Voicing Assimilation

Level 4

In Hungarian, when two consonants meet, the second one usually decides if the first one is voiced (vibrating vocal cords) or voiceless (whispered). This happens automatically in speech, even if the spelling doesn't change.

How It Works

Voicing assimilation is a fundamental phonological process in Hungarian. It is obligatory and occurs both within words (when adding suffixes) and across word boundaries in natural speech.

1. The Rule of the Second Consonant:
The voicing of a sequence of obstruents (stops, fricatives, and affricates) is determined by the last member of the group.

  • Voicing: A voiceless sound becomes voiced (e.g., k → [g]) before a voiced sound like b, d, g, z, zs, dz, dzs, gy.
  • Devoicing: A voiced sound becomes voiceless (e.g., z → [s]) before a voiceless sound like p, t, k, s, sz, cs, ty, c, f, h.

2. Iterativity (The Ripple Effect):
Voicing assimilation is iterative, meaning the voicing of the final obstruent in a cluster ripples back through all preceding obstruents. For example, in testben (in a body), the voiced b causes the t to become [d], which then causes the s to become [z], resulting in the pronunciation [tezdben].

3. The Special Case of 'v':
The letter v is unique. It can be devoiced by a following voiceless consonant (e.g., hév + től sounds like [héftől]), but it never triggers voicing in the consonant before it. For example, in hatvan (sixty), the 't' remains voiceless [t], it does not become [d].

4. The Special Case of 'h':
The letter h triggers devoicing in the consonant before it (e.g., adhat sounds like [athat]), but it does not undergo voicing itself.

5. Sonorants and the 'j' Exception:
Most sonorants (m, n, ny, l, r) are neutral; they do not trigger regressive assimilation and are not affected by it. However, the sonorant j is a special case: while it does not trigger voicing in preceding sounds, it undergoes progressive devoicing when it follows a voiceless consonant (e.g., lépj sounds like [lépch]).

6. Spelling vs. Pronunciation:
Unlike full assimilation, voicing assimilation is not reflected in spelling. You must apply the rule mentally while reading.

Examples (5)

  1. zsákban

    in a sack (pronounced [zságban])

    Word by Word: sack-in

    The voiceless 'k' is followed by the voiced 'b' of the inessive suffix. Therefore, 'k' is pronounced as its voiced pair [g].

  2. kéztől

    from a hand (pronounced [késztől])

    Word by Word: hand-from

    The voiced 'z' is followed by the voiceless 't' of the ablative suffix. The 'z' is pronounced as its voiceless pair [sz].

  3. egy nagy fa

    a big tree (pronounced [egy naty fa])

    Word by Word: a big tree

    Assimilation happens across word boundaries. The voiced 'gy' [ɟ] in 'nagy' becomes the voiceless [c] because the next word starts with the voiceless 'f'.

  4. szubjektív

    subjective (pronounced [szubjektíf])

    Word by Word: subjective

    Even in loanwords, a voiced consonant at the end of a word (like 'v') is devoiced if the speaker pauses or if the next sound is voiceless. Here, 'v' becomes [f].

  5. hatvan

    sixty (pronounced [hatvan], NOT [hadvan])

    Word by Word: sixty

    This demonstrates that 'v' does not trigger voicing. The 't' remains voiceless even though 'v' is a voiced consonant.

Common Mistakes

  • Pronouncing words exactly as they are spelled (staccato speech).

    nép-dal [nép-dal]
    népdal [nébdal]

    Learners often try to be 'too clear' by pronouncing every letter. In natural Hungarian, the 'p' must become [b] before the 'd' for the word to sound native.

  • Voicing a consonant before the letter 'v'.

    hatvan [hadvan]
    hatvan [hatvan]

    While 'v' is voiced, it is a 'one-way' consonant in Hungarian phonology. It does not cause the preceding consonant to change. Pronouncing 't' as 'd' before 'v' is a common foreign accent marker.

How It Connects

Tap a pattern to see what it builds on and what it leads to.

Voicing AssimilationLevel 4

Learn Next

Patterns that build on this one.