Executive verdict
Tussentaal (literally, “in-between language”) is the informal Dutch used across much of everyday life in Flanders, combining standard Dutch with features inherited from dialects. It is a normal part of the linguistic landscape that learners need to understand. [1]
The difficulty is that there is no single tussentaal accent or fixed recipe. Regional pronunciation survives, and speakers select different combinations of features. [2] Meanwhile, Belgian standard Dutch has legitimate vocabulary and grammar of its own. A difference from usage in the Netherlands does not automatically make something tussentaal. [3]
Our verdict for learners is to build a foundation in standard Dutch while deliberately training comprehension of local informal speech. Observed interactions with adult learners show that nonstandard Dutch can create real comprehension difficulties. [4] Ignoring it leaves a gap between classroom preparation and conversation.
However, standard Dutch retains practical value. VRT applies its strictest language requirements to news and current affairs, while allowing broader variation in fiction and humour. [5] Perception research also shows that judgments about tussentaal depend on accent and context. [6]
The useful goal is therefore flexibility: understand informal speech, speak clearly, and choose forms that suit your audience. You do not need to reproduce every regional feature to participate in Flemish life.
What is tussentaal? Meaning in English
Tussentaal means informal Flemish Dutch that falls between local dialect and standard Dutch. It describes a variable area of language use, rather than a separately codified language.
The Flemish government's explanation treats the boundaries as gradual. Speakers adjust their Dutch according to the situation and their relationship with the listener. Someone may use dialect with relatives, more standard language in a presentation, and tussentaal with colleagues. [7]
That makes “the Dutch actually spoken in Flanders” a useful description of the learner's problem, but an incomplete description of Flanders. It should not imply that everyone speaks the same way, or that standard Dutch exists only in textbooks.
A register, a continuum, and a contested label
The word can mislead English speakers in another way. “Interlanguage” often describes the developing system of someone learning another language. Tussentaal is also used by native speakers and can be the familiar speech they grew up with. Scholarly accounts explain that it did not disappear as a temporary stage on the way to standard Dutch. [8]
A helpful working distinction is between an accent and a bundle of grammatical choices. A Belgian accent can occur in standard Dutch. Tussentaal can additionally involve pronouns, verb forms, articles and sentence structures that differ from the standard. [2]
Similarly, Vlaams (Flemish) is too broad a label to identify the register of a conversation. Ask what someone means by it: Belgian standard Dutch, informal speech, or a particular local dialect. Those answers imply different listening demands.
For learners, the diagnostic question is practical: which part prevented understanding? Was it an unfamiliar word, a shortened familiar word, a grammatical form, or local pronunciation? Calling everything “Flemish” hides those differences and makes it harder to decide what to practise.
Tussentaal examples: what changes in spoken Dutch?
Recognisable tussentaal features include shortened word endings, the ge/gij system (informal singular “you”), dialectal article forms and diminutives with -ke (a smallness or affection suffix). They occur in varying combinations, rather than as a compulsory checklist. [9]
The Flemish government's advice confirms that ge/gij remains common in informal speech across much of Belgium and the southern Netherlands. It generally recommends je/jij (informal “you”) or u (polite “you”) for standard usage, while recognising special traditional uses of gij. [10]
A listening guide, not a conversion formula
Linguist Anne-Sophie Ghyselen describes speakers using different mixtures with different interlocutors. Even familiar features can be retained or suppressed depending on the conversation. [11]
The examples below are editorial illustrations of documented patterns, not quotations from recorded speakers. Their English meanings also serve as glosses.
| Feature | Informal example and English meaning | Standard equivalent | What to recognise |
|---|---|---|---|
| Loss of final consonants | Ik weet da nie: “I don't know that” | Ik weet dat niet | Short familiar words can lose their final consonant. [2] |
| Informal pronoun and verb | Ge zijt moe: “You are tired” | Je bent moe | Both pronoun and verb can differ, without making the relationship more formal. [8] [10] |
| Contracted question form | Komde morgen?: “Are you coming tomorrow?” | Kom je morgen? | The question can contain a contracted verb and pronoun sequence. [8] |
| Article inflection | Ne stoel: “a chair” | Een stoel | Masculine article forms can differ from standard een (“a”). [9] |
| Diminutive | Boekske: “little book” | Boekje | Recognise a familiar noun inside a different diminutive ending. [1] |
| An additional conjunction | Hoe dat ge werkt: “how you work” | Hoe je werkt | An extra dat (“that”) can appear after a question word. [9] |
| Double negation | Ik heb niemand nie gezien: “I haven't seen anyone” | Ik heb niemand gezien | Multiple negative elements can express a single negation. [8] |
Initial h can also disappear in speech, and personal or possessive forms may be inflected differently. The government's feature list explicitly warns that usage differs by region, speaker and situation. [9]
Our recommendation is to learn these as recognition patterns. When you hear a short form, try restoring the standard word before concluding that you lack the vocabulary. When you hear an unfamiliar ending, look for a familiar noun or verb inside it.
Why “just replace jij with gij” fails
The pronoun system extends beyond the subject. Informal speech can combine ge/gij with u (object “you”) and uw (possessive “your”). In that setting, hearing u does not by itself establish a formal relationship. [8]
This matters when following conversations between friends. A learner trained to interpret every occurrence of u as respectful distance may misread the relationship.
Avoid turning the table into a rule for manufacturing tussentaal. Copying isolated features without their local grammatical and social context can produce an unnatural mixture. Use recurring expressions from people you actually speak with, and ask them to explain both the standard equivalent and the setting in which they use it.
Tussentaal words: is every Belgian word nonstandard?
No. Belgian standard Dutch, informal Belgian vocabulary and disputed usages must be distinguished. A word's national distribution and its suitability for formal communication are different questions. [3]
This is where simple “Dutch versus Flemish” word lists often become unhelpful. They place regional standard words beside informal expressions without telling the learner which can safely go into a professional email.
| Word or expression | Meaning | What the consulted advice says | Learner implication |
|---|---|---|---|
| Kleedje | Dress | Standard Dutch in Belgium in this meaning; jurkje is standard across the language area. [12] | A Belgian word can be entirely appropriate without being used in the Netherlands. |
| Tas koffie | Cup of coffee | Taaladvies.net classifies tas meaning “cup” as nonstandard; kopje koffie is standard. [13] | Understand the expression, and choose the standard equivalent when the task calls for it. |
| Goesting | Desire, appetite or inclination | Taaladvies.net says its standard status in Belgium is unclear, despite widespread spoken and informal written use. [14] | Learn the meaning, but check the register rather than assuming universal formal acceptance. |
These are deliberately different classifications. The sources support neither “all Belgian words are dialect” nor “everything widely used is standard.”
The uncertainty around goesting is especially instructive. The advice records both use by standard speakers and continuing objections from other language users. [14] A reference work can acknowledge that a boundary remains disputed.
Our inference is that learners need vocabulary notes with context, not just translations. Alongside a new word, record whether it is Belgian standard usage, informal usage, a local expression, or a disputed case. Leave room to revise that note when you encounter a better source.
This also changes how to request corrections. Ask a teacher whether an expression is unsuitable for formal writing, unfamiliar outside Belgium, or simply different from their own preference. Those explanations lead to different learning decisions.
For everyday conversation, recognition has value even when you choose a different expression yourself. For a text intended for both Belgium and the Netherlands, an expression shared across the language area may be the clearest choice. Taaladvies.net explicitly presents its labels as tools for making such contextual decisions. [3]
Is tussentaal the same across Flanders?
No. Brabant-influenced forms are prominent, but there are regional and personal tussentaal repertoires. A television example cannot establish how everyone in West Flanders, East Flanders, Antwerp, Flemish Brabant or Limburg speaks.
Dialectloket describes both regional variation and the prominence of Brabant-coloured speech in entertainment. It also cautions that the extent and future direction of convergence are uncertain. [15]
The implication is to treat widely recognisable forms as a starting point, then adapt your listening practice to your actual surroundings. Exposure to Antwerp-based entertainment may help, but it does not replace exposure to colleagues in another region.
The familiar three-way diagram has limits
A peer-reviewed case study of Ieper published in 2018 found that a simple division into dialect, tussentaal and standard Dutch did not adequately describe the observed repertoire. The researchers distinguished dialect, a largely virtual standard and different intermediate groupings. [16]
That is evidence about the location and speakers studied. It is not a new national classification to memorise. Its practical lesson is that useful teaching categories can be less tidy than actual speech.
How many people speak tussentaal?
The sources reviewed do not establish a current population percentage for tussentaal alone.
The Taalunie's public report published in 2021 covered survey work conducted in 2020 and included 4,007 Flemish respondents. It reported that 52% said they could speak a dialect or something between dialect and Dutch. That combines categories and measures reported ability, so it is not a tussentaal prevalence figure. [17]
The distinction matters. Ability, habitual use, self-identification and the proportion of nonstandard features in recordings are different measurements. They should not be combined into a single popularity statistic.
Research does nevertheless document the importance of intermediate speech. A study published in 2009 discussed corpus recordings of Flemish young adults aged 18 to 24. Most used intermediate forms even when asked to converse informally in standard Dutch. [18]
That finding helps explain the classroom-to-conversation gap. It cannot establish how every age group speaks now, and it does not mean that those speakers lacked knowledge of the standard.
Why is verkavelingsvlaams controversial?
The debate concerns access, identity and authority as well as linguistic forms. Support for a common standard coexists with criticism of the idea that familiar informal speech is evidence of incompetence.
Verkavelingsvlaams (roughly, “housing-estate Flemish”) carries a dismissive social association. The Letterenhuis identifies writer Geert van Istendael as the inventor of the term and documents his opposition to tussentaal and advocacy of standard Dutch. [19]
The label therefore does argumentative work. It does not merely identify a pronunciation or grammatical feature. Describing someone's speech with it can also convey a judgment about their aspirations, education or taste.
Recognising a feature is not the same as judging a person
A perception study published in 2014 asked 80 informants to evaluate seven recordings of standard Dutch and regional tussentaal. It investigated how listeners identified varieties and which features they noticed, emphasising differences between listeners. [20]
This suggests caution when a speaker declares that a particular form is obviously standard or obviously dialect. The listener's experience and expectations also affect the classification.
A further distinction is between saying that an expression falls outside a formal norm and saying that its speaker is careless or unintelligent. The former can be useful editorial advice. The latter requires evidence that the grammatical feature itself cannot provide.
What the latest experiment adds
A study published on 16 October 2026 used 614 respondents and four manipulated versions of a government video, varying accent and tussentaal features. Unaccented standard Dutch retained the greatest prestige, but tussentaal features paired with a neutral accent were not automatically downgraded and could carry dynamic prestige. [6]
The researchers used convenience and snowball recruitment targeting highly educated native speakers, excluding language specialists. This was a contextualised evaluation experiment, not a representative survey of everyday speech. [6]
Our inference is that acceptability cannot be predicted from a list of forms alone. The setting, the speaker's delivery and the listener's expectations can change the judgment.
For learners, that supports a practical middle course: maintain access to the standard while learning to understand local speech without absorbing the contempt sometimes attached to it. A learner should be told which expression fits a task, rather than being made to feel that everyone around them speaks defective Dutch.
Tussentaal on television: what VRT actually allows
VRT does not operate a universal ban on informal Dutch, and its published policy does not treat every programme alike. News, entertainment, fiction and humour have different language expectations. [5]
Its policy gives news and current-affairs output the strictest requirements. It allows all Dutch varieties in fiction and humour, with an exception for Ketnet, and recognises that contributors with particular expertise require consideration beyond language alone. [5]
Policy and observed speech answer different questions
Research published in 2015 by Jürgen Jaspers and Sarah Van Hoof argues that mixed Dutch in Flemish screen discourse can evoke meanings associated with either standard language or vernacular speech, depending on the unfolding context. [21]
The implication for viewing is to attend to who is speaking to whom. A character's language, a presenter's scripted introduction and an interviewee's spontaneous answer serve different purposes.
A soap can provide useful exposure to informal patterns, but it is also scripted dialogue. It should not be treated as a representative recording of Flemish households. Likewise, a carefully delivered news bulletin offers a listening model with different demands from overlapping conversation.
VRT stated on 13 February 2023 that 95% of its broadcasts used standard Dutch. That is the broadcaster's historical claim. The statement does not supply enough coding detail to reproduce the percentage or convert it into the share of individual words or speakers using standard forms. [22]
There is no need to force that statement and sociolinguistic research into a numerical contest. Institutional reporting, speaker judgments and analysis of linguistic features can classify different things.
Subtitles can conceal what you need to learn
VRT's T888 subtitle stylebook, dated 2017, instructs subtitlers to replace informal pronouns and some other nonstandard forms with standard equivalents, subject to exceptions. Its examples include normalising ge/gij and diminutives. [23]
This documents an editorial practice, rather than guaranteeing how every current subtitle is produced. It makes subtitles useful for understanding the message, but potentially misleading as an exact transcript. The sound and written line may differ because of editorial policy rather than because you misheard.
Our recommendation is to use the discrepancy as an exercise. Listen first, read the subtitle, then identify whether the difference concerns a pronoun, an ending, a contraction or vocabulary. Keep the spoken form and its standard counterpart together in your notes.
Why can learners understand class but struggle outside?
A course can teach useful Dutch while leaving learners underprepared for the forms they hear locally. The problem can involve variation in familiar language, as well as missing vocabulary or fast delivery.
Lybaert and van Hest's study published in 2020 combined classroom research with recordings involving four adult learners outside class. It documented comprehension difficulties caused by nonstandard Dutch in interactions with native speakers. [4]
This is direct evidence that the problem can occur. Its small qualitative sample does not establish how often it occurs among all learners, or prove that a particular teaching method fixes it.
Recognition and imitation are different learning goals
A Ghent University project listed for 2023 to 2027 investigates adult second-language learners' attitudes towards variation and how they incorporate nonstandard forms. Its description identifies a tension between widespread nonstandard variation and the ideological superiority assigned to the standard, including in language teaching. [24]
That is a research programme, not a completed effectiveness trial. It supports taking the issue seriously without supplying a universal syllabus.
Our recommendation is to separate understanding others from deciding how to speak. A learner can recognise an informal question and answer in standard Dutch. Conversational success does not require both participants to use identical forms.
This approach also gives teachers a clear task. Instead of marking a local expression only as wrong, explain its meaning, identify the setting in which it occurs, and teach the standard alternative for tasks that require it.
Responsibility also belongs to the learner's interlocutor
Preparation for variation should not become an excuse to leave newcomers struggling. In a conversation, useful adjustments include rephrasing, reducing overlapping speech and checking which expression caused difficulty.
The learner's request can be specific: explain the unfamiliar phrase in standard Dutch while continuing the conversation in Dutch. An editorial example is Kun je dat anders zeggen? (“Can you say that differently?”). That produces more useful feedback than a general request to “speak more clearly.”
For a partner or close friend, shared practice can make this easier. Our Dutch for partners guide offers a broader framework for making Dutch part of the relationship. The particular recognition work described here can fit into ordinary conversations without turning every exchange into a correction session.
What this means for learners
Keep standard Dutch as your foundation, and make local listening a deliberate part of learning. The sequence below is our practical recommendation, not a tested prescription or a set of proficiency requirements.
If you are beginning Dutch
Use a coherent course to build sentence structure, common vocabulary and clear pronunciation. Add awareness of local forms when you encounter them, without trying to reproduce a complete regional repertoire.
Start a small listening notebook. Record the expression you heard, its standard equivalent, its meaning and the person or situation in which it occurred. This keeps a recognisable pattern attached to a real context.
Ask a teacher to help distinguish a shortened familiar word from unfamiliar vocabulary. That diagnosis determines whether your next task is listening practice or vocabulary learning.
If you can converse but keep losing the thread
Choose a short recording from a relevant Belgian setting. Listen for the message, then revisit the places where comprehension failed. Classify the problem before repeating the whole clip.
Useful categories include pronouns, reduced endings, word boundaries, additional grammatical elements and local vocabulary. Work on the pattern that repeatedly obstructs you, rather than collecting expressions simply because they sound Flemish.
For supported practice, KU Leuven describes NedBox as a platform where newcomers can practise Dutch while learning about life in Flanders. [25] It is a source of contextual language practice; the consulted description does not establish it as a dedicated tussentaal curriculum.
Use broadcast speech and informal dialogue for different purposes. Clear presentation can strengthen comprehension of structured language. Conversation can train you to handle the local forms and interactional conditions that a prepared script avoids.
If you want help from a tutor
Request practice built around actual comprehension failures. Ask the tutor to explain a spoken form, provide its standard counterpart and identify how local its use is.
When comparing Dutch tutors, look for someone familiar with your destination and daily setting. A tutor's regional knowledge is relevant to this task, but nationality alone does not demonstrate skill in explaining variation.
For speaking, adopt expressions gradually when you understand their grammar and social use. Preserve the ability to shift towards standard Dutch for a presentation, unfamiliar audience or formal written task.
If you want a book about tussentaal
The scholarly collection Tussentaal: over de talige ruimte tussen dialect en standaardtaal in Vlaanderen (“Tussentaal: on the linguistic space between dialect and standard language in Flanders”), edited by Johan De Caluwe and colleagues, was published in 2013. [26]
Its bibliographic record supports its existence and academic scope. It does not establish that it is a beginner course. Choose a resource according to the task you need help with: linguistic explanation, supported listening or practice with your own interlocutors.
What we could not verify
We could not establish a current population percentage for tussentaal alone. The survey measure discussed above combines dialect with intermediate speech, while the recording and perception studies address different questions. [17] [18] [20]
We could not independently reproduce VRT's historical broadcast percentage or establish an equivalent current figure. Its published statement supplies the claim, without the methodological detail needed for an external audit. [22]
We found no evidence in the reviewed material establishing a universal proficiency threshold for starting tussentaal comprehension, or a fixed amount of exposure that reliably produces understanding. The learner evidence establishes a difficulty; it does not quantify a universally effective remedy. [4] [24]
We also could not verify that there is a single regionally neutral spoken tussentaal norm replacing the existing standard. The grammar reference and regional research instead document variation and limits to a simple national classification. [2] [16]
The practical recommendations in this report are therefore editorial inferences from the evidence. They should be adapted to the people, places and situations in which you use Dutch.