/ instituut voor de Nederlandse taal / Taalportaal
bibliography glossary about
  • Dutch
  • Frisian
  • Saterfrisian
  • Afrikaans
Show all
    Final devoicing
    quickinfo

    In Dutch, voiced obstruents are generally not found in syllable-final position. As a consequence, phonologically voiced obstruents occurring syllable-finally are devoiced (Zonneveld 1994,2007; Booij 1995). The resulting alternations are given in table 1:

    Table 1
    Word-initial, intervocalic, syllable-initial Word-final, syllable-final
    Voiced obstruent /b, d, v, z, ɣ/ → [b, d, v, z, ɣ] /b, d, v, z, ɣ/ → [p, t, f, s, x]
    Voiceless obstruent /p, t, k, f, s, x/ → [p, t, k, f, s, x] /p, t, k, f, s, x/ → [p, t, k, f, s, x]

    This process is known as final devoicing (or Auslautverhärtung). The following examples show why Dutch exhibits final devoicing, rather than a rule that voices obstruents when realized between two sonorant segments:

    1
    hand /hɑnd/ [hant] hand vs. handen /hɑndən/ [hɑndən] hands
    kant /kɑnt/ [kɑnt] side vs. kanten /kɑntən/ [kɑntən] sides
    huis /hœyz/ [hœys] house vs. huizen /hœyzən/ [hœyzən] houses
    kous /kaus/ [kaus] stocking vs. kousen /kausən/ [kausən] stockings

    The singular form of the stems hand and huis surfaces with a voiceless obstruent, whereas the plural form surfaces with a voiced obstruent. In the examples kant and kous, however, both the singular and plural are realized with a voiceless obstruent. If Dutch had a voicing rule across the board, then the plurals of kant and kous would have to surface with voiced obstruents, just like the plurals of hand and huis, yet they surface as voiceless. Therefore, it seems that syllable-final obstruents have to be voiceless in Dutch. Since /d/ and /z/ are in syllable-final position in hand and huis, they are devoiced, whereas in the plural forms handen and huizen, /d/ and /z/ are in onset position and thus need not be devoiced.

    Final devoicing does not only apply to simple codas, but also to complex codas; alternating forms occur when the final obstruent of a stem is followed by a vowelless suffix:

    2
    krabben /krɑbən/ to scratch vs. krabt krab-t /krɑbt/ [krɑpt] scratch-PRS.3SG
    broeden /bruden/ to breed vs. broeds broed-s /bruds/ [bruts] broody

    As a result, syllable-final (complex) codas in Dutch are usually phonetically voiceless. Note, however, that matters are more complex in compounds in which the quality of the word-final consonant of the first constituent interacts with that of the first segment of the second constituent.

    extra
    The feature ±voicein Dutch

    In Dutch, there exists an opposition between voiced and voiceless plosives, and voiced and voiceless fricatives. The feature±voiceis thus distinctive in the class of obstruents. No opposition exists between voiced and voiceless sonorants: all sonorants are phonetically voiced. The feature ±voice is thus not distinctive in the class of sonorants.

    Dutch has the following obstruents: /p/, /t/, /k/, /f/, /s/, /x/, /b/, /d/, /v/, /z/, /ɣ/, /h/. These can be subdivided according to voicing:

    • -voice: /p/, /t/, /k/, /f/, /s/, /x/ ,
    • +voice: /b/, /d/, /v/, /z/, /ɣ/, /h/ .

    The glottal fricative /h/ has no voiceless counterpart; furthermore, the voiceless velar plosive /k/ has no voiced counterpart in the native vocabulary, as /g/ only occurs in loanwords:

    i
    goal /gol/ goal
    oregano /ɔregano/ oregano

    The following list shows minimal pairs for plosives and for fricatives:

    ii
    paard /pard/ [part] horse vs. baard /bard/ [bart] beard
    pak /pɑk/ package vs. bak /bɑk/ vat
    pit /pɪt/ kernel vs. bit /bɪt/ bridle bit
    paal /pal/ pole vs. baal /bal/ bale
    tak /tɑk/ branch vs. dak /dɑk/ roof
    tik /tɪk/ tap vs. dik /dɪk/ fat
    top /tɔp/ peak vs. dop /dɔp/ button
    taal /tal/ language vs. daal /dal/ decline
    c /se/ <c> vs. zee /ze/ sea
    faal /fal/ fail vs. vaal /val/ pale

    For some Dutch dialects, amongst others Standard Dutch, the voicing contrast in fricatives appears to be less stable than the voicing contrast in plosives. Many speakers do not make a distinction between /x/ and /ɣ/, and between /f/ and /v/ in word-initial position. The distinction between /s/ and /z/ appears to be the most stable one. In cases of neutralization, both the voiced and the voiceless fricative are realised as voiceless (Collins and Mees 1981; Van de Velde et al. 1996; Kissine et al. 2003). (See also topics about velar fricatives /x/ and /ɣ/ and labiodental fricatives /f/ and /v/.)

    Status of /g/ in Dutch
    Whether the voiced velar plosive /g/ is a phoneme of Dutch is subject to some debate, as it only occurs in loanwords; however, there are some minimal pairs, e.g. the native word kool /kol/ cabbage versus the loanword goal /gol/ goal (cf. topic velar plosives /k/ and /g/).

    Phonetics of the voicing contrast

    Dutch is a so-called pre-voicing language: in word-initial positions, the voicing contrast in stops/plosives rests upon voice onset time (VOT; -70 ms for voiced, 20 ms for voiceless, cf. Lisker and Abramson 1964; Slis and Cohen 1969; -80 ms for voiced, 0-25 ms for voiceless, cf. Kager et al. 2007). So, Dutch distinguishes between phonetically voiced and phonetically voiceless plosives. In this respect, Dutch (together with Yiddish and Afrikaans) differs from the other Germanic languages, which are said to be aspiration languages, i.e. they have a contrast between phonetically voiceless unaspirated and phonetically voiceless aspirated obstruents (Kager et al. 2007; Van Alphen 2007; Kerkhoff 2007).

    Dutch also differs from other Germanic languages in that the VOT distinction mentioned above does not hold for all obstruents but primarily for stops. In Dutch fricatives, the phonetic correlates of the contrast are somewhat different from those in plosives: while they are traditionally labelled as +voice, they do not always have phonetic voicing (Slis and Van Heugten 1989). Phonetic voicing thus is not a distinctive feature of phonologically +voice obstruents; yet, as reported in Slis and Cohen (1969), Slis and Van Heugten (1989) and Kissine et al. (2003), there is a significant difference in duration: voiceless (or fortis) fricatives have a significantly longer duration than voiced (or lenis) fricatives. None of the experiments, however, found significant duration differences for the velar fricatives /x/ and /ɣ/.

    Distribution of voiced and voiceless fricatives

    There are some restrictions on the distribution of voiced versus voiceless fricatives, which do not hold for voiced versus voiceless plosives. Voiced fricatives can only occur after A-class vowels, whereas voiceless fricatives can only occur after B-class vowels:

    i
    knuffel /knʏfəl/ hug but *kneufel /*knøfəl/
    heuvel /høvəl/ hill but *huvvel /*hœvəl/

    This restriction does not hold for plosives. Both voiced and voiceless plosives can occur after A-class vowels and B-class vowels:

    ii
    ratel /ratəl/ rattle
    radar /radɑr/ radar
    adder /ɑdər/ adder
    otter /ɔtər/ otter

    For more information see the topic on general co-occurrence restrictions in rhymes.

    extra

    The spelling of singular forms with final fricatives does not represent underlyingly voiced forms: underlying /z/ of huis is written as <s> rather than <z>. Yet the spelling of the singular forms with final plosives represents the underlying forms: underlying /d/ of hand is written as <d>, rather than reflecting the pronunciation with phonetic [t]. (See topic about Abstract phonological forms in Dutch orthography.)

    readmore
    [+]Morphological aspects in final devoicing

    Final devoicing in compounds
    In compounds, the (non-)application of final devoicing depends on the first segment of the second constituent in the compound.

    If this first segment is a sonorant consonant or a vowel, final devoicing applies:

    3
    handrem hand-rem /hɑnd.rɛm/ [hɑnt.rɛm] handbrake
    ooglapje oog-lapje /oɣ.lɑpjə/ [ox.lɑpjə] little eyepatch
    goudader goud-ader /xɑud.adər/ [xɑut.adər] gold vein
    huidarts huid-arts /hœyd.ɑrts/ [hœyt.ɑrts] dermatologist

    There are, however, some counterexamples:

    4
    tandarts tand-arts /tɑnd.ɑrts/ [tɑndɑrts]   or [tɑnt.ɑrts] dentist

    Kerkhoff (2007) suggests that this inconsistency might be the result of a frequency effect: the combination of tand tooth and arts doctor is much more frequent than the combination of the the two words huid skin and arts doctor. Furthermore, the syllabification of [tɑndɑrts] might be [tɑn.dɑrts], rather than [tɑnd.ɑrts] as a result of a possible loss of transparent morphological structure.

    Matters are more complex if the second element begins with an obstruent: if the second element starts with a plosive, it depends on the voicing of that plosive whether final devoicing applies: if it is voiceless, the whole consonant cluster surfaces as voiceless. Yet if it is voiced, no final devoicing occurs, and the consonant cluster is voiced. This process has been referred to as regressive voicing assimilation. Furthermore, if the second constituent starts with a fricative, the voicing specification depends on the voicing quality of the coda consonant.

    In cases where a geminate consonant could arise in compounding (handdoek towel, cf. also morphological topics about compounding compounding), degemination takes place, since geminate consonants are not allowed in Dutch:

    5
    handdoek hand-doek /hand.duk/ [han.duk] towel

    The influence of suffixes on final devoicing
    Nominal stems surfacing with suffixes behave in various ways with regard to final devoicing, depending on the suffix attached to the stem. There are suffixes that trigger final devoicing and others that do not.

    Suffixes that do not trigger final devoicing are the consonant-initial past tense suffix –de (Booij 1977; Booij 1995; Ernestus and Baayen 2003), due to regressive voicing assimilation, and all Dutch vowel-initial suffixes, with the exception of –aardig and –achtig. These suffixes are generally referred to as cohering suffixes. Some examples are given in table 2:

    Table 2
    -air miljard /mɪljɑrd/ [mɪljɑrt] billion miljardair [mɪljɑrdɛr] billionaire
    -de (PST) krab /krɑb/ [krɑp] scratch krabde [krɑbdə] scratched
    -e lief /liv/ [lif] sweet, dear lieve [livə] dear
    -elijk dood /dod/ [dot] dead dodelijk [dodələk] lethal
    -en (PL) hand /hɑnd/ [hɑnt] hand handen [hɑndən] hands
    -en (INF) lees /lez/ [les] I read lezen [lezən] read
    -ig rood /rod/ [rot] red rodig [rodəɣ] reddish

    The vowel-initial suffixes in this list syllabify with the stem, so that the stem-final consonant surfaces in the onset position of the syllable containing the suffix, and therefore final devoicing does not apply.

    The suffixes triggering final devoicing are all Dutch consonant-initial suffixes (except the past tense suffix –de), as well as –aardig and –achtig (Booij 1977; Booij 1995; Ernestus and Baayen 2003). These suffixes are generally referred to as non-cohering suffixes. Some examples are given in table 3:

    Table 3
    affix stem derivation
    -aardig goed /gud/ [gut] good goedaardig [gutardəx] good-natured
    -achtig rood /rod/ /rot/ red roodachtig [rotaxtəx] reddish
    -loos draad /drad/ [drat] wire draadloos [dratlos] wireless
    -je hond /hɔnd/ [hɔnt] dog hondje [hɔntje] little dog

    The suffix -(e)nis behaves ambivalently with respect to final devoicing. In some cases, the suffix does trigger final devoicing, whereas in other cases it does not:

    Table 4
    affix stem derivation
    -(e)nis /beɣrav/ [bəɣraf] begrafenis [bəɣrafənɪs] funeral
    /vərbɪnd/ [vərbɪnt] verbintenis [vərbɪntənɪs] alliance, commitment
    /ɛrv/ [ɛrf] erfenis [ɛrfənɪs] inheritance
    /hɛrɛiz/ [hɛrɛis] herrijzenis [hɛrɛizənɪs] resurrection
    [+]Incomplete neutralization: phonetic aspects of final devoicing

    Phonetically speaking, the neutralization of voicing in final obstruents is not a complete neutralization. Production and perception experiments have shown that the phonetic neutralization between underlyingly voiceless obstruents /p, t, f, s, x/ and devoiced obstruents /b, d, v, z, ɣ/ is phonetically incomplete, i.e., phonetically, they are not completely identical. Ernestus and Baayen (2006) found that the release noise of underlyingly voiced word-final plosives is significantly shorter than that of underlyingly voiceless word-final plosives, and that codas with labiodental or alveolar fricatives are significantly longer with respect to the underlyingly voiced fricatives than the underlyingly voiceless fricatives. Warner et al. (2004) also found a significant difference in vowel length: vowels preceding underlyingly voiced obstruents are longer than vowels preceding underlyingly voiceless obstruents.

    Perception experiments have shown that listeners are able to distinguish between voiceless and devoiced segments (Ernestus and Baayen 2006, 2007; Warner et al. 2004): when presented a (pseudo-)word (recorded by a native speaker of Dutch) with either a voiceless or a devoiced final obstruent, listeners chose the correct past tense suffix /-də/ or /-tə/ at a significantly above-chance level, and they chose the correct form of the minimal pair when asked to at a significantly above-chance level. The authors claim that this has to do with the minor phonetic differences between ‘true’ voiceless segments and devoiced segments.

    extra

    There is some debate whether incomplete neutralization exists at all and what the implications are for phonological theory (see Kharlamov 2012 for a comprehensive overview of the studies on this subject and discussion of experimental settings; for studies on Dutch that did not find incomplete neutralization, see Jongman 1998; Baumann 1995). In general, the reported incomplete neutralization effects in Dutch are (extremely) small (e.g. a difference in vowel duration of 3.5 ms before voiceless and devoiced obstruents), varying between studies and task dependent.

    It has been claimed that the existence of a phenomenon such as incomplete neutralization posits a serious threat for formal theoretical phonology since an apparent three-way surface distinction of segments (voiced, voiceless and devoiced segments) is assumed incompatible with the two-level model (underlying representation vs. surface representation) of standard phonological theory combined with just one underlyingly voiced and one underlyingly voiceless segment. However, Van Oostendorp (2008) presents a theoretical account for incomplete neutralization based on Turbidity Theory (Goldrick 2001), a branch of Optimality Theory, which can capture a possible three-way distinction of surface representations while maintaining two underlying segments (a voiced and voiceless segment).

    [+]Exceptions to final devoicing

    There are a few lexical exceptions to final devoicing. The adjective bijdehand [bɛidəhɑnt] bright, sharp is derived from the noun hand /hɑnd/ [hɑnt] hand, with an underlyingly voiced obstruent. In the inflected variant of this adjective, where the adjectival suffix –e follows the underlyingly voiced obstruent, we would expect this obstruent to surface as [d], as bijdehande [*bɛidəhɑnde] (see also The influence of suffixes on final devoicing). This is, however, not what we find: the word is pronounced with a voiceless obstruent, as bijdehande [bɛidəhɑnte]. In the case of the comparative form, the voiceless obstruent even shows in the spelling bijdehanter brighter, sharper

    Also, the verb statten [stɑtən] to go shopping, which is derived from the noun stad [stɑt] city, displays a similar behaviour. The noun stad has the plural form steden /stedən/ [stedən], which indicates that at least historically, the noun must have had an underlying /d/ in its representation. We thus expect the verb derived from it, to surface with [d] as well, since the infinitival suffix does not trigger final devoicing (see also The influence of suffixes on final devoicing above); yet the verb surfaces with a voiceless [t].

    Booij (2002) suggests that speakers of Dutch no longer associate the word bijdehand bright, sharp with the noun hand hand. Therefore, there is no reason to assume an underlying voiced plosive (other than the spelling with <d>). For statten to go shopping, he assumes that the plural noun steden cities is no longer computed on the basis of the singular stad city, but rather that both singular and plural are stored independently. There is thus no reason for assuming underlying /d/ for stad (apart from the spelling), and the verb statten is derived from the singular stad with an underlying voiceless plosive.

    References
    • Alphen, Petra M. van2007Prevoicing in Dutch initial plosives: production, perception, and word recognitionVoicing in DutchAmsterdam/PhiladelphiaJohn Benjamins Publishing Company
    • Baumann, Monika1995The production of syllables in connected speechUniversity of NijmegenThesis
    • Booij, Geert1977Dutch morphology. A study of word formation in generative grammarLisse / DordrechtThe Peter de Ridder Press / Foris Publications
    • Booij, Geert1977Dutch morphology. A study of word formation in generative grammarLisse / DordrechtThe Peter de Ridder Press / Foris Publications
    • Booij, Geert1995The phonology of DutchOxfordOxford University Press
    • Booij, Geert1995The phonology of DutchOxfordOxford University Press
    • Booij, Geert1995The phonology of DutchOxfordOxford University Press
    • Booij, Geert2002The balance between storage and computation in phonologyNooteboom, Sieb, Weerman, Fred & Wijnen, Frank (eds.)Storage and computation in the language facultyDordrechtKluwer115-138
    • Collins, B. & Mees,I1981The Phonetics of English and DutchLeiden/Boston/KölnBrill
    • Ernestus, Mirjam T.C. & Baayen, R. Harald2003Predicting the unpredictable: Interpreting neutralized segments in DutchLanguage795-38
    • Ernestus, Mirjam T.C. & Baayen, R. Harald2003Predicting the unpredictable: Interpreting neutralized segments in DutchLanguage795-38
    • Ernestus, Mirjam T.C. & Baayen, R. Harald2006The functionality of incomplete neutralization in Dutch: the case of the past tense formationLaboratory PhonologyBerlinMoulton de Gruyter27-49
    • Ernestus, Mirjam T.C. & Baayen, R. Harald2006The functionality of incomplete neutralization in Dutch: the case of the past tense formationLaboratory PhonologyBerlinMoulton de Gruyter27-49
    • Ernestus, Mirjam T.C. & Baayen, R. Harald2007Paradigmatic effects in auditory word recognition: The case of alternating voice in DutchLanguage and Cognitive Processes221-24
    • Goldrick, Matthew2001Turbid output representations and the unity of opacityHirotani, M., Coetzee, A., Hall, N. & Kim, J.-Y. (eds.)NELS 30Amherst231-245
    • Jongman, A1998Effects of vowel length and syllable structure on segment duration in DutchJournal of Phonetics26207--222
    • Kager, René, Feest, Suzanne van der, Fikkert, Paula, Kerkhoff, Aannemarie & Zamuner, Tania S2007Representations of [Voice]: Evidence from aquisitionWeijer, Jeroen M. van de & Torre, Erik Jan van der (eds.)Voicing in Dutch: (de)voicing - phonology, phonetics, and psycholinguisticsAmsterdam/PhiladelphiaJohn Benjamins Publishing Company
    • Kager, René, Feest, Suzanne van der, Fikkert, Paula, Kerkhoff, Aannemarie & Zamuner, Tania S2007Representations of [Voice]: Evidence from aquisitionWeijer, Jeroen M. van de & Torre, Erik Jan van der (eds.)Voicing in Dutch: (de)voicing - phonology, phonetics, and psycholinguisticsAmsterdam/PhiladelphiaJohn Benjamins Publishing Company
    • Kerkhoff, A2007Acquisition of morpho-phonology: the Dutch voicing alternationUniversity of UtrechtThesis
    • Kerkhoff, A2007Acquisition of morpho-phonology: the Dutch voicing alternationUniversity of UtrechtThesis
    • Kharlamov, Viktor2012Incomplete neutralization and task effects in experimentally-elicited speech: evidence from the production and perception of word-final devoicing in RussianUniversity of OttawaThesis
    • Kissine, Mikhail, Velde, Hans van de & Hout, Roeland van2003An acoustic study of standard Dutch /v/, /f/, /z/ and /s/Linguistics in the NetherlandsJohn Benjamins Publishing Company93-104
    • Kissine, Mikhail, Velde, Hans van de & Hout, Roeland van2003An acoustic study of standard Dutch /v/, /f/, /z/ and /s/Linguistics in the NetherlandsJohn Benjamins Publishing Company93-104
    • Lisker, Leigh & Abramson, Arthur S1964A cross-language study of voicing in initial stops: Acoustical measurementsWord20384-422
    • Oostendorp, Marc van2008Incomplete devoicing in formal phonologyLingua1181362-1374
    • Slis, Iman H. & Cohen, Antonie1969On the complex regulating the voiced-voiceless distinction ILanguage and Speech1280-102
    • Slis, Iman H. & Cohen, Antonie1969On the complex regulating the voiced-voiceless distinction ILanguage and Speech1280-102
    • Slis, Iman H. & Heugten, M. van1989Voiced-voiceless distinction in Dutch fricativesLinguistics in the Netherlands 1989
    • Slis, Iman H. & Heugten, M. van1989Voiced-voiceless distinction in Dutch fricativesLinguistics in the Netherlands 1989
    • Velde, Hans van de, Gerritsen, Marinel & Hout, Roeland van1996The devoicing of fricatives in Standard Dutch: A real-time study based on radio recordingsLanguage Variation and Change8149-175
    • Warner, Natasha, Jongman, Allard, Sereno, Joan & Kemps, Rachel2004Incomplete neutralization and other sub-phonemic durational differences in production and perception: evidence from DutchJournal of Phonetics32251-276
    • Warner, Natasha, Jongman, Allard, Sereno, Joan & Kemps, Rachel2004Incomplete neutralization and other sub-phonemic durational differences in production and perception: evidence from DutchJournal of Phonetics32251-276
    • Zonneveld, Wim1994Fonologie van het Nederlands: een overzichtsartikelTijdschrift voor Nederlandse Taal- en Letterkunde1102-33
    • Zonneveld, Wim2007Issues in Dutch devoicingWeijer, J. M. van de & Torre, E. J. van der (eds.)Voicing in DutchAmsterdam/PhiladelphiaJohn Benjamins Publishing Company
    About this article
    Author(s): Nina Ouddeken
    Category: Dutch Phonology
    Keywords: phonological processfinal devoicing
    Version: 1.0, December 2015
    Version history:
    version editor(s) date remarks
    1.0 Nina Ouddeken December 2015
    printreport errorcite
    Jump to:
    X