Data on language knowledge in Brussels
The Brussels Council for Multilingualism provides this data hub to help citizens, researchers, and policymakers understand the linguistic landscape of Brussels. French remains the dominant lingua franca though gradually declining, English has surged to second place, and Dutch has rebounded after a historic low. Alongside these three primary contact languages, Brussels is home to vibrant communities speaking Arabic, Spanish, Italian, Portuguese, Turkish, Romanian, and over 100 other languages across the city. The hub brings together two complementary sources: the VUB/BRIO Taalbarometer, a representative scientific survey conducted five times since 2001, and language exposure maps drawn from aggregated social media data across the 19 communes.
Explore the Data
Taalbarometer Survey Findings
A representative scientific survey of 2,500 Brussels residents conducted across five editions since 2001 by VUB/BRIO. On the survey page, you will find interactive charts tracking 20+ years of contact language trends (French, English, Dutch), multilingual combination profiles, home language diversity, generational shifts, and language use in public institutions.
Mapping languages in Brussels using social media data
Using anonymized social media usage, we map language presence across the Brussels-Region and its communes. On the map page, you can explore geographic maps, filter by individual languages or communes, analyze local rankings, and discover the territorial footprint of over 50 languages.
Understanding the Data Sources
Each source illuminates a distinct dimension of multilingualism: the Taalbarometer measures self-reported proficiency and demographic shifts, while the social media map visualizes real-time digital presence and localized language exposure across the 19 communes.
| Taalbarometer (VUB / BRIO) | Social Media Language Exposure | |
|---|---|---|
| Methodology | A representative sample-based survey conducted every 5–6 years since 2001 (latest edition TB5 in 2024/2025). Measures self-assessed proficiency ('good to excellent') and captures language practice across private, professional, and civic spheres. | An analysis of aggregated and anonymized active user profile data and language settings across social media platforms (retrieved in January 2026) in the Brussels-Capital Region, mapped across all 19 communes. |
| Strengths | Robust representative sample (2,500+ adults), 23-year longitudinal comparative baseline, detailed demographic cross-referencing by age, education, and origin. | Very large active digital sample, high geographic granularity across all 19 communes, high sensitivity to minor language communities. |
| Limitations | Subjective self-assessment of proficiency; limited sample resolution for very small language communities. | Overstates English due to default device fallback settings and the heavy dominance of English content on social media. Conversely, may understate languages primarily used in professional settings versus leisure/family/social contexts given the personal nature of Facebook and Instagram. Not an official census or test of proficiency. |