INTRODUCTION
Digital discourse, often referred to as computer-mediated communication, has revolutionized communication practices since the advent of the personal computer (Crystal, Reference Crystal2011), with ever-growing effects, as smart devices proliferate in every aspect of twenty-first-century human existence. Linguists in the francophone world first took note of these new digital interactional practices in the 1980s with the onset of conversation via Minitel (Levy, Reference Lévy1993), a French-born service that consisted of a computer terminal that connected via telephone lines to remote services like chatrooms, interactive games, and purchasing platforms, years before most Americans had ever heard of the world wide web (Mailland, Reference Mailland2017). Minitel terminals remained functional some 30 years later; the service was ultimately decommissioned in 2012 on account of outdated modems, an inability to support advancements in graphics and the earlier mass migration (in France and beyond) to the present-day internet (Mailland, Reference Mailland2017). But the introduction of virtual interactional spaces – in the francophone world and beyond – had marked the beginning of a new linguistic trend in which users found themselves regularly engaged in the production of written language infused with vernacular tendencies.
As the internet specialized in the 2000s and 2010s, so did the number of sub-genres of digital discourse that populated it – chatrooms gave way to instant messengers, blogs morphed into microblogging tools (i.e. Twitter), and a multitude of social media platforms emerged, each with its own technical niche – Facebook for affinity groups, Instagram for photos, Snapchat for disappearing content, Tiktok for video – each one supported either primarily (i.e. in posts) or secondarily (i.e. in captions and comments) by the new written vernacular. The emphasis on the visual dimension saw text being fortified through the use of all sorts of iconic measures – emoticons, emoji, capital letters, repeated letters, repeated punctuation – many with an affective value seeking to embody some aspect of the would-be prosody of spoken utterances. For sociolinguists, a new data source was born – one that layered novel representations of vernacular with (extra)linguistic affective flair, was accessible to users 24/7 and minimized the observer’s paradox (Labov, Reference Labov1972), a common limitation of traditional sociolinguistic inquiry. The big looming question: would the variation present in digital vernacular signal the same kinds of things about speakers and interactions as the variation in analog vernacular?
Early sociolinguistic inquiries into the genre of digital discourse first sought to establish if users encoded their digital vernacular with the same kinds of non-standard elements as they did their speech, and if the parameters that conditioned these types of variation in digital vernacular were the same as in speech. The results were uniformly affirmative, at least in terms of digital speech on Twitter. Eisenstein (Reference Eisenstein2015) and Tatman (Reference Tatman2015) reported that pragmatic and sociolinguistic variables seemed to impact spoken and written variation in similar ways, and that the same contextual factors that favoured a variant in speech could also favour it in writing (Eisenstein, Reference Eisenstein2013). On the representation of variation, Tatman (Reference Tatman2015) documented the use of several non-standard orthographies associated with a particular language variety, alongside the appearance of multiple dialectal features in a single tweet. Eisenstein (Reference Eisenstein, Boberg, Nerbonne and Watt2018) revealed the diffusion of non-standard orthographies representing linguistic variants in American-based tweets to follow the same trajectory as many socially-conditioned innovations in speech: coastal urban areas > inland urban areas > rural areas. The novel and non-standard orthographies described by Eisenstein (Reference Eisenstein2013, Reference Eisenstein, Boberg, Nerbonne and Watt2018) and Tatman (Reference Tatman2015), also known as “neographies” (Roche et al., Reference Roche, Verine, Lopez, Panckhurst, De Sterck, Linder, Nava, Sánchez Ibáñez and Torres del Rey2016; Cougnon et Beaufort, Reference Cougnon, Beaufort, Granger and Paquot2009), emerge therefore as the beacons of this genre, as speakers may use them in unestablished but decipherable ways to encode social information, dialectal information, affective information, or some combination of them all (Tatman, Reference Tatman2015; Lamontagne & McCulloch, Reference Lamontagne and McCulloch2022; Law, Reference Law2022).
The goal of this special issue is to examine the representations of a diverse collection of socially conditioned variables from twenty-first-century varieties of French as they appear in different types of digital media. To do this, the articles will each explore the variation of one or more structural components in French – morphosyntax, lexicon, phonology, phonetics – in an effort to describe its distribution in digital settings and, when relevant, compare and contrast its distribution in spoken language. A variety of digital media types will be explored – text messages, chatrooms, Twitter – as will a multitude of French varieties – Quebec French, Metropolitan French, Midi French, European French. Given the nature of data that have been extracted and organized into digital corpora, several articles will employ one or more big data methods (i.e. data mining, predictive modelling) in their analyses.
NEW GRAMMAR, WHO DIS?
Our discussion begins with Blondeau & Tremblay’s Écrire son vernaculaire: variation et normes communautaires dans les messages textes en français québécois , an article which lays the groundwork for our special issue’s exploration of the digital vernacular by first establishing the genre’s structural independence from the written standard and colloquial speech. Focusing on a collection of Quebecois text messages taken from the Texto4Science corpus, the authors investigate the structural tendencies of text messages to determine if the genre should be conceived of as a hybridized system of written Standard French with colloquial influences or an independent system with its own structural tendencies and community norms. Considering syntactic, morphosyntactic, lexical, and phonological variables, they find evidence for a grammar that is systematically distinct from Standard French and demonstrates both variability and permeability. Among the defining practices of this new genre of spontaneous written vernacular is the use of a variable, affect-driven neography, in which users deliberately engage in creative orthographic representations when rendering their speech on screen.
WHAT IF I TOLD YOU THAT WORD CHOICE WAS REVEALING
Zsombok’s Official new terms in the age of social media: the story of hashtag on French Twitter explores lexical variation on Twitter in two corners of the francophone world – Quebec and France – via an analysis of the implantation of two new and different official French-language equivalents prescribed to replace the English word hashtag: mot-dièse (France) and mot-clic (Quebec). To do so, he examines the role of city size, tweeter gender and tweeter following size in predicting the use of the prescribed forms, revealing almost entirely divergent landscapes in Quebec and France. French tweeters seemed to display an internal purism regarding their preference for hashtag-word, showing more concern for the authenticity of the adopted variant than the fact that it comes from English. Contrastively, Quebecois tweeters, likely due to the different sociopolitical linguistic landscape they live in more generally, exhibited external purism, showing more concern for the avoidance of English than the artificial prescription of a neologized term engineered of French parts. This divergent behaviour explains why France’s use of mot-dièse declined steadily while Quebec’s use of mot-clic rose slowly in the years following their prescription.
NEOGRAPHY, NEOGRAPHY EVERYWHERE
Two articles in this issue examine instances of neography in digital vernacular that serve as a lens for various phonological phenomena. Lamontagne & McCulloch’s Phonological variation on Twitter: Evidence from letter repetition in three French dialects investigates Twitter data from three different French varieties (Laurentian, Metropolitan, Midi) to see if the use of letter repetition in non-standard writing (interpreted here as the tweeter’s intent to communicate prosody) can serve as an indirect signal to phonological properties. Through an examination of three crossdialectally variable phonological traits – pronounced and unpronounced final consonants, the realisation of final schwa and prosodic conditioning – Lamontagne & McCulloch demonstrate both the presence of dialect-based differences in the realization of prosodic letter repetitions and individual differences within each variable type, reflective of their differing status in each dialect. This creative strategy of repeating some letters and not others, they argue, gives tweeters a means to coerce French orthography into more justly conveying localized modern phonological patterns.
Similarly, Law’s Reflections of the French nasal vowel shift in orthography on Twitter examines instances of non-standard orthography in search of evidence that twenty-first-century French tweeters might be creatively using shifted spellings to signal the ongoing nasal chainshift in Northern Metropolitan French’s nasal series, in which, most famously, /œ̃ / is said to be unrounding to [ɛ̃] or [æ̃/ɑ̃], but also /ɛ̃/ is lowering to [æ̃/ɑ̃], /ɑ̃/ is backing to [ɔ̃] and /ɔ̃/ is raising to [ũ] (Nicholas, Fagyal & Carignan, Reference Nicholas, Fagyal and Carignan2019; Nicholas, Reference Nicholas2018 most recently; among others). Examining a collection of 306 creatively spelled forms taken from an 870-million-word corpus, the author finds shifted spellings for all four nasal vowels but reveals they are not conditioned by the same stress, grammatical category, frequency, or phonological contexts as in speech. Instead, he finds that frequent words show more indications of intentional misspelling, via strategies such as letter repetition (as in Lamontagne & McCulloch, this issue) and/or capitalization of the target vowel, suggesting that some speakers are conscious enough of the variation to comment overtly on it in their online exchanges. The results also contribute to an ongoing debate about a possible merger between /ɛ̃/ and /œ̃/, supporting the hypothesis of an incomplete merger where /ɛ̃/ shifts towards [ɑ̃] but /œ̃/ does not.
The final two articles of the issue examine instances of neography that, rather than representing phonological phenomena, instead serve as a lens for various social phenomena. Williams & Blattner’s The variable use of diacritics in synchronous computer-mediated French discourse: Replication research compares the variable use of diacritics in synchronous French chat discourse taken from the EuropNet chat server in 2008 and 2016. To do so, they examine the presence/absence of the French’s five diacritics: acute (é), grave (à, è, ù), circumflex (â, ê, î, ô, û), cedilla (ç) and diaeresis (ë, ï, ü), relative to their host letter, the self-reported age of the chatters, and the year the interaction took place. When comparing years, they found that the cinquantaines ‘people-in-their-fifties’ favoured the use of diacritics in 2008 when compared to younger chatters, but that it was the vingtaines ‘people-in-their-twenties’ who favoured their use in 2016, a change they theorize might have been conditioned by the normalization of newer smarter devices (e.g., smartphones, smartwatches, tablets) and the improvement of correction technologies for easy, error-free digital expression (e.g., spellcheck, predictive text, swyping, i.e. slide-texting, which allows users to swipe from one letter to the next without lifting their finger when typing, and voice recognition software).
Finally, Dalola’s #YouAreWhatYouTweetCHHH: Identity and fricative epithesis in French-language tweets takes to Twitter to track the usage of the sociophonetic variable known as phrase-final fricative epithesis (AKA phrase-final vowel devoicing) in Hexagonal French. She reports not only that the variable has become salient enough for graphical representation in tweets but that it has appeared regularly in this setting for the last 10+ years. Testing previous perceptual work that reported the variable to be associated with two different affect types – Formality and Intense Affect – she compiles a comprehensive corpus of every occurrence on Twitter over a 12-year span, classifies them as instances of Formality, Intense Affect or Other, then examines the role of tweeter gender, hostword phrasal location and lexical frequency at predicting their occurrence. Although both affect types were represented in the data, she documents a clear predominance for its usage on Twitter in situations of Intense Affect (over 90% of occurrences). Additionally, a new iconic usage of the variable is observed, which she reasons is likely motivated by Twitter’s text-centric environment, in which an epithesized fricative may appear in any position (à la emoji), rather than just phrase-finally as in spoken language.
The special issues ends with a brief outro by French sociolinguist Zsuzsanna Fagyal, who has devoted her career to tracking French variation in both traditional and digital spaces.
Competing Interests
The author declares none.
