Patterns of age‐based linguistic variation in American English¹

Abstract

In prior sociolinguistic research, speaker age has been considered the principal correlate of language change, but it ‘has not yet been explicitly studied as a sociolinguistic variable’ (Eckert 1997: 167). Consequently, little is known about how language varies across the life span. The present study employs key word analysis on a large corpus of casual conversation in American English to explore age‐based linguistic variation in spontaneous conversation. Analyses of the key words point to two major patterns of age‐based lexico‐grammatical variation: use of slang, and use of stance and involvement markers. Younger speakers' talk is characterized by an unusually frequent use of slang and swear words, and by a marked use of features indexing speaker's stance and emotional involvement, including intensifiers, stance adverbs, discourse markers, personal pronouns, and attitudinal adjectives; older speakers favor modals. These patterns are suggestive of functional differences in the discourse of youth and adults. It is argued that the expression of personal stance is more explicit and plays a key role in younger speakers' discourse.

Demographic Factors Improve Classification Performance

“…The subcorpora for the discrete variable gender are relatively straightforward (although see (Bamman et al, 2014b)), but the split for the continuous age variable are less clear. While the effect of age on language use is undisputed (Barke, 2000;Barbieri, 2008;Rickford and Price, 2013), providing a clear cut-off is hard. We therefore use age ranges that result in roughly equally sized data sets for both groups, and that are not contiguous.…”

“…In other words, we do have certain expectations as to who uses "super cute," "rather satisfying," or "rad, dude." Sociolinguistics has long since studied the interplay between demographic factors and language use (Labov, 1964;Milroy and Milroy, 1992;Holmes, 1997;Macaulay, 2001;Macaulay, 2002;Barbieri, 2008;Wieling et al, 2011;Rickford and Price, 2013, inter alia). 1 These factors greatly influence word choice, syntax, and even semantics.…”

Discriminative subprofile-specific representations for author profiling in social media

“…The AP problem has been approached from different areas, including psychology (Pennebaker and Graybeal, 2001;Pennebaker and Stone, 2003), linguistics (Holmes and Meyerhoff, 2008;Barbieri, 2008), sociolinguistics (Eckert, 1997), and natural language processing (NLP) (Koppel et al, 2002;Schler et al, 2006). The rest of this section reviews related work from the latter field, where studies have traditionally focused on the use of specific textual features and machine learning techniques for AP.…”

Stylistic variation over 200 years of court proceedings according to gender and social class

“…Recently, also the possible interplay between social variables is considered (Prabhakaran and Rambow, 2017), but is mostly confined to age and gender (see e.g. Ardehaly and Culotta (2015); Argamon et al (2007); Barbieri (2008); Burger et al (2011);Eckert and McConnell-Ginet (2013); Holmes and Meyerhoff (2003); Hovy and Søgaard (2015); Nguyen et al (2014);Peersman et al (2011);Schwartz et al (2013);Wagner (2012)) as other social variables -such as social class -are not easily available (cf. Sloan et al (2015)).…”