Linguistic Variation in Social Settings
a cognitive and behavioural exploration of the interplay between language and identity
We all adapt and taper our language to the unique context of our specific social environment, adopting different morphosyntactic constructions, registers, slang and even accents based on who we are talking to. Exploring this social function of language is the domain of sociolinguists, who explore the effects of society on language and vice-versa.
In this blog post, I aim to explore cognitive and behavioural evidence exploring the link and motivations between linguistic variation and identity expression, as well as how various facets of identity are expressed and perceived by the speaker and the listener respectively.
Language Variation
Inherent to any social context are a range of different communicative needs and contexts that lead speakers to make various choices: these range from using pronunciation variants (eg. talkin’ vs talking) to choosing different languages within different speech communities. Speech is influenced by factors such as the speaker’s sex and gender, regional background, ethnicity, socioeconomic status, as even external contexts such as the degree of formality of the interaction (eg. at a bank or a market). Such markers are encoded in speech by competent speakers, and trained listeners in turn are able to exploit this meaning to interpret this speech.
A classic example from the Hebrew Bible details the Gileads who, following the people of Ephraïm, ask them to pronounce the word shibboleth: anybody who pronounced the initial consonant as a [ʃ] was a friend and anybody who said [s] revealed themselves to be an enemy to be killed. The term shibboleth is now used to refer to any choice of phrasing or words that distinguishes a linguistic subgroup of peoples.
The encoding of such meaning can also be highly intentional: a sketch by the American comedy duo Key & Peele illustrates the use of speech to assert certain identities in different contexts:
Linguistic variance, conscious or otherwise, thereby encodes a variety of indexical meaning in a speaker’s speech, indexing properties (eg. “educated”, “non-native”, “macho”) that are highly subjective and variable. Chevrot, Drager & Foulkes (2018) note this in their review of recent work on indexical meaning:
For example, releasing /t/ in positions where it is most commonly unreleased or produced as a glottal stop can index both prissiness (Podesva, 2004) and geekiness (Bucholtz, 2001) … variation thus may be seen as a resource for achieving individual communicative goals, such as manipulating the social distance between a speaker and interlocutor (Giles & Powesland, 1975), accommodation to one’s actual or imagined audience (Bell, 1984), or expressing aspects of social identity (Le Page & Tabouret-Keller, 1985).
Case Study: Social Stratification in NYC
Another experiment that evidences the relationship between perceived social standing and dialect is Labov’s seminal 1966 study “The Social Stratification of (r) in New York City department stores”. By visiting a number of department stores from various price ranges — Saks, Macy’s and Klein’s (in descending order of the price and fashion scale) — Labov elicits utterances (one casual, the second time more emphatically) from employees of the phrase “fourth floor”, in order to elicit (r) the presence or absence of consonantal [r] in postvocalic positions.
Analysing the trend from the data collected, Labov uncovers the use of (r) as an element of high prestige speech and the dialect differences that exist in different social strata. Stores of perceived higher class and calibre were more likely to engage in rhoticity at the time. Such a conclusion suggests that rhoticity is a marker of high social prestige and points further to linguistic variation as a carrier of indexical meaning and social aspiration.
Such an experiment was also worthwhile in that it avoided the issue of the observed effect. By collecting the data through covert and anonymous methods in natural, Labov was able to record unforced speech and control the context of speech production to elicit the desired responses accurately and effectively. These avoided the issues that often crop up in formal research settings, whereby participants adapt their speech and change their behaviour owing to the unnatural nature of the setup.
Motivations for Linguistic Variation
It is evident that current models of language and cognition therefore need to account for the myriad social factors that are interlinked with speech production. In order to understand the links between linguistic variation and social factors, we also need to consider the motivations that might exist for engaging in linguistic variation, obvious or otherwise.
Prestige
Central to understanding how social factors affect linguistic variation is the concept of prestige, i.e. the level of regard given to certain languages, dialects or even certain speech forms within a speech community. Often, the most prestigious varieties of speech correlate to the standard form, i.e. those promoted actively by governmental or religious authorities, and are associated with social standing, educational level, and professionalism. By contrast, lower prestige forms—anything deviating from the imposed standard—such as, for instance, morphosyntactic variants in American English as documented in the Yale Grammatical Diversity Project, are perceived as more informal and less educated.
Negative consequences of perceptions of what constitute lower prestige speech is most evident in the phenomenon of linguistic profiling, or speech subcues that reveal indexical information about individual’s socioeconomic or racial background. Typically listeners are very quickly able to identify accented speech in their native language: one study found that English speakers were able to identify French accent accurately within the first 30 milliseconds of the utterance (Flege, 1984).
The response of accented speech, in particular that of ethnic and racial minorities, is to typically rank speakers lower on categories of status (intelligence, education, and social class) and solidarity (friendliness, trustworthiness and kindness) (Feurtes, Potere, and Ramirez, 2002). A seminal study by John Baugh, studying this phenomenon through hundreds of phone calls revealed companies racially discriminated against voices that sounded African-American or Mexican-American. In response to such ‘dialect prejudice’, AAVE-speaking populations are often pressured to gain bi-dialectal competence, code-switching to standard English to avoid systemic racial bias at the cost of their psychological health.
Covert Prestige
In certain informal settings, however, it is often the non-standard form of the language that lends social currency to the speaker and is regarded with high value: this is called covert prestige. For instance, in my home country of Singapore, the local English-based creole of Singlish is often derided as a language with low prestige. Various governmental campaigns, such as the recent ‘Speak Good English’ movement, frown upon what is perceived to be an inferior form of the English language and attempt to eliminate it from popular use. At the same time, the ability to use Singlish fluently in informal settings—in markets, homes, schoolyards—continues to be ‘the quintessential mark of Singaporean‐ness’ (Chng, 2003), establishing an informal rapport and intimacy with a Singaporean listener that lends the speaker great social currency. As such, angmos (foreigners) who are able to codeswitch to Singlish fluently are viewed and respected as ‘true’ locals, as in the video below:
As creole languages invariably tend to face pressure from their lexifier languages that hold “higher” linguistic prestige, languages such as Singlish, Kristang or Haitian Creole may face a lot of sociopolitical pressure to undergo decreolisation, dropping nonstandard features and converging with the standard language (English, Portuguese and French in the above examples). In fact, certain linguistic theories hold that modern AAVE has already undergone the process of decreolization in the US.
Empowerment of Marginalised Communities
The use of low-prestige, nonstandard language as a means of constructing a positive identity and reclaiming pejorative terms has also been extensively documented, particularly in the reappropriation of negative slurs by minority racial groups and queer individuals. Mutonya (2007) also discusses the positive construction of identity in his analysis of a restructured variety of Swahili, Kinoki, spoken by Nairobi’s street community in “Redefining Nairobi’s Streets: A Study of Slang, Marginalization, and Identity”, where reactions were elicited by school-going pupils to a variety of pejorative terms.
Results indicate that street children, unlike their school-going peers living in the city’s low-income neighborhoods, redefine pejoratives that devalue and stigmatize street people and their lifestyle. Instead, Kinoki empowers the marginalized community to construct a positive identity, to ameliorate representations of street lifestyle, and to redefine neologisms that reference in-group (us) and out-group (them) experiences.(Mutonya, 2007)
Such findings reveal divergent attitudes and demarcations for in-group and out-group members, revealing the use of language for social inclusion and exclusion. Such examples can be found all over the world, and contest simplistic models of prestige as directly correlating with “standard-ness”: in doing so, they reveal the often arbitrary, self-imposed, and complex nature of prestige in the myriad social settings of any individual.
Priming Effects in the Processing and Production of Speech Variation
Experiments in cognitive sociolinguistics have also attempted to discern cognitive mechanisms in the processing of sociolinguistic variation: plenty of evidence points to priming effects on listeners caused by an individual’s perceived gender or race. Evans, Munson & Edwards (2018) discovered that both trained and untrained listeners perceived and interpreted African American English (AAE) speech more accurately when primed with African American faces, mediated by individuals’ knowledge and beliefs about AAE.
In a similar experiment, Strand (1999) revealed that for listeners classifying acoustically ambiguous utterances between [s] and [ʃ], awareness of speaker gender effects our categorisation. Female faces induce the perceptual boundary to move up in frequency and male faces move the perceptual boundary down (evident in the picture to the right). This is evidence that listeners preemptively account for the different acoustic properties of typical male and female voices when processing speech.
It seems, then, that cognitive mechanisms employ the formation, processing, and encoding of various stereotypes and general social beliefs for the sake of efficiency. Such priming effects can be particularly insidious, and are even evident in speech production: a study by Dupree revealed that “white Americans who hold liberal socio-political views use language that makes them appear less competent in an effort to get along with racial minorities”. Such patronising (albeit well-intentioned) language redirects us to back to the problem of linguistic profiling.
Dupree and Fiske suspect that the behavior stems from a liberal person’s desire to connect with other races … it’s also possible that “this is happening because people are using common stereotypes in an effort to get along,” Dupree says. (from Yale Insights)
Conclusion
The complex and bidirectional relationship between language and identity and colourful insights from observing sociolinguistic variation lead us develop a deeper understanding of human cognition, self-expression, and the language production. Language is part of a cognitive toolkit that allows us to invariably express facets of our identity competently in different social settings, and many external factors and motivators—class aspirations, local loyalty, educational background—play a role in determining the kind of language we speak: this non-modular relationship forces us to rethink speech processing and how we should account for these learned associations in models of speech and social cognition. Ultimately, we are left with the clear impression that linguistic and social knowledge are inextricable from each other. Future research endeavours that continue to explore this link would be mutually beneficial to both cognitive scientists and sociolinguists alike, and it would also be interesting to gain clarity on cognitive processes and social motivators that instigate such language change. Another interesting question that merits serious academic exploration would be the process of acquisition of comptency in sociolinguistic variation among native and non-native learners. Ultimately, and more importantly, our discussion of discrimination and linguistic profiling (however malignant or well-intentioned it may pretend to be) encourages us to research further into the role of media, education, and exposure to the formation of inaccurate schema and ignorant socially constructed beliefs that negatively discriminate against out-group members and engender self-demeaning beliefs.
Bibliography
Chng, Huang Hoon (2003). “You see me no up: Is Singlish a problem?”. Language Problems & Language Planning. 1: 45–62.
Evans, K. E., Munson, B., & Edwards, J. (2018). Does Speaker Race Affect the Assessment of Children’s Speech Accuracy? A Comparison of Speech-Language Pathologists and Clinically Untrained Listeners. Language, speech, and hearing services in schools, 49(4), 906–921. https://doi.org/10.1044/2018_LSHSS-17-0120
Flege JE. The detection of French accent by American listeners. J Acoust Soc Am. 1984 Sep;76(3):692–707. doi: 10.1121/1.391256. PMID: 6491043.
Fuertes, J. N., Potere, J. C., & Ramirez, K. Y. (2002). Effects of speech accents on interpersonal evaluations: implications for counseling practice and research. Cultural Diversity and Ethnic Minority Psychology, 8(4), 346–356. https://doi.org/10.1037/1099-9809.8.4.347
Jean-Pierre Chevrot, Katie Drager, Paul Foulkes. Editors’ Introduction and Review: Sociolinguistic Variation and Cognitive Science. Topics in cognitive science, Wiley, 2018, 10 (4), pp.679–695. hal- 01969828
Le Page, R. B., & Tabouret-Keller, A. (1985). Acts of Identity : Creole-Based Approaches to Language and Ethnicity. Cambridge ; New York : Cambridge University Press.