# Evolving linguistic divergence on polarizing social media arXiv: https://arxiv.org/abs/2309.01659 ยท HTML: https://mlsanity.com/paper/2309.01659/ ## Key facts - Cited by 6 papers in the map (data through 14 Oct 2026). - Builds on 32 of its 122 references that are in the map. - First posted on arXiv 4 Sep 2023. ## Builds on (most cited first, 30 of 32) - [Enriching Word Vectors with Subword Information](https://mlsanity.com/paper/1607.04606/) (2016) - [ChatGPT Outperforms Crowd-Workers for Text-Annotation Tasks](https://mlsanity.com/paper/2303.15056/) (2023) - [Distributed Representations of Sentences and Documents](https://mlsanity.com/paper/1405.4053/) (2014) - [Can Large Language Models Transform Computational Social Science?](https://mlsanity.com/paper/2305.03514/) (2023) - [Diachronic Word Embeddings Reveal Statistical Laws of Semantic Change](https://mlsanity.com/paper/1605.09096/) (2016) - [Is ChatGPT better than Human Annotators? Potential and Limitations of ChatGPT in Explaining Implicit Hate Speech](https://mlsanity.com/paper/2302.07736/) (2023) - [Top2Vec: Distributed Representations of Topics](https://mlsanity.com/paper/2008.09470/) (2020) - [SemEval-2020 Task 1: Unsupervised Lexical Semantic Change Detection](https://mlsanity.com/paper/2007.11464/) (2020) - [Analyzing Polarization in Social Media: Method and Application to Tweets on 21 Mass Shootings](https://mlsanity.com/paper/1904.01596/) (2019) - [Algorithmic Amplification of Politics on Twitter](https://mlsanity.com/paper/2110.11010/) (2021) - [Diachronic Usage Relatedness (DURel): A Framework for the Annotation of Lexical Semantic Change](https://mlsanity.com/paper/1804.06517/) (2018) - [Factors Influencing the Surprising Instability of Word Embeddings](https://mlsanity.com/paper/1804.09692/) (2018) - [A Wind of Change: Detecting and Evaluating Lexical Semantic Change across Times and Domains](https://mlsanity.com/paper/1906.02979/) (2019) - [Temporal Attention for Language Models](https://mlsanity.com/paper/2202.02093/) (2022) - [Contextualized language models for semantic change detection: lessons learned](https://mlsanity.com/paper/2209.00154/) (2022) - [Making "fetch" happen: The influence of social and linguistic context on nonstandard word growth and decline](https://mlsanity.com/paper/1802.04140/) (2018) - [Storywrangler: A massive exploratorium for sociolinguistic, cultural, socioeconomic, and political timelines using Twitter](https://mlsanity.com/paper/2007.12988/) (2020) - [Dialectometric analysis of language variation in Twitter](https://mlsanity.com/paper/1702.06777/) (2017) - [Conceptual similarity and communicative need shape colexification: an experimental study](https://mlsanity.com/paper/2103.11024/) (2021) - [Identification of Biased Terms in News Articles by Comparison of Outlet-specific Word Embeddings](https://mlsanity.com/paper/2112.07384/) (2021) - [(Mis)alignment Between Stance Expressed in Social Media Data and Public Opinion Surveys](https://mlsanity.com/paper/2109.01762/) (2021) - [Quantifying the dynamics of topical fluctuations in language](https://mlsanity.com/paper/1806.00699/) (2018) - [Detecting Political Biases of Named Entities and Hashtags on Twitter](https://mlsanity.com/paper/2209.08110/) (2022) - [Frequency patterns of semantic change: Corpus-based evidence of a near-critical dynamics in language change](https://mlsanity.com/paper/1703.00203/) (2017) - [How Polarized Have We Become? A Multimodal Classification of Trump Followers and Clinton Followers](https://mlsanity.com/paper/1711.00617/) (2017) - [Semantic Journeys: Quantifying Change in Emoji Meaning from 2012-2018](https://mlsanity.com/paper/2105.00846/) (2021) - [American cultural regions mapped through the lexical analysis of social media](https://mlsanity.com/paper/2208.07649/) (2022) - [Polarization and reliability of news sources in Wikipedia](https://mlsanity.com/paper/2210.16065/) (2022) - [Polarization of Climate Politics Results from Partisan Sorting: Evidence from Finnish Twittersphere](https://mlsanity.com/paper/2007.02706/) (2020) - [Unpacking polarization: Antagonism and Alignment in Signed Networks of Online Interaction](https://mlsanity.com/paper/2307.06571/) (2023) ## Built on by (most cited first, 6 of 6) - [Machine-assisted quantitizing designs: augmenting humanities and social sciences with artificial intelligence](https://mlsanity.com/paper/2309.14379/) (2023) - [HistLens: Mapping Idea Change across Concepts and Corpora](https://mlsanity.com/paper/2604.11749/) (2026) - [Causal Effects of Trigger Words in Social Media Discussions: A Large-Scale Case Study about UK Politics on Reddit](https://mlsanity.com/paper/2405.10213/) (2024) - [Are Large Language Models the future crowd workers of Linguistics?](https://mlsanity.com/paper/2502.10266/) (2025) - [Content Moderation in TV Search: Balancing Policy Compliance, Relevance, and User Experience](https://mlsanity.com/paper/2505.17207/) (2025) - [Measuring Narrative Polarization in Online Discourse](https://mlsanity.com/paper/2601.07398/) (2026) ## Abstract (arXiv) Language change is influenced by many factors, but often starts from synchronic variation, where multiple linguistic patterns or forms coexist, or where different speech communities use language in increasingly different ways. Besides regional or economic reasons, communities may form and segregate based on political alignment. The latter, referred to as political polarization, is of growing societal concern across the world. Here we map and quantify linguistic divergence across the partisan left-right divide in the United States, using social media data. We develop a general methodology to delineate (social) media users by their political preference, based on which (potentially biased) news media accounts they do and do not follow on a given platform. Our data consists of 1.5M short posts by 10k users (about 20M words) from the social media platform Twitter (now "X"). Delineating this sample involved mining the platform for the lists of followers (n=422M) of 72 large news media accounts. We quantify divergence in topics of conversation and word frequencies, messaging sentiment, and lexical semantics of words and emoji. We find signs of linguistic divergence across all these aspects, especially in topics and themes of conversation, in line with previous research. While US American English remains largely intelligible within its large speech community, our findings point at areas where miscommunication may eventually arise given ongoing polarization and therefore potential linguistic divergence. Our methodology - combining data mining, lexicostatistics, machine learning, large language models and a systematic human annotation approach - is largely language and platform agnostic. In other words, while we focus here on US political divides and US English, the same approach is applicable to other countries, languages, and social media platforms.