1. The WordReference Forums have moved to new forum software. (Details)

Old English: Schwa

Discussion in 'Etymology, History of languages and Linguistics (EHL)' started by nemurenai, May 17, 2013.

  1. nemurenai Junior Member

    English
    Were unstressed vowels in Old English pronounced purely or as schwas? Did the schwa exist in Old English? I've never been able to find a straight answer for this.

    Thank you
     
  2. entangledbank

    entangledbank Senior Member

    London
    English - South-East England
    I believe it depends on the exact period. What were originally final [ɛ], [a], and fell together and became schwa sometime in late Old English. Likewise [ɛs] and [as], and other OE endings that had merged by the time of Middle English. Scribal variations would give some information about when this had happened in Mercia, Wessex, and so on.
     
  3. olena907 New Member

    Russia
    Russian
    books say the process started in Old English, but data's poor, poems can help, rhymes. May be this can help: The Cambridge history of the English language. Vol.1. The beginning to 1066, Lass R. On schwa. Stellenbosch Papers in Linguistics 15, 1986: 1-30
     
  4. Walshie79 Junior Member

    Shropshire, UK
    English (British)
    Almost certainly some Old English "e"s as early as the 9th century (possibly before that) were schwas, such as in the prefix ge- and the dative singular -e.

    Later, it is clear that many short /o/, /a/, /u/ were also schwa by the early 12th century (the Old/Middle transition period), particularly in inflectional endings which would then have been harder to distinguish: the genitive singular -esand nominative/accusative plural -as; the class of nouns with u in the nominative from those with a and e. The dative plural -um and weak declension gen/dat singular and nom plural -​an both seem to have merged as -en, with a schwa, early in Middle English.

    The text that shows it most clearly is ​Layamon's Brut (c.1200) which in vocabulary and morphology is basically late Old English, but in phonology more Middle English:

    An preost wes on leoden, Layamon wes ihoten, he wes leovenadhes sone, lidhe him beo drihten. He wonede at Ernleye on sevarne stathe at aedhelen are chirechen.

    (Old English) An preost waes on leodum, Laguman waes gehatan, he waes leofnothes sunu, lithe him beo dryhten. He wunede aet Earnleage on saeferne stathe aet aethelum anre cyrican

    So it is basically the same language as Old English, except for the vowels- in the endings all become e, which was a schwa.
     
  5. berndf Moderator

    Geneva
    German (Germany)
    That is an interesting hypothesis. It is an nearly risk-free assumption that the ME merger of vowel in unstressed suffix syllables means that they were reduced to Schwas. Interestingly, the same merger can also be observed for all the continental W-Germanic languages at the same time although English wasn't part of the W-Germanic dialect continuum any more. Would you think that this is (circumstantial) evidence for the hypothesis that the weakening of unstressed vowels in suffix/ending and prefix syllables already started much earlier and only on the 11th/12th century reached the point of phonemic neutralization in W-Germanic languages?
     
  6. jmx

    jmx Senior Member

    Barcelona
    Spain / incorrect Spanish
    I have a naïve question: Is there any Germanic language with no schwa?

    And then a second question comes to my mind: Are schwas particularly common in languages with lots of different vowel phonemes, as for example the Germanic languages?
     
  7. Walshie79 Junior Member

    Shropshire, UK
    English (British)
    Very much so, although 8th/9th century Old English seems to be more advanced than contemporary Old High German in this, often showing e form where OHG has a, i, u, o. (OE Tobecume thin rice, geweorthe thin wille vs OHG quaeme richi thin, uuerthe willeo thin). According to the Wikipedia article on OHG these vowels had all been reduced to e (schwa) by the 11th century, which if correct would certainly mean the same had happened by then in English, as it appears to have started earlier there.

    The stress patterns/accents of the WG languages seem to have had a big part to play in this, short vowels in unstressed syllables, particularly those endings which were never stressed, are always liable to be reduced.
     
  8. berndf Moderator

    Geneva
    German (Germany)
    Hmmm.... The OE version you quote is about 200 years younger than the OHG one. Older OE versions also have willa rather than wille.
     
  9. entangledbank

    entangledbank Senior Member

    London
    English - South-East England
    There probably isn't really any correspondence globally. The Germanic languages are a familiar example with many vowels including schwas, but you could probably furnish contrary systems if you examined enough. Even if English and West Germanic went down that road independently, it wouldn't mean it was a universally natural process.

    Neither Old Norse nor modern Icelandic have schwa, though the vowels have changed significantly, and are numerous, there is strong initial stress, and stress is now associated with length, all factors that might favour schwa. ON of course also gave rise to the Scandinavian dialects, which do have schwa.

    Consider Spanish with its small vowel inventory and no vowel reduction, and compare to Catalan and Portuguese where there's massive vowel reduction. Again, Russian has reduction, Polish and Czech lack it. Strong initial stress led to reduction in Old Latin (though not schwa) and Etruscan (I believe), but have not done so in Czech, Hungarian, or Finnish.

    I can think of some African languages with large vowel inventories but no schwa, but don't know enough to quote specifics.
     
  10. Peter94 Junior Member

    Abcdefgh
    Some Mexican dialects do have vowel reduction though. Check this Wikipedia page for sources.

    Some German dialects have [ɛ] instead of schwa. I'm not sure which ones though, Bühnendeutsch is for sure among them.
     
  11. berndf Moderator

    Geneva
    German (Germany)
    There are so many German dialects that you can't really exclude anything but that sounds rather weird to me. At any rate, [e] would make much more sense than [ɛ].
     
  12. Peter94 Junior Member

    Abcdefgh
    Maybe it would, but from what I remember I've seen it more often than not transcribed as <ɛ>. I'm not going to argue though, because I don't know which symbol is more accurate.

    And there are of course many English dialects without schwa. Among them are Caribbean, Black South African (it's not racist, that's how they're called), Irish accents. Not all of them, but some.

    I'm not sure about Scottish English. I think some accents have full vowels where RP would use a schwa.

    And on the other hand, younger speakers of Netherlandic Dutch (mostly in the central and central-west region) tend to use the schwa a lot more often than it's used in other accents, so that for example waarschijnlijk (probably, probable) is pronounced [ʋərˈsχɛɪ̯nlək] instead of [ʋaˑrˈsχɛɪ̯nlək]. But that's just a quick aside.
     
  13. berndf Moderator

    Geneva
    German (Germany)
    You seem to be riding the wrong horse here. We are talking about the phonemic Schwas created in MHG, like in eine or Gesang. I accept there are dialects where these Schwas either mute (in casual pronunciation) or [e] (in articulate, slightly hyper-corrective language) but [ɛ] is hard to imagine. There are dialects where final -er is [ɛ] or [ɛ:] but that's a different story.
     
  14. Peter94 Junior Member

    Abcdefgh
    As I said:

    You say it's <e>. I've checked my Duden Aussprachewörterbuch. It says that in Bühnenaussprache /ə/ is pronounced [e]. Sylvia Moosmüller's Vowels in Standard Austrian German also uses [e] where the vowel is phonemically /ə/ in Standard German. So it looks you're right.

    The only thing I'm not sure about is the phonemicity you're talking about. How is that relevant to my post? If you're talking about Dutch, it was an aside, not the main theme of my message. I'm aware that it has no connection to the German schwa-fronting, but it's a reverse process worth mentioning. I'd argue though that waarschijnlijk could be analyzed as /ʋərˈsχɛɪ̯nlək/ for speakers that have such heavy vowel reduction, as most phoneticians consider [ə] a vowel phoneme in Dutch.
     
  15. berndf Moderator

    Geneva
    German (Germany)
    You are again talking about -er. That is a completely different ballgame. I though we had already agreed that Germanic languages have a completely separate vowel system in front of /r/. Dutch has a phonemic Schwa but not in that word. It exist, e.g., in de.
     
    Last edited: Jul 11, 2013
  16. Peter94 Junior Member

    Abcdefgh
    No, I am not. It doesn't happen just before /r/, but in all environments. Not in Standard ("Posh") Dutch though.

    Standard Dutch does have a phonemic schwa in the suffix -lijk. Some speakers just go further and reduce the unstressed <aa> (and all other vowels) to [ə]. That would justify treating <aa> in waarschijnlijk as /ə/ for these speakers. /r/ is irrelevant here, though some Dutch accents of course do have pre-/r/ (and pre-/l/ before consonants and word-finally) allophones.
     
    Last edited: Jul 11, 2013
  17. berndf Moderator

    Geneva
    German (Germany)
    Ok, I see. Back to the question know, if all W-Germanic languages have a Schwa: I think so. It's realization might be in disguise, e.g. Swiss-German g'sie (=been) where the realization is the null-vowel, but I would think it is phonemically still present or at least there is an easily identifiable reflex of a former Schwa.
     
  18. Walshie79 Junior Member

    Shropshire, UK
    English (British)
    I've seen both, but even a suggests reduction compared to o. That OE version is variously given as late 10th/early 11th century; at the very latest OE seems to have developed schwa by about 950*, though as there was a "standardised" West Saxon spelling it may not fully show in writing, especially if it was more advanced in other dialects e.g. Mercian.

    *Personally I reckon more like 800 for the dative singular e and genitive singular es, and the 1st. person singular verb.
     
  19. berndf Moderator

    Geneva
    German (Germany)
    The dative singular -e is regular in OHG as well and can already by found as early as the 8th century in OHG, e.g. in the St Gall Paternoster Fater unseer, thu pist in himile, The variant willia instead of willio can already be found in the Gothic paternoster of the 4th century.

    I find it really difficult to make such a general statement that the weakening happened earlier in OE than in OHG.
     

Share This Page