Implicit-explicit language attitudes in England: Attitude change in apparent time data?

Robert M. McKenzie, University of Northumbria

Research conducted by social psychologists has frequently demonstrated low correlations between explicit and implicit attitude measures in a range of socially sensitive domains. Since implicit and explicit evaluations do not change at the same rate, any implicit–explicit attitudinal discrepancy (IED) may indicate attitude change in progress (see Charlesworth & Banaji 2019). However, (socio)linguists have yet to investigate whether differences between implicit and explicit attitudes towards language use can determine the direction of any language attitude change underway; surprising given recent evidence community language attitude change can result in micro-level language change over time.

This talk details the results of a recent study (McKenzie & Carrie 2018), employing an Implicit Association Test (IAT) and self-report attitude scale, examining the relationship between 90 English nationals' implicit and explicit ratings of Northern English and Southern English speech in England. Multivariate analysis demonstrated a significant implicit-explicit attitude discrepancy (IED), providing evidence of language attitude change in progress, led by younger females, with explicit attitudes changing more rapidly towards a greater tolerance, if not unreserved approval, of the English spoken in the north of England. The study findings are discussed in relation to the potential changing status of Northern and Southern English speech as well as the potential value of measuring implicit-explicit attitudinal discrepancy in apparent time data to help uncover language attitude change in progress.


  • Charlesworth, T.E.S. & M.R. Banaji. 2019. Patterns of implicit and explicit attitudes: 1. Long-term change and stability from 2007 to 2016. Psychological Science 30(2): 830-844. doi: 10.1177/0956797618813087
  • McKenzie, R.M. & E. Carrie. 2018. Implicit-explicit attitudinal discrepancy and the investigation of language attitude change in progress. Journal of Multilingual and Multicultural Development 39(9): 174-192. doi: 10.1080/01434632.2018.1445744

Rethinking the role of the broadcast media in (de)standardization: Glasgow dialect meets ‘Mockney’

Jane Stuart-Smith, University of Glasgow

In the UK, as in other western industrialised countries, which have experienced substantial geographical and social shifts during the course of the 20th century, substantial dialect levelling towards standard varieties has taken place. A typical assumption about processes of this kind of dialect standardization, has been that exposure to standard varieties through the broadcast media are likely to promote shifts towards that standard (e.g. Chambers 1998; Trudgill 1986). But changes in UK urban accents also show other patterns, which suggest that alongside standardization, other processes are at work which also affect non-standard dialects (e.g. Kerswill 2003).

This paper reviews the fate of a traditional dialect, Glasgow English, whose phonological development across the course of the 20th century indicates two kinds of shift: (1) illustrated by vowel monophthongs, e.g. LOT, GOAT, GOOSE, which is distinctive to Glaswegian and not easily linked to changes in Anglo-English, standard or non-standard; (2) illustrated by some consonants, which result in reflexes similar to non-standard Anglo-Englishes, e.g. TH-fronting, L-vocalisation, weakening of coda-/r/ (Stuart-Smith & Lawson 2017; Stuart-Smith, José, Rathcke, Macdonald & Lawson 2017). Taken together, and in the context of the city’s shifting social aspect across the same period, these sound changes reflect a process of continued de-standardisation of Glasgow dialect. The second group of changes also help illuminate the role played by the broadcast media in sound change in general, and in dialect shifts away from, as opposed to towards, standard English (Stuart-Smith, Pryce, Timmins & Gunter 2013).


  • Chambers, J. 1998. TV makes people sound the same. In L. Bauer & P. Trudgill (Eds.), Language Myths (pp. 123–131). New York: Penguin.
  • Kerswill, P. 2003. Models of linguistic change and diffusion: new evidence from dialect levelling in British English. In D. Britain & J. Cheshire (Eds.), Social Dialectology. In Honour of Peter Trudgill (pp. 223–243). Amsterdam: Benjamins.
  • Stuart-Smith, J. & E. Lawson 2017. Scotland: Glasgow and the central belt. Listening to
    the Past: Audio Records of Accents of English. doi: 10.1017/9781107279865.009
  • Stuart-Smith, Jane, B. José, T. Rathcke, R. Macdonald & E. Lawson. 2017. Changing sounds in a changing city: An acoustic phonetic investigation of real-time change over a century of Glaswegian. In C. Montgomery & E. Moore (Eds.), Language and a Sense of Place: Studies in Language and Region (pp. 38–65). Cambridge: Cambridge University Press.
  • Stuart-Smith, Jane, G. Pryce, C. Timmins & B. Gunter. 2013. Television can also be a factor in language change: Evidence from an urban dialect. Language, 89(3), 501–536.
    Trudgill, P. 1986. Dialects in Contact. Oxford: Blackwell.

When late-modernity is… late. Best language in post-communist Lithuanian media and community

Loreta Vaicekauskienė, Vilnius University

Among theoretical considerations of the SLICE project is the impact of the spoken media on the conception of a language standard of a community. For instance, the late-modern Danish TV and radio when producing a type of a fascinating, cool and self-assured public personality, that speaks modern Copenhagen, might have contributed to the ideological construction of this speech variety as a standard for the new public style, the so called “‘dynamic’ way of speaking” (as opposed to the “‘superior’ way of speaking”, reserved to the conservative Copenhagen).

In my presentation of SLICE-Lithuania, which closely followed the Danish research design, I would like to continue the discussion of interrelations between a community’s idea of “best language” and the development of new public speech styles mediatized by the spoken media. The Lithuanian community can be an excellent experimental arena for the study of this issue. While the Western societies and media underwent processes of democratization and informalization that apparently started in the 60s, the turning point for the post-WW2 Lithuania’s development was the fall of the Berlin Wall.

During my presentation I will discuss attitudinal data collected in 23 secondary schools in 8 regions of Lithuania during 2012-2013 and match them with the findings of language development in Lithuanian TV and radio 1960-2017. We shall see that young Lithuanians subconsciously produce a slight association of modern capital Vilnius speech with personality dynamism, but there is no evidence of such clear evaluative distinction between modern and conservative (orthoepic) standard as indices of resp. dynamic and superior speaker identities as it has been documented in the Danish investigations. It seems also that regardless of evident informalization, the Lithuanian media still consider the “‘superior’ way of speaking” handy. It might well be then that the Lithuanian data offer a version of the ‘norm-and-variation’ issue that represents an earlier stage of the late-modern development that has been progressing in at least some Western societies. If so, we have an additional clue for the study of how a standard language is being made.

State broadcasting as evidence of sociolinguistic change. How media language changes, and symbolic meaning changes with it

Jacob Thøgersen, Nicolai Pharao & Janus Mortensen, University of Copenhagen

Denmark had a broadcast monopoly from the 1920s until the 1980s. Its stated aim was to serve the nation and act as an educator of the people. In terms of language ideology, the state broadcasting company (Denmark’s Radio, DR) was expected to lead by example, and even to this day, public service broadcasters are required by the government to: “emphasize a correct and comprehensible Danish in the programs and to lead an active language policy” (Ministry of Culture 2018).

Through all the time broadcasting has existed in Denmark, a particular variant of /a/ acoustically closer to long /a:/, the so-called ‘flat’ a, has been vilified not only among linguists, rhetoricians, and pedagogues, but also in popular press. It should come as little surprise, then, that the use of ‘flat a’ in media language was (and is) sharply proscribed.

This paper aims to describe how media speakers handle the tension between on the one hand language variation and change (in this case the ‘flat a’), and on the other language prescriptivism. Is proscription able to keep a change-in-progress in check? Or conversely, will the vowel change win out in the sense that a formerly stigmatized variant becomes acceptable and no longer debated? The data will show that neither is the case: Language changes but apparently prescriptivism stays more-or-less stable. Data for this part is drawn partly from prescriptive literature, partly from approx. 7 hours of news reading spanning six decades.

A second goal of the paper is to investigate how the meaning potential of a (formerly) stigmatized variant changes as language changes. As the previously highly marked variant slowly gains acceptance in formal news reading, what happens to the previously unmarked, or ‘standard’, variant? Does it gain new symbolic meaning? The argument is that it does. It comes to mean ‘old’ and ‘conservative’; but not old and conservative in a neutral way (if there is such a thing), but old and conservative as linked with the ethos of the era of state monopoly media. Data for this part is drawn from radio satire which uses stylized ‘old speech’ to frame the comedy and social commentary, and possibly as a tool of Verfremdung.

Recent apparent time evidence shows that the change in news reading concerning short /a/ has affected younger middle class speakers: they now also use a raised short /a/, i.e. one that is acoustically closer to long /a:/, something that could only be observed for working class speakers in older generations. The paper proposes that these patterns can be seen as a case of sociolinguistic change, not just a change in linguistic form, ‘language change’, but a change in the interplay between linguistic forms and social meaning – a re-valorization of particular ways of speaking. Sociolinguistic changes expose social changes, here e.g. the view of the nation-state as exhibited through different ideologies for broadcast media, and at the same time, sociolinguistic changes have the potential to themselves promote social changes, e.g. in the hierarchy of social classes and class-based ways of speaking.


Linguistic diversity and standard ideology in Denmark

Lian Malai Madsen & Andreas Stæhr, University of Copenhagen

Sociolinguists have recently suggested that the changing social conditions across Europe results in a relativisation of sociolinguistic values (Coupland 2009: 45), which disturbs the ideology of ‘one best language’ and potentially results in “de-standardisation” (Coupland 2009: 44, cf. Coupland & Kristiansen 2011). Denmark, however, has often been characterised as having a sociolinguistic history of particularly strong standardisation of the spoken language (e.g. Kristiansen 2009), and in this national context is it is argued that while a previously ‘low prestige’ variety has gained a new status of ‘best’ speech, the ideological investment in a ‘standard’ variety is still intact (Coupland & Kristiansen 2011: 28).

In our presentation, we add to the picture the kind of vernacular and hybrid language use we are witnessing in larger European cities. Since this type of language use involves linguistic forms originating from a range of national languages, one might expect that it contributes to the development of “less determinate and more complex” sociolinguistic value associations (Coupland 2009: 43), which may question the position of a standard language ideology and the bounded code-model of language it implies. We discuss if this is the case through investigating how the urban vernacular is placed in the Danish sociolinguistic order. We draw on different data types from an ethnographic sociolinguistic study (Madsen et al. 2016) as well as social media and mass media productions involving hybrid urban language use and discuss the metapragmatic acts we find here in relation to the socio-linguistic history and the wider language ideological environment in Denmark.

Basing our argument on Agha’s discussion of ‘slang’ registers (Agha 2015), we suggest that although the contemporary urban linguistic practices we attend to, certainly in the situations of their use may undermine a bounded code-model of language, the wider processes of enregisterment across various communication platforms, in fact, suggest that this hybridity does not pose any significant challenge to the standard language ideology, but rather contribute to its reproduction.


  • Agha, A. 2015. ‘Tropes of slang’. Signs and Society 3(2), 306-330.
  • Coupland, N. 2009. Dialects, standards and social change. In Maegaard et al. (Eds.), 27-50.
  • Coupland, N. & T. Kristiansen. 2011. SLICE: Critical perspectives on language (de)standardisation. In T. Kristiansen & N. Coupland (Eds.) Standard Languages and Language Standards in a Changing Europe (pp. 11-38). Oslo: Novus Press.
  • Kristiansen, T. 2009. The Macro-level Social Meanings of Late Modern Danish Accents. Acta Linguistica Hafniensia 41, 167-92.
  • Madsen, L.M., M.S. Karrebæk & J.S. Møller (Eds.). 2016. Everyday Languaging: Collaborative Research on the Language Use of Children and Youth. Trends in Applied Linguistics. Mouton de Gruyter.

Extending SLICE. From top-down to bottom-up, from survey to social media corpora

Stefan Grondelaers, Radboud University Nijmegen

Over the past decade, collaborative SLICE-research has unveiled some of the main determinants of fluidity and change in European standard languages: in communities which have preserved their standard language ideal, new “best languages” have appeared alongside the norm variety in the form of regional and modern standards (or neo-standards, Auer 2017 & 2018). Crucially, neo-standards such as “Low Copenhagen” speech in Denmark, or Tussentaal in Flanders have been shown to be boosted by a novel sort of modern prestige which pertains to dynamism and media cool.

The bulk of experimental SLICE-research has built on listener attitudes extracted with the matched-guise technique (MGT), a tool which has been criticized as ecologically dubious, because it relies on a limited number of researcher-defined scales, and naïve, for targeting enduring, cross-contextually stable social meanings such as “modern prestige”. In addition, there is almost no “convincing evidence that would link attitudes established via quantitative social psychological testing with observable human behaviour” (Soukup 2012: 216).

In this paper, we tackle these three criticisms in an attempt to bridge the gap between the experimental and (qualitative) media strand in SLICE.

An ecologically more valid bottom-up alternative to the MGT is the enriched free response (EFR) tool. We have asked respondents in Flanders and The Netherlands to return the first three adjectives which spring to mind in reaction to labels or recorded samples of the varieties in their repertoire, and we have enriched these keywords with valence and vector-based synonymy measures to extract positive and negative evaluation dimensions from them. Correspondence analysis is then used to visualize the correlations between these evaluative dimensions and the investigated stimuli. This new tool advances our knowledge of standard language dynamics, by laying bare the dimensionality and conceptual content of competing language ideologies and their prestige reference points. The EFR-data show us that the modern prestige-dimension which was identified as the major driving-force for a number of European neo-standards is not a cross-contextually stable social meaning, because it assumes quite different guises across varieties in the
investigated communities. An important question to ask in this respect is whether the modern prestige attributed to ethnolectal ways of speaking – notable the Moroccan accent of Belgian and Netherlandic Dutch – is the same as the “yuppie dynamism” associated with modern Copenhagen speech.

The second part of the talk zooms in on the nature of the causal relation between language evaluation and production: is positive evaluation really a driving-force for language change? We study the diffusion of Citétaal, a Flemish multi-ethnolect, on the basis of a corpus of tweets. While modern prestige is a hearer category – to the extent that whether or not a speaker is considered cool, macho or dynamic depends on the listener –, speakers can of course try to enact a modern prestige personality (a process known as “stylisation”, Coupland 2007), which listeners may subsequently interpret as such. If we can prove that speakers who stylise themselves as somewhat provocative, streetwise, and cool dudes and gals manifest a higher preference for the Cité-features in the enactment of this style, then it is a specific social meaning which triggered the non-standard features.


  • Auer, P. 2017. The neo-standard of Italy and elsewhere in Europe: Theoretical and empirical studies on the restandardization of Italian. In M. Cerruti, C. Crocco & S. Marzo (Eds.), Towards a new standard: theoretical and empirical studies on the restandardization of Italian (pp. 365-374). Boston: Walter de Gruyter.
  • Auer, P. 2018. The German neo-standard in a Europa context. In G. Stickel (Ed.), National language institutions and national languages. Contributions to the EFNIL Conference 2017 in Mannheim. Budapest: Research Institute for Linguistics, Hungarian Academy of Sciences.
  • Coupland, N. 2007. Style: Language Variation and Identity. Cambridge: Cambridge University Press.
  • Soukup, B. 2012. Current issues in the social psychological study of ‘language attitudes’: constructionism, context, and the attitude–behavior link. Language and Linguistics Compass 6(4), 212–224