German: -chs- (pronunciation)

AndrasBP

Senior Member
Hungarian
Hello,

In German words such as "Dachs", "Fuchs" or "Wechsel", the sequence -chs- is pronounced /ks/, not /xs/ or /çs/ as the spelling suggests.
Is it just a spelling convention, the fricative sounds being phonotactically impossible before /s/, or did the pronunciation of these words change to /ks/ some time in history, the spelling representing an older phonology?
 
  • berndf

    Moderator
    German (Germany)
    There was probably a PGm [-xs] which in some WGm languages became [-ks] and in others [-s]/[-ss], e.g., Eng. fox, HGerm. Fuchs, LGerm/Dutch vos. There are two forms for a female fox: German Füchsin/English vixen, which has an s and the hardening to [k]. But there is also a second female form without s and without [k]: Fehe. The oldest OHG forms of Fuchs and Fehe are fuhs and foha. By contrast in Old English we find fox and fyxen (corresponding to Füchsin). So, the hardening was probably already complete in Old English but not in OHG. That would at explain why [ks] is still spelled chs and not x as in English.
     

    berndf

    Moderator
    German (Germany)
    Occasionally, yes. It is regular in Swiss German, like e.g. in Zurich German Sächsilüüte (six o'clock bells). The is a form of the second Germanic consonant shift k>kx>x occurring only in very South as in the probably most famous Swiss German word Chuchichäschtli (Küchenkästlein).
     
    Last edited:

    berndf

    Moderator
    German (Germany)
    That is why I compared it to the k>x in other cases in Swiss dialect. I would assume this to be a later development but I am not sure.
     

    Riverplatense

    Senior Member
    German — Austria
    Apparantly Austrians pronounce Daxs as /daxs/.
    https://en.wiktionary.org/wiki/Dachs#Pronunciation
    I've never heard this pronunciation, except for some isolated cases that I'd consider hypercorrection. This pronunciation appears to be even less likely taking into account that in dialectal speech (I'm not sure for Vorarlberg, but in Bavarian dialects definitely) the pronunciation with /x/ is rather unthinkable. Besides, also Duden – Das Aussprachewörterbuch only gives [daks], without any alternative pronunciation.
     

    eamp

    Member
    German (Austria)
    I think in Austria this is purely a spelling pronunciation, all the Bavarian dialects have /ks/ here and the few samples I could find from Vorarlberg also had /ks/ (not to say there could not exist pockets that indeed have /xs/).
    From personal experience I would say the pronunciation with /xs/ is on the rise here in Vienna though, especially among younger women. And then it is applied most consistently in the word 'sechs' to avoid homophony with 'Sex'. In this regard I have even heard some go as far as voicing the initial s - only in this word - to maximize the contrast: /zexs/ vs. /seks/ ...

    In Switzerland almost all dialects have /xs/, exceptions I found include some areas near Lake Constance, Basel and Walser German, so its use in spoken Swiss Standard German is not surprising.
    I don't think this pronunciation can easily be related to the simplification of initial /kx/ to /x/, and considering there does exist a contrast /xs/ - /ks/ in these dialects, I reckon an archaism is most likely here.
     

    berndf

    Moderator
    German (Germany)
    considering there does exist a contrast /xs/ - /ks/ in these dialects
    Not that I know of. High Alemannic (except for the regional exeptions you mentioned) have lost the /k/ phoneme completely (there is only <gg> as in Weggli or Schoggi as a reflex of original /k:/). Any [-ks-] would occur in foreign words only.
     

    AndrasBP

    Senior Member
    Hungarian
    Even in Standard German, the fricative is preserved is sechzehn /ˈzɛçtseːn/.
    Has the "hardening" to /k/ never happened here or is it a secondary "softening" somehow caused by /ts/?
     

    berndf

    Moderator
    German (Germany)
    Even in Standard German, the fricative is preserved is sechzehn /ˈzɛçtseːn/.
    Has the "hardening" to /k/ never happened here or is it a secondary "softening" somehow caused by /ts/?
    The s of sechs is lost in sechzehn. And it was lost relatively early, still in OHG. So there is no reason for /x/>/k/ to happen any more.
     

    eamp

    Member
    German (Austria)
    Not that I know of. High Alemannic (except for the regional exeptions you mentioned) have lost the /k/ phoneme completely (there is only <gg> as in Weggli or Schoggi as a reflex of original /k:/). Any [-ks-] would occur in foreign words only.
    Well, yes, original (West-Germanic) /k/ has certainly disappeared, but that wasn't really what I wanted to signify...
    What I mean is that these dialects distinguish between a cluster consisting of a velar stop plus s and one of a velar fricative plus s while Standard German doesn't (except at morpheme boundaries).
    Of course the first kind is secondary, arising from syncope of a vowel between a stop and s and later from loanwords, but in any case only Swiss German seems to maintain a pronunciation difference.
    Though the actual situation is complex as there are 5 potential ohg. velar sequences /kk/, /gg/, /g/, /hh/ and /h/ plus the two kinds of sibilant sounds, all of which could result in a different outcome once in contact. And dialects indeed seem to show a variety of outcomes, but I find it too difficult to come by reliable transcripts for some quick insights ...
     

    ahvalj

    Senior Member
    Last edited:

    berndf

    Moderator
    German (Germany)
    I meant the original Germanic sound transliterated at the turn of the eras by Romans as ch and Greeks as χ in cases like Chattī > Hesse, Chamavī > Hamaland, Chaucī, Cheruscī, Charudēs > Hordaland, Chalī, Chaemae, Chaedinī > heathen etc.
    Yes, that is the same as h. What we express by "h" today was an allophone of this sounds that developed some time in the middle ages. When exactly is not clear because Germanic languages started to differentiate them in writing only around 1200.
     

    ahvalj

    Senior Member
    Yes, phonemically, but I meant the actual sounds. Wulfila uses his h for the Greek h: Helias, Herodes, Haileisaius and also Abraham, Beþlahaim (but Iairusalem and Wmainaius) vs. x found only in Xristus and, strangely, in Xreskus pro Crescens, otherwise χ becomes k: Iairiko, Antiaukia, Twkeikus. This suggests that in his speech this consonant sounded rather as h. In contrast, Slavic at approximately the same time consistently borrows this Gothic sound as x: xlěbъ "bread" (hlaifs), xlěvъ "stall" (hlaiw "tomb, grave", cp. Slavic xlěvina "pit"), xъlmъ "hill", xǫdogъ "skillful" (from *handags, an alternative form of handugs), lixva "rest" (from an otherwise unattested outcome of PIE leı̯kʷ- "to leave" — cp. leiƕan) (the Slavic forms are given in their attested shape of 6–8 centuries later).
     
    Last edited:

    berndf

    Moderator
    German (Germany)
    Yes, phonemically, but I meant the actual sounds. Wulfila uses his h for the Greek h: Helias, Herodes, Haileisaius and also Abraham, Beþlahaim (but Iairusalem and Wmainaius) vs. x found only in Xristus and, strangely, in Xreskus pro Crescens, otherwise χ becomes k: Iairiko, Antiaukia, Twkeikus. This suggests that in his speech this consonant sounded rather as h. In contrast, Slavic at approximately the same time consistently borrows this Gothic sound as x: xlěbъ "bread" (hlaifs), xlěvъ "stall" (hlaiw "tomb, grave", cp. Slavic xlěvina "pit"), xъlmъ "hill", xǫdogъ "skillful" (from *handags, an alternative form of handugs), lixva "rest" (from an otherwise unattested outcome of PIE leı̯kʷ- "to leave").
    We don't really know how the Gothic X was supposed to be pronounced. It was mainly used to transcribe Greek words. I don't think you can deduce from this that Gothic distinguished phonetically between h and x.

    In West Germanic languages there was definitely no written distinction between the sounds before about 1200.

    Anyway, it doesn't matter for this thread. We are talking about a sequence that only occurs at the end of a syllable. [h] is phonologically completely impossible in any Germanic languages. In the spelling saihs, <h> cannot represent [h] but only [x].
     
    < Previous | Next >
    Top