Most common Dutch words: top 100, plus nouns with de or het

The ranking on this page was computed on 2026-10-31 from SUBTLEX-NL, a word-frequency database that Ghent University’s Center for Reading Research built from about 44 million words of Dutch film and television subtitles (Keuleers, Brysbaert and New, 2010). Nothing here comes from PolyYapp’s own word list; the method is described in full below the table.

What are the 100 most common Dutch words?

The 100 most common Dutch words in SUBTLEX-NL are almost all grammar words: pronouns such as ik, je and het, the verbs zijn, hebben, gaan and kunnen, prepositions such as van, in and op, and short adverbs such as niet, nog and wel. Together they make up about 64% of all the words in the corpus. The list counts dictionary forms, so zijn includes is, ben, was and waren. Only one word that is mainly a noun makes the cut, and it is shown with its article: de man. Haar and weg also have noun senses (het haar, hair; de weg, road), but those are the smaller share of each count.

The table ranks each word by how often it occurs per million words of subtitles. The English column covers the senses that account for most of each count, not every meaning in a dictionary.

How was this list made?

This ranking of the most common Dutch words was built from the SUBTLEX-NL lemma file, which groups every inflected form under its dictionary form (the lemma) and tags each one with a part of speech. According to the Center for Reading Research, the tags were assigned automatically with TADPOLE, software from Tilburg University. The corpus itself is 43,729,424 words from 8,443 subtitle files, collected between 10 and 19 March 2009, and the FAQ adds that the subtitles cannot be shared because the researchers do not hold their copyright.

The steps, in order:

1. Source file. SUBTLEX-NL.master.cd-above2.txt, the version limited to lemmas that occur in more than two subtitle files (89,564 entries, as the download page labels them; each entry is one lemma with one part of speech). The Center’s own site was offline on 2026-10-31, so the file came from the Internet Archive’s copy of its download page.
2. One row per spelling. Counts for the same lemma under different parts of speech were added together, so dat the pronoun and dat the conjunction are one row. Four rows join unrelated words that share a spelling: zijn (to be; his), haar (her; hair), waar (where; true) and weg (away; road). The table gives both meanings, and in each the first sense is the larger share.
3. Contractions folded in. The spoken spellings ‘t, m’n and z’n were counted as het, mijn and zijn. D’r was left as its own entry, because it can stand for either haar or er.
4. Split plurals rejoined. Where the tagger left a plural as its own noun lemma (mensen, mannen, dagen, levens, dingen, jongens, vrienden, meisjes, ideeën, landen), it was counted under the singular. Other tagging slips were left as published.
5. Per million. Each count was divided by the 43,112,394 tagged words in the lemma file and multiplied by one million.

Word classes in the tables are plain English for learners. SUBTLEX-NL’s own tags follow the Dutch CGN tagset, which files hier, daar, er and waar under pronouns and mee under prepositions. The archived pages ask anyone who uses the data to cite the 2010 paper and state no other licence terms; this page reproduces only the top ranks, with its own translations.

Which Dutch nouns are most common, and do they take de or het?

The most common Dutch nouns in SUBTLEX-NL are man, mens, dag, vrouw, tijd, jaar, leven, vriend, huis and jongen. Of the 30 most frequent nouns, 18 take de and 12 take het. Two of the het-words are diminutives ending in -je (meisje, beetje) and one is a verb used as a noun (leven), and both patterns always take het. For the other nine there is no shortcut: het jaar, het huis and het geld have to be learned with their article.

This table counts noun uses only, and it leaves out entries the tagger mislabelled as nouns: een, oké, hè, the expletive verdomme and the abbreviation mr.

Why are the most common Dutch words mostly grammar words?

The top of any Dutch frequency list is grammar words because articles, pronouns, prepositions and auxiliary verbs turn up in almost every sentence, whatever it is about. Nouns and adjectives depend on the topic, so their counts are spread across tens of thousands of words. In SUBTLEX-NL that gives a steep curve: ten words cover more than a quarter of the corpus, and each further thousand adds less than the one before.

The coverage figures below were calculated from the same lemma file, using the method described above: the share of all 43.1 million tagged words taken up by the most frequent words.

Coverage is not comprehension. The top 100 dictionary forms account for about two words in three of running subtitle text, counting all their inflected forms, but the remaining third holds almost every word that says what a sentence is about: the nouns, most main verbs, the names. The figures say where the effort pays off first, not how much Dutch you will follow.

What does a subtitle frequency list get wrong?

A subtitle list measures what people say on screen, which is close to conversation but not the same thing. The Center for Reading Research’s FAQ states that 5,966 of the 8,443 subtitle files were translations of American films and television series. That shows in the ranks: vermoorden (to murder) occurs about as often as drie (three), and schieten (to shoot) is 211th.

Four other limits are worth knowing before you treat the ranking as exact:

• Spoken, not written. SUBTLEX-NL does not measure letters, forms or news, so it says little about the formal Dutch in a letter from the gemeente.
• Automatic tagging. The parts of speech were assigned by software, and it slips: een was tagged as a noun about 62,500 times, so treat ranks a few places apart as roughly equal.
• Dated material. The subtitles were collected in March 2009, so words that became common after 2009 are undercounted.
• Not the exam list. SUBTLEX-NL is not the list the Staatsexamen Nt2 names. The College voor Toetsen en Examens (CvTE) builds its tasks on A Frequency Dictionary of Dutch (Tiberius and Schoonheim, 2013) plus its own supplementary list (CvTE word list).

The case for subtitles is that they track how quickly people recognise a word. The Center reported that SUBTLEX-NL frequencies explain up to 10% more variance in how fast and how accurately people recognise Dutch words than CELEX, the frequency database that was the previous standard for Dutch (general information, archived).

Does PolyYapp teach the most common Dutch words first?

No. PolyYapp does not teach words in frequency order. Each lesson is filled first with words that are due for review, and any remaining slots go to new words picked at random from its curated list of about 3,900 Dutch words (as of October 2026); if the lesson is still short, already-mastered words fill the rest, and the order is then shuffled. It has no frequency ranking, no CEFR levels and no topic decks, so a new learner can meet a rare word before a common one. That is a real gap if you want a frequency-first path, and the list above is one way to supply it yourself.

What PolyYapp does once a word comes up is show its card: the article (de or het) on nouns where the dictionary has it, forms such as the plural or the past participle, and text-to-speech audio in a Netherlands Dutch voice. The Add Word screen searches Dutch words (not English meanings), so you can look up a word from this list and add it to your own vocabulary, where it joins the same review schedule. PolyYapp is free today, runs on iPhone and iPad (and on Apple silicon Macs as the iPad app), needs an internet connection, and teaches vocabulary only: no grammar lessons and no speaking practice.

How should I use a list of the most common Dutch words?

Use a frequency list as a checklist, not a syllabus. Words such as er, wel, toch, nou and even have no one-word English equivalent, and they make sense in sentences more than on their own. The nouns are where a list helps most directly, because each one has to be learned together with de or het.

The rest of this site picks up from here:

• Which article? The rules that do work, and the patterns with exceptions, are in de or het.
• Making the words stick. Spaced repetition, self-testing and separable verbs are covered in how to learn Dutch words.
• For the inburgering exam. SUBTLEX-NL is not the exam’s list. Whether an official word list exists, what the CvTE’s frequency dictionary covers, and what to learn for the exam are in the inburgering word list guide.

The full ranking behind this page goes well past 100 words. Anyone who wants the next thousand can download the archived SUBTLEX-NL files, repeat the five steps in the method section, and get the same numbers.