Barbiers, Sjef, Hans Bennis, Gunther De Vogelaer, Magda Devos & Margreet Van der Ham. 2006. Dynamische syntactische atlas van de Nederlandse dialecten (DynaSAND). Amsterdam: Meertens Instituut. http://meertens.knaw.nl/sand/.

De Caluwe, Johan. 2009. Tussentaal wordt omgangstaal in Vlaanderen. Nederlandse taalkunde 14(1). 8–25. https://doi.org/10.5117/NEDTAA2009.1.TUSS339.

De Caluwe, Johan & Evelien Van Renterghem. 2011. Regiolectisering en de opkomst van tussentaal in Vlaanderen. Taal en Tongval 63(1). 61–77. https://doi.org/10.5117/TET2011.1.DECA.

Hastie, Trevor & Robert Tibshirani. 1987. Generalized Additive Models: Some Applications. Journal of the American Statistical Association 82(398). 371–386. https://doi.org/10.1080/01621459.1987.10478440.

Ooms, Miet. 2022. Brabants als overgangstaal tussen het Nederlandse Nederlands en het Belgische Nederlands. Brabants Erfgoed. http://www.brabantserfgoed.nl/page/14595/brabants-als-overgangstaal-tussen-het-nederlandse-nederlands-en-het. (7 April, 2023).

If one wishes to say ‘you are’ (singular) in Belgian Dutch, there are generally two options: the formal ‘jij bent’ from Standard Dutch, or the informal ‘gij zijt’ found in Flemish dialects. Recently, however, a special form of the second person singular has appeared in Colloquial Belgium Dutch (CBD): ‘gij bent’. This new construction blends the historically traditional ‘gij’ pronoun with the ‘bent’ conjugation from Standard Dutch. ‘Gij bent’ is divisive among Flemish language users as some find the form comes naturally, while others despise its contaminated nature.

Though ‘gij bent’ now appears in CBD, it is said to originate from northern Antwerp dialects (Ooms 2022). Curiously, however, ‘gij bent’ does not appear in any Flemish dialects found in DynaSand (Barbiers et al. 2006), a database of Dutch dialect interviews. To find a definitive answer to where ‘gij bent’ comes from and where it can be found in Flanders, I collected over 5000 posts from X (formerly Twitter) and geo-coded them using the posters’ locations found in their user profiles. In addition, a distinction was made between public statement posts and conversational posts to distinguish between formal and informal posts respectively. Finally, I estimated the gender of post authors by matching their usernames with a database of male and female first names from the Belgian government.

Through a logistic regression analysis, it is found that ‘gij bent’ is popular only in the Brabant dialect area. Additionally, the regression model shows that ‘gij bent’ is more likely to appear in posts with general statements than in conversations, which hints at a formality difference. Posts thought to come from women also feature more ‘gij bent’. To complement the regression analysis, I built a Generalised Additive Model (Hastie & Tibshirani 1987) and generated a heatmap of ‘gij bent’ usage in Flanders. This heatmap makes the findings from the regression analysis more explicit: ‘gij bent’ indeed seems to have its centre in the north of the Antwerp province and is confined only to the Brabant dialect area, though it does not reach past Brussels in the south.

It is surprising that ‘gij bent’ was able to ‘escape’ from northern Antwerp, since this area is not generally seen as a leading language area in Flanders (De Caluwe 2009: 15). I argue that while the form likely originated from the northern Antwerp dialects, it was possibly not recognised as such and was subsequently reinterpreted as an innovative, hybrid and less informal construction in CBD. The restriction of ‘gij bent’ to the Brabant area is puzzling, but might be related to the highly accelerated dialect loss specific to the area (De Caluwe & Van Renterghem 2011).