Digital discourse, often referred to as computer-mediated communication, has revolutionized communication practices since the advent of the personal computer (Crystal, 2011), 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, 1993), 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, 2017). 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terminals remained functional some 30 years later; the service was ultimately decommissioned in 2012 on account of outdated modems, an inability to support advancements in graphics and the earlier mass migration (in France and beyond) to the present-day internet (Mailland, 2017). 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.

Cet article, qui porte sur les messages textes en français québécois, considère cette pratique langagière comme un lieu privilégié de l’expression du vernaculaire. L’analyse situe la langue des textos dans le débat qui a cours sur la diglossie et la variation en français. S’appuyant sur une analyse de textos tirés du corpus Texto4Science, l’article explique comment cette pratique vernaculaire s’appuie, comme l’oral familier, sur une grammaire qui se distingue de façon systématique du français de référence. De l’examen des données se dégage une grammaire à la fois variable et perméable. Enfin, l’article examine l’émergence d’une scripta vernaculaire et une néographie qui révèlent un effort délibéré de représenter des variantes vernaculaires.

As other chapters in this special issue demonstrate, social media proposes widely available data for sociolinguistic analysis. Twitter is an ideal resource to implement variationist approaches regarding regional differences, features specific to gender, and metrics of social media influence. At the same time, official intervention on language use, while somewhat studied in other corpora, is less explored on Twitter. French shows a long tradition of purist and prescriptive ideologies, embodied by the Académie française in France and the Office québécois de la langue française in Québec. The injection of recommended terminology aimed to eradicate foreign influence often has questionable success rate, especially in such an informal setting as Twitter. This article thus investigates lexical variation, in particular, the implantation of official new French translations mot-dièse and mot-clic between 2010 and 2016 that are meant to replace the English word hashtag. Results corroborate previous findings on the lacklustre implantation of the prescribed terms, while also revealing that users in Québec are more inclined to adapt them. Furthermore, diffusion online reflects face-to-face patterns that is cascading spread from large urban areas to smaller cities.

Writing on social media often departs from prescriptive norms through the use of non-standard words, spellings and punctuation. Amongst these traits is the repetition of letters (e.g. <ouiiiii> for oui ‘yes’). In this study, we draw upon a corpus of over 65 million tweets from three dialects of French (Laurentian, Metropolitan and Midi) to test phonological motivations for the choice of repeated letter in a word with repetition. Using mixed-effects multinomial regression, we compare dialectal differences in whether repetition targets final consonants (silent or pronounced), word-final orthographic <e> corresponding to phonological schwa, and prosodically accented penults. We demonstrate that repetition covertly signals phonological properties. We conclude that prosody mediates morphological and phonological effects and that grapheme-to-phoneme correspondences vary between regions, thereby producing phonological patterns that writers likely did not intend to convey at the time of writing. We also propose that orthographic repetition on Twitter has two prosodic sources: the default pitch accent in French (shifted or not) and focus.

Non-standard orthography on social media provides a useful supplementary data source for sociophonetic research. Regarding an ongoing chain shift in Northern Metropolitan French nasal vowels, spellings reflecting shifted vowel targets are observed on Twitter. These non-standard spellings, e.g. avont [avɔ̃] for avant /avɑ̃/ ‘before’, provide insight into speakers’ awareness of this change and its lexical distribution. Tweets with shifted and standard spellings of 306 word forms containing the phonemes /ɛ̃/, /œ̃/, /ɑ̃/ and /ɔ̃/ were collected from an 870-million word Internet Archive corpus of French tweets from 2011–2017. Shifted spellings were found for all four vowels and 168 words. The shifted spelling rate is lower than that of comparable variables in English and is not conditioned by stress, grammatical category, frequency, or phonological context, which affect the distribution of shifted nasal vowels in speech. However, frequent words show more indications of intentional misspelling, such as repetition and capitalization of the target vowel, suggesting that some speakers are conscious of the variation and comment on it using salient words. 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.

This study promotes replication research as a methodological approach that is needed in order to compare earlier and more recent analyses of digital discourse. When much of the existing research was conducted, the primary means of communication included the use of a computer keyboard, (presumably) less bandwidth, and fewer devices. However, with an increase of the range of device types, the study of diacritics deserves another look within the Digital Media landscape. The present study examines the variable use of diacritics in synchronous (i.e., real-time) French chat discourse. We have replicated a study with different data sets from the same chat corpus, which is composed of data from a European chat server. We have also compared the data from the 2008 half of the corpus to data from the same chat channels collected in 2016 (just over 60,000 words in each half of the corpus, which included a total of 7,569 tokens that were coded). Our analysis of the 2008 corpus showed that one main finding was not the same as ours (from a different part of the 2008 corpus). Moreover, a diachronic analysis (2008 vs. 2016) revealed reversed trends between the two age-based channels (i.e., 20s vs. 50s).

Phrase-final fricative epithesis (PFFE), often indicated in informal writing with a final -h or -ch, e.g. beaucoup_h, oui_ch, is a sociophonetic variable of Hexagonal French in which utterance-final vowels give way to intense fricative-like whistles. Production research has found PFFE to be used at equal rates among men and women, and perception research has found that native French speakers perceive it to mark either formal speech or intense affect. This research furthers the special issue’s line of inquiry on French variation in forms of digital media by extending the analysis to a sociophonetic variable with a robust life on Twitter. The study compares the pragmatic value of tweets containing PFFE with previously described values and then examines interactions of gender, hostword phrasal location and lexical frequency on its realization. 96.8% of PFFE occurrences in the 2060-token corpus were classified into the pragmatic categories of Formality and Intense Affect. Results suggest that PFFE has become a salient enough sociophonetic variable that 21st-century French users represent it graphically in their tweets, however, its usage is structurally more permissive than in spoken language, signaling that it has taken on an iconic value in digital spaces.

Papers in this Special Issue fill a gap in French sociolinguistics by providing a coherent and yet diverse sample of empirical studies on a variety of structural aspects of French digital media. Collectively, their thematic focus reflects a traditional framing of language variation, well known from variationist sociolinguistics, correlating empirically observed phonological, morphological, and lexical patterns with social and linguistic variables, among them age, gender, and genre. In this commentary, I reflect on each paper’s unique contribution to recent sociolinguistic research on digital media, and the volume’s goal to examine socially conditioned variation in selected contexts of twenty-first-century digital writing in French.