Echoes of power: language effects and power differences in social interaction

By Cristian Danescu-Niculescu-Mizil, Lillian Lee, Bo Pang, Jon M. Kleinberg. The Web Conference, 2011

372 citations recorded by the source corpus.

Understanding social interaction within groups is key to analyzing online communities. Most current work focuses on structural properties: who talks to whom, and how such interactions form larger network structures. The interactions themselves, however, generally take place in the form of natural language --- either spoken or written --- and one could reasonably suppose that signals manifested in language might also provide information about roles, status, and other aspects of the group's dynamics. To date, however, finding domain-independent language-based signals has been a challenge. Here, we show that in group discussions, power differentials between participants are subtly revealed by how much one individual immediately echoes the linguistic style of the person they are responding to. Starting from this observation, we propose an analysis framework based on linguistic coordination that can be used to shed light on power relationships and that works consistently across multiple types of power --- including a more "static" form of power based on status differences, and a more "situational" form of power in which one individual experiences a type of dependence on another. Using this framework, we study how conversational behavior can reveal power relationships in two very different settings: discussions among Wikipedians and arguments before the U. S. Supreme Court.

Subfields: Communication, Statistical and Nonlinear Physics

View Paper: https://www.semanticscholar.org/paper/884a2aed62a7afe78e0b6a3d08f7f98ad2c2db1e

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Communication

In Statistical and Nonlinear Physics

Citations in the corpus

Entity-Centric Contextual Affective Analysis (2019, background): “Studies of power have largely been limited to a dialog setting (e.g. Danescu-Niculescu-Mizil et al. (2012), see Prabhakaran (2015) for an overview), and almost no work has examined agency, with the exception of connotation frames.”
No country for old members: user lifecycle and linguistic change in online communities (2013, background): “More recent studies have shown that similar mechanisms are at work in online communities, revealing the emergence of linguistic norms in e-mail [35], Twitter [7, 19, 20, 38, 45] and internet forums [8, 14].”
Computational Sociolinguistics: A Survey (2016, methodology): “Depending on the type of power and the used dataset, labels for the relations or roles of individuals have been collected in different ways, such as based on the organizational structure of Enron [29, 85], the number of followers in Twitter [49], standings in state and national polls to study power ”
Winning Arguments: Interaction Dynamics and Persuasion Strategies in Good-faith Online Discussions (2016, background): “While stopwords may be related to how challengers coordinate their style with the OP [14, 35], content words can be a good signal of new information or new perspectives.”
Write Like I Write: Herding in the Language of Online Reviews (2014, background): “The language of communicators of a lower social status is expected to converge to the patterns of more powerful or prestigious interlocutors (Herring 1996; Wang, Fussell, and Setlock 2009; Danescu-Niculescu-Mizil et al. 2012).”
Conversations Gone Awry: Detecting Early Signs of Conversational Failure (2018, methodology): “To determine the prompt types of comments in our dataset, we ﬁrst apply the above procedure to derive a set of prompt types from a disjoint (unlabeled) corpus of Wikipedia talk page conversations (Danescu-Niculescu-Mizil et al., 2012).”
Does It Capture STEL? A Modular, Similarity-based Linguistic Style Evaluation Framework (2021, background): “…similarity between the 8 dimensional binary LIWC style vectors (1 if a word of the category is present in the sentence, 0 otherwise) proposed in Danescu-Niculescu-Mizil et al. (2012), (c) LIWC (function) similarity by taking 1− the difference between the relative frequencies of function words.”
A computational approach to politeness with application to social factors (2013, background): “Ad-mins have a higher status than common editors ( non-admins ), and this distinction seems to be widely acknowledged by the community (Burke and Kraut, 2008b; Leskovec et al., 2010; Danescu-Niculescu-Mizil et al., 2012).”
It's (Not) Simply a Matter of Time: The Relationship Between CMC Cues and Interpersonal Affinity (2016, background): “, 42] as well as large-scale online environments such as Twitter [7] and Wikipedia pages [8].”
Text Preprocessing for Text Mining in Organizational Research: Review and Recommendations (2020, background): “…of speech are considered vital for understanding language use (Pennebaker et al., 2003), and counting various parts of speech can provide insight into how language patterns shift depending on contextual and relational dynamics (e.g., Danescu-Niculescu-Mizil et al., 2012; Yang & Srinivasan, 2016).”
Semantic Variation in Online Communities of Practice (2018, neutral): “mbers. For example, Cassell and Tversky (2005) and Huffaker et al. (2006) investigate the correlation between linguistic features and the strength of the relations among users in newborn communities; Danescu-Niculescu-Mizil et al. (2012) and Noble and Fern´andez (2015) show how variations in linguis”
Characterizing and curating conversation threads: expansion, focus, volume, re-entry (2013, background): “In thinking about how to bring long threads to users’ attention, though, a natural question is whether there are sub-classes of such conversations that should be treated differently.”
Can Large Language Models Transform Computational Social Science? (2023, methodology): “Specifically, text analysis can uncover power relationships in the degree to which one speaker accommodates to the linguistic style of another (Danescu-Niculescu-Mizil et al., 2012).”