Sloman et al, 2021) and online discourse (Cantini et al, 2020;Esteve Del Valle et al, 2021;Jiang et al, 2020;Kursuncu et al, 2019;Makrehchi, 2016;Serrano-Contreras et al, 2020). Four papers use US congress speech, being the congress transcripts the majority of IRL data text, and three works use Twitter data.…”

“…From the ones selected, five measure political polarization specifically (Belcastro et al, 2020;Gentzkow et al, 2015;, while the rest present a polarization model as part of a bigger model. The latter type of models aim at predicting election results (Belcastro et al, 2020), finding communication factors affecting polarization , understanding the interaction between polarization and participation (Serrano-Contreras et al, 2020), determining if people can detect ideology through expressions (Sloman et al, 2021), predicting extremist text content (Kursuncu et al, 2019), understanding interaction between covid-19 and polarization in online discourse (Jiang et al, 2020), and predicting political conflicts (Makrehchi, 2016).…”

“…This approach provides a methodology to look for further polarized terms on Twitter and takes into consideration the dynamic characteristic of political polarization, inputting the time dimension. Sloman et al (2021) showcase both the Clustering and Scaling models. The political polarization model is part of a bigger question to investigate whether socially conditioned variation in speech is a factor to identify others based on their political identity.…”

The Discursive Dimensions of Pernicious Polarization. Analysis of Right-Wing Populists in Western Europe on Twitter

“…Jiang et al (2020) examine COVID-19 discourse polarization on Twitter, adding a temporal dimension to track polarization evolution in response to pandemic developments. Sloman et al (2021) explore Clustering and Scaling Models to uncover how linguistic variations reflect political identities, demonstrating that specific linguistic patterns can signal political affiliations and that polarity can be deduced from the lexical proximity or distance in speech. Despite advancements-as the authors of the work themselves also claim-challenges remain in analysis depth and adapting to textual context specificities.…”

The Pronunciation of the Capital of Ukraine in English-Speaking Media

“…Studies on the language of politicians suggest that phonetic variations could not only help in forming speaker's identity (Hay 2018) but also index political meaning (Hall-Lew et al 2010;Hall-Lew et al 2017). 'Kiev' to 'Kyiv' change, thus, could be ascribed to politically conditioned variation, which Sloman et al (2021: 2) define as "linguistic variation that can be anticipated on the basis of the speaker's political identity." In the 'Kiev' / 'Kyiv' dichotomy, one's political identity can be defined as pro-Russian or pro-Ukrainian.…”