When to Use Accent 2

The overview page established what accent 1 and accent 2 are and that the contrast is lexical. This page answers the harder, more useful question: can you predict which accent a word takes? The distinguishing claim here — one almost no learner resource attempts — is yes, mostly. Pitch accent feels like an unmemorisable property you must learn word by word, but in reality it falls out of word structure with high reliability. Learn the morphological triggers and you can assign the correct accent to thousands of words you have never heard.

The starting point: accent 1 is the default, accent 2 is triggered

The cleanest way to think about the system is markedness: accent 1 is the unmarked default, and accent 2 is added by certain morphological conditions. A bare monosyllable has no choice — it is accent 1, because accent 2 physically needs a second syllable to host its dip-and-rise. Everything interesting happens when a word becomes polysyllabic: which kind of polysyllabicity decides whether the default holds or accent 2 is switched on.

💡
Don't ask "is this word accent 1 or accent 2?" as if flipping a coin. Ask "what is this word built from?" Accent 2 is a signal of structure — of a verb stem, a derived form, a historically-disyllabic root, or (above all) a compound. Read the morphology and the accent follows.

Rule 1 — Monosyllables and their definite forms keep accent 1

All one-syllable words are accent 1: bil (car), hund (dog), hus (house), sol (sun). The genuinely surprising part — and the one that distinguishes a careful account from a sloppy one — is what happens when you add an inflectional ending to such a noun. Adding the definite-article ending does not switch on accent 2. The word grows a syllable but keeps accent 1: bilbilen, hundhunden, solsolen.

This is why the overview's minimal pairs work the way they do. Anden "the duck" is accent 1 because it is the monosyllable and "duck" plus the definite -en; anden "the spirit" is accent 2 because it is the already-disyllabic ande plus -n. The accent "remembers" how many syllables the base had.

BaseDefiniteAccentWhy
bil (car)bilen1monosyllable + definite ending → still accent 1
sol (sun)solen1same
and (duck)anden1monosyllable + definite → accent 1
ande (spirit)anden2already disyllabic base → accent 2

Bilen står utanför huset.

The car is parked outside the house. — bilen keeps accent 1 even though the definite ending added a syllable.

Solen gick ner bakom bergen.

The sun set behind the mountains. — solen: monosyllable sol + -en, still accent 1.

The plural is a different story: bilar "cars" (bil + plural -ar) is accent 2, because the plural ending counts as the kind of derivation that triggers it, whereas the definite ending does not. So a single noun can show both accents across its forms — bil (1), bilen (1), bilar (2), bilarna (2) — and that split is itself predictable.

Det står tre bilar på gatan.

There are three cars in the street. — bilar (plural) is accent 2, unlike singular bil/bilen (accent 1).

Rule 2 — Most native polysyllables with non-final stress take accent 2

Native Swedish words of two or more syllables with stress on a non-final syllable are overwhelmingly accent 2: flicka (girl), sommar (summer), gata (street), kvinna (woman), pojke (boy). Many of these descend from words that were already disyllabic in Old Norse — the very class the rules above keep distinguishing from monosyllable-plus-ending words. The secondary stress lands on the final syllable, which is where the accent-2 rise is realised.

Flickan och pojken bor på samma gata.

The girl and the boy live on the same street. — flickan, pojken, gata: native disyllables, all accent 2.

Det är en vacker sommar i år.

It's a beautiful summer this year. — sommar and vacker carry accent 2.

Rule 3 — Verb infinitives and present forms take accent 2

This is one of the most reliable triggers and the most useful for production, because verbs are everywhere. The infinitive and present tense of polysyllabic verbs are accent 2: tala (to speak) → talar (speaks); handla (to shop) → handlar; öppna (to open) → öppnar. All Group 1 verbs (the -ar class — see Group 1 verbs) behave this way, and so do Groups 2–3 in their polysyllabic forms.

Jag talar svenska och hon talar norska.

I speak Swedish and she speaks Norwegian. — talar (present) is accent 2; hold the dip-and-rise across tal-ar.

Vi handlar mat på lördagar.

We do the grocery shopping on Saturdays. — handlar (present) accent 2; lördagar (plural) accent 2 too.

Kan du öppna fönstret?

Can you open the window? — öppna (infinitive) is accent 2.

The flip side: a monosyllabic finite verb form is accent 1, because it is a monosyllable. Tar (takes), ser (sees), går (goes) are accent 1; their polysyllabic infinitives or compounds are accent 2.

Rule 4 — Compounds take accent 2, with the peak on the second element

Compounding is the single biggest accent-2 generator in Swedish, and the most systematic. When two words join into a compound, the result is accent 2, with primary stress on the first element and a strong secondary stress on the last element — and the accent-2 melody stretches across the whole word, its second peak landing on that secondary-stressed final element. This is so regular that the accent itself becomes a cue that a word is a compound: a listener hears the two-peak melody and parses the word as built from parts.

Take lampskärm "lampshade," from lampa "lamp" + skärm "screen/shade." It is accent 2: primary stress on lamp-, secondary stress (and the second tonal peak) on -skärm. Crucially, even when a compound is built from two monosyllables — each of which alone would be accent 1 — the compound is accent 2. Compounding creates the polysyllabic structure that accent 2 requires.

CompoundBuilt fromAccent
lampskärm (lampshade)lampa
  • skärm
2 — peak on -skärm
solglasögon (sunglasses)sol
  • glas
    • ögon
2
tandborste (toothbrush)tand
  • borste
2
biltvätt (car wash)bil
  • tvätt
2 — though both parts are accent-1 monosyllables alone

Kan du köpa en ny lampskärm?

Can you buy a new lampshade? — lampskärm: compound, accent 2, with the second peak on -skärm.

Glöm inte solglasögonen!

Don't forget the sunglasses! — solglasögon(en): compound, accent 2 across the whole word.

Vi tog bilen till biltvätten.

We took the car to the car wash. — bilen is accent 1 (monosyllable + definite), but biltvätten is accent 2 because it's a compound. Same root bil, different accent.

💡
The compound contrast is the clearest proof that accent is structural, not lexical: bilen (accent 1) and biltvätten (accent 2) share the root bil, yet differ in accent — purely because one is a definite monosyllable and the other is a compound. If you hear the two-peak melody, you are almost certainly hearing a compound or a derived polysyllable.

Rule 5 — Most loanwords default to accent 1

Borrowed words, even polysyllabic ones, mostly take accent 1 rather than the accent 2 their syllable count might suggest. Telefon, museum, restaurang, biljett, kontor tend toward accent 1 (final-stressed loans especially). This is why accent 1 functions as the unmarked default: it is what a word "gets" when no native morphological trigger fires. The category is fuzzy — long-naturalised loans can drift toward accent 2, and there is regional variation — so treat it as a strong tendency, not a hard law.

Jag jobbar på ett kontor i centrum.

I work at an office downtown. — kontor: a loan, accent 1 despite being disyllabic.

Har du en biljett till tåget?

Do you have a ticket for the train? — biljett: a French loan, accent 1.

How predictable is it, really? An honest accounting

The synthesis above is genuinely powerful: monosyllable-default, definite-keeps-1, plural/derivation/infinitive/present switch to 2, compounds force 2, loans default to 1. With those rules you can assign the correct accent to the large majority of Swedish words from their structure alone — far better than the "just memorise each word" advice competitors fall back on.

But be honest about the limits:

  • There is genuine residual unpredictability. A handful of disyllabic words resist the rules, and some minimal pairs (like anden/anden) exist precisely because two different structures collapsed into one spelling.
  • The rules describe which accent, not how it sounds. Realisation is heavily dialect-dependent (see the overview): the Central Swedish two-peak is one timing among several, South Swedish and Gothenburg differ, and Finland Swedish has no contrast to assign at all. The morphological triggers are stable across Sweden; the melody they produce is local.
  • Pitch accent is still never written. No rule here is visible on the page; you apply them to the structure of the word, not to its spelling.

I finlandssvenska finns ingen regel att tillämpa — kontrasten saknas.

In Finland Swedish there's no rule to apply — the contrast is absent. — the prediction system presupposes a dialect that has the contrast.

Common Mistakes

❌ Giving 'bilen' accent 2 because it now has two syllables

Incorrect — the definite ending on a monosyllable keeps accent 1: bilen, solen, anden 'the duck'. Only some endings (plural, derivation) trigger accent 2.

✅ bilen — accent 1 (monosyllable + definite)

the car

❌ Assuming every polysyllabic word is accent 2

Incorrect — most loanwords (kontor, biljett, restaurang) are accent 1 despite their length. Accent 2 is triggered by native structure, not mere syllable count.

✅ kontor — accent 1 (a loan)

office

❌ Giving the compound 'biltvätt' accent 1 because 'bil' is accent 1

Incorrect — compounding forces accent 2 regardless of the parts' own accents: biltvätt is accent 2, with the second peak on -tvätt.

✅ biltvätt — accent 2 (compound)

car wash

❌ Saying the present tense 'talar' with a flat accent-1 melody

Incorrect — polysyllabic verb infinitives and present forms take accent 2: talar, handlar, öppnar all carry the dip-and-rise.

✅ talar — accent 2 (present tense)

speaks

❌ Trying to read the accent off the spelling

Incorrect — pitch accent is never written. Predict it from the word's structure (monosyllable? compound? verb? loan?), not from any mark on the page.

✅ Read the morphology, not the spelling, to assign the accent

structure predicts the accent

Key Takeaways

  • Accent 1 is the default; accent 2 is triggered by structure. A monosyllable is always accent 1, because accent 2 needs a second syllable.
  • The definite ending keeps accent 1 on a monosyllabic noun (bil → bilen), but the plural and other derivations switch to accent 2 (bilar) — so one noun can show both across its forms.
  • Native polysyllables, verb infinitives/present forms, and above all compounds take accent 2; the compound's second peak lands on the secondary-stressed final element (lampskärm, biltvätt).
  • Most loanwords default to accent 1 despite being polysyllabic (kontor, biljett).
  • The rules predict the large majority of cases from morphology — but realisation is dialect-dependent, Finland Swedish has no contrast, and the accent is never spelled.

Now practice Swedish

Reading grammar gets you part of the way. The exercises are where it sticks — free, no signup needed.

Start learning Swedish

Related Topics

  • Pitch Accent: Accent 1 and Accent 2B1Swedish's flagship feature: a tonal word accent. Most dialects contrast accent 1 (acute, a single fall) with accent 2 (grave, the famous 'two-peak' rise-and-fall) on stressed words — distinguishing minimal pairs like anden 'the duck' vs 'the spirit'. It is never written, and Finland Swedish drops it entirely.
  • CompoundingB1Swedish builds new words by fusing existing ones into a single solid word — fotbollsplan, tvättmaskin, skrivbord. Compounds are RIGHT-HEADED: the last element decides the word class, the gender, and the core meaning, while everything before it just modifies. Only the final element inflects. Master that one rule and you can parse, gender, and inflect almost any compound, however long.
  • Present Tense: Group 1 (-ar)A1The single most useful conjugation rule in Swedish: for the giant, fully regular Group 1 class, the present tense is just the infinitive plus -r (tala → talar, arbeta → arbetar, fråga → frågar). No stem change, no person endings. Because every new and borrowed verb joins Group 1, mastering this one rule unlocks the bulk of the Swedish verb lexicon.