When a Syllable Takes Stød

Stød — the brief creaky catch in the voice that English speakers often hear as a tiny "hiccup" — is one of the hardest features of Danish to predict. The good news is that it is not pure chaos. There are real, learnable regularities about where stød can land and when it disappears. The bad news, which we will be honest about, is that those regularities only get you part of the way: a residue of stød placement is simply lexical and must be memorised word by word. This page maps out the conditions that make stød possible, the grammatical endings that toggle it on and off, and the famous minimal pairs that prove it is carrying real information.

The stødbasis: what a syllable needs to carry stød

Stød is not a free-floating accent you can sprinkle anywhere. A syllable can only carry stød if it has the right phonetic material to support it — what Danish linguists call the stødbasis. There are exactly two configurations that qualify:

1. A long vowel (e.g. hus [huˀs] "house"), or
2. A short vowel followed by a voiced consonant in the same syllable. The voiced consonants that count are the sonorants and a few others: r, l, m, n, ŋ, v, d, g (e.g. barn [ˈbɑˀn] "child").

The logic is acoustic. Stød is a phonation gesture — a momentary tightening of the vocal folds — and it needs something voiced and sustainable to ride on. A long vowel gives it that runway; a short vowel plus a voiced sonorant gives it the same thing in two pieces. A syllable that ends in a short vowel plus a voiceless consonant (like kat [kʰad̥] "cat", which despite its spelling ends in a voiceless-ish stop) has no stødbasis at all, so the question of stød never even arises there.

Stød lands on stressed syllables

The second condition is about prosody. Stød only appears on a stressed syllable. Unstressed syllables — including the schwa-laden endings that pile up at the ends of Danish words — never carry stød, even when they technically contain a voiced consonant.

This is why so much of Danish word-stress and stød-placement move together. In a word like betale [beˈtʰæːlə] "to pay", the stress is on the middle syllable, and that is the only place stød could appear. The final -e is unstressed schwa and is permanently out of the running.

How endings add or remove stød

Here is where Danish grammar and Danish phonetics tangle together. Adding an inflectional ending can create a stødbasis, destroy one, or shift the picture entirely — so the same root may have stød in one form and lose it in another.

A classic pattern: a monosyllable with stød loses it when an ending adds an unstressed syllable, because the consonant that used to close the stressed syllable now begins the following (unstressed) syllable.

The reverse also happens. Some definite and inflected forms gain stød that the indefinite singular lacked, and a number of verb forms pattern this way too. There is no single ending that always adds stød or always removes it — the effect depends on how the ending reshapes the syllable. This is exactly why you cannot reliably read stød off the spelling.

The minimal pairs: bønder vs. bønner

The strongest evidence that stød is a real, meaning-bearing feature — not just an accent or a personal habit — is the existence of minimal pairs that differ only in stød. The most cited is:

These two words are spelled almost identically and have the same vowels and consonants — the only spoken difference is that bønder carries stød on the first syllable and bønner does not. To a Danish ear the two are as distinct as English "ship" and "sheep". Get the stød wrong and you have genuinely said a different word.

Other well-known pairs include hun [hun] "she" vs. hund [hunˀ] "dog", and læser "reader" vs. læser "reads" in some speakers' systems. The list is short enough that no learner ever resolves a real sentence by stød alone — context almost always disambiguates — but it is long enough to prove the feature is phonemic.

Where the rules run out

Honesty section. Everything above tells you where stød is possible and gives you strong tendencies for the common inflections — but it does not let you derive every word. Many monosyllables with a perfect stødbasis nonetheless have no stød for purely historical reasons (hun "she" has none; many loanwords have none), and the behaviour of compounds and derived words is notoriously irregular. The fully predictive "stød algorithm" that 20th-century Danish phoneticians chased does not exist as a tidy rule you can memorise on one page.

The practical consequence: treat stød as part of the word, the way you treat noun gender. Learn each new noun and verb with its stød pattern from audio, lean on the stødbasis test to rule out impossible cases, and accept that the rest is lexical. Native speakers themselves don't compute it — they store it.

Common Mistakes

English speakers tend to err in one of two opposite directions: either flattening Danish out entirely (no stød anywhere, which makes the speech sound foreign and occasionally merges minimal pairs) or, after learning that stød exists, over-applying it to every stressed syllable. The cure for both is the same: anchor stød to the stødbasis test, tie it to stress, and learn the lexical exceptions from listening rather than from spelling.