Loading ...
Sorry, an error occurred while loading the content.
 

Re: Vowel Harmony?

Expand Messages
  • Jim Henry
    ... gjâ-zym-byn has nasal/oral vowel harmony; I don t know of any natlangs that do that. The only ones I know of have vowel harmony based on front/backness.
    Message 1 of 22 , Sep 1, 2009
      On Mon, Aug 31, 2009 at 2:22 AM, <deinx nxtxr><deinx.nxtxr@...> wrote:
      > I'm just curious about any conlangs out there that have vowel harmony and

      gjâ-zym-byn has nasal/oral vowel harmony; I don't know of any natlangs
      that do that. The only ones I know of have vowel harmony based on
      front/backness.

      --
      Jim Henry
      http://www.pobox.com/~jimhenry/
    • <deinx nxtxr>
      ... I ve seen some, like Turkic languages with rounding harmony. The language I m playing with doesn t have nasal vowels so I m not sure I ll have to worry
      Message 2 of 22 , Sep 1, 2009
        Jim Henry wrote:
        > On Mon, Aug 31, 2009 at 2:22 AM, <deinx nxtxr><deinx.nxtxr@...> wrote:
        >> I'm just curious about any conlangs out there that have vowel harmony and
        >
        > gj�-zym-byn has nasal/oral vowel harmony; I don't know of any natlangs
        > that do that. The only ones I know of have vowel harmony based on
        > front/backness.

        I've seen some, like Turkic languages with rounding harmony. The
        language I'm playing with doesn't have nasal vowels so I'm not sure
        I'll have to worry about that. I'm thinking maybe a
        front-center-back system.
      • Jim Henry
        ... ...I should clarify that I ve heard, vaguely, that some natlangs have nasal vowel harmony. But I don t know which they are or any more details. Wikipedia
        Message 3 of 22 , Sep 1, 2009
          On Tue, Sep 1, 2009 at 3:26 AM, Jim Henry<jimhenry1973@...> wrote:
          > On Mon, Aug 31, 2009 at 2:22 AM, <deinx nxtxr><deinx.nxtxr@...> wrote:
          >> I'm just curious about any conlangs out there that have vowel harmony and
          >
          > gjâ-zym-byn has nasal/oral vowel harmony; I don't know of any natlangs
          > that do that.  The only ones I know of have vowel harmony based on
          > front/backness.

          ...I should clarify that I've heard, vaguely, that some natlangs have
          nasal vowel harmony. But I don't know which they are or any more
          details. Wikipedia says vowel harmony can be based on backness,
          height, rounding, advanced/retracted tongue root, or nasalization, but
          it only gives natlang examples for backness and rounding.

          --
          Jim Henry
          http://www.pobox.com/~jimhenry/
        • taliesin the storyteller
          ... WALS makes no attempt to cover all possible features of all possible languages, it started out as a paper book after all, with all the limitations of that.
          Message 4 of 22 , Sep 1, 2009
            Amanda Babcock Furrow wrote:
            > On Mon, Aug 31, 2009 at 02:22:46AM -0400, <deinx nxtxr> wrote:
            >
            >> I'm just curious about any conlangs out there that have vowel harmony and
            >> how it's used. I'm considering it for a conlang I've been contemplating
            >> for a few days now.
            >
            > I tried checking on CALS, but for some reason neither CALS nor WALS seems
            > to have vowel harmony listed as one of the searchable features! Is WALS
            > hiding it behind some unfamiliar terminology? I can't imagine that they've
            > simply overlooked it...

            WALS makes no attempt to cover all possible features of all possible
            languages, it started out as a paper book after all, with all the
            limitations of that.

            Maybe WALS will add something on vowel harmony one day. In the meantime,
            if you're interested in testing something in CALS for me *hinthint*,
            send me a mail off-list.


            t.
          • Peter Bleackley
            ... Rounding harmony is one I ve always wanted to try out. Another thing I like the idea of (I think I tried it out in the Conlang Evolotion Experiment, but
            Message 5 of 22 , Sep 1, 2009
              staving <deinx nxtxr>:
              > Jim Henry wrote:
              >> On Mon, Aug 31, 2009 at 2:22 AM, <deinx
              >> nxtxr><deinx.nxtxr@...> wrote:
              >>> I'm just curious about any conlangs out there that have vowel harmony
              >>> and
              >>
              >> gj�-zym-byn has nasal/oral vowel harmony; I don't know of any natlangs
              >> that do that. The only ones I know of have vowel harmony based on
              >> front/backness.
              >
              > I've seen some, like Turkic languages with rounding harmony. The
              > language I'm playing with doesn't have nasal vowels so I'm not sure I'll
              > have to worry about that. I'm thinking maybe a front-center-back system.
              >

              Rounding harmony is one I've always wanted to try out.
              Another thing I like the idea of (I think I tried it out in the Conlang
              Evolotion Experiment, but talideon wiki is down at the moment) is what I
              call a "vowel melody" system, mainly on the ground that I don't know a
              better term for it. There's an underlying three-vowel system, but each
              vowel assimilates towards the following one, eg
              katiru > ketyru
              surami > soremi
              So we have a total of six vowels
              i y u
              e o
              a

              where subsequent syllables must contain vowels which are either
              identical or adjecent in the triangle, and the final vowel of a word
              must be at an apex of the triangle.

              Any real-life examples of such a system, and what would it be called?

              Pete
            • Michael Poxon
              For some time I ve considered a language with vowel-consonant harmony (i.e., voiceless consonants can only occur in the neighbourhood of front vowels, and vice
              Message 6 of 22 , Sep 1, 2009
                For some time I've considered a language with vowel-consonant harmony (i.e.,
                voiceless consonants can only occur in the neighbourhood of front vowels,
                and vice versa) so /te/ but not **/de/. I hope there aren't any natlangs
                like this!
                Mike

                > I'm just curious about any conlangs out there that have vowel
                > harmony and how it's used. I'm considering it for a conlang I've
                > been contemplating for a few days now.
              • Alex Fink
                ... Guarani is one example. In a nasal word, AIUI, all vowels and sonorants will be nasalised, in a non-nasal word none will be. Also prenasalised voiced
                Message 7 of 22 , Sep 1, 2009
                  On Tue, 1 Sep 2009 03:33:16 -0400, Jim Henry <jimhenry1973@...> wrote:

                  >On Tue, Sep 1, 2009 at 3:26 AM, Jim Henry<jimhenry1973@...> wrote:
                  >> gjâ-zym-byn has nasal/oral vowel harmony; I don't know of any natlangs
                  >> that do that. The only ones I know of have vowel harmony based on
                  >> front/backness.
                  >
                  >...I should clarify that I've heard, vaguely, that some natlangs have
                  >nasal vowel harmony. But I don't know which they are or any more
                  >details. Wikipedia says vowel harmony can be based on backness,
                  >height, rounding, advanced/retracted tongue root, or nasalization, but
                  >it only gives natlang examples for backness and rounding.

                  Guarani is one example. In a nasal word, AIUI, all vowels and sonorants
                  will be nasalised, in a non-nasal word none will be. Also prenasalised
                  voiced stops (/mb)/) only occur in non-nasal words, nasals proper (/m/) only
                  in nasal words.

                  On Tue, 1 Sep 2009 13:38:00 +0100, Michael Poxon <mike@...> wrote:

                  >For some time I've considered a language with vowel-consonant harmony (i.e.,
                  >voiceless consonants can only occur in the neighbourhood of front vowels,
                  >and vice versa) so /te/ but not **/de/. I hope there aren't any natlangs
                  >like this!

                  No, I think there aren't. This is basically what started the current
                  "Sketch of New Conlang" thread, which see for a bit more.

                  But of course there are several attested types of vowel-consonant harmony
                  out there; it's just that the vowels and consonants will agree in their
                  value of a single feature, not entirely dissimilar ones like voice and
                  frontness. The nasal harmony above is one example. For another, Mongolian
                  and I think also some Turkic languages have their front-back harmony carry
                  through to back consonants, so velar consonants appear in front words and
                  uvulars in back ones. Or pharyngealisation; some Arabic varieties have
                  this, as does Chilcotin.

                  Alex
                • Benct Philip Jonsson
                  ... AFMOC Classical Sohlob has vowel height harmony. The vowel system looks like this: high: i i u low: & a Q A word can have only all high or all
                  Message 8 of 22 , Sep 4, 2009
                    <deinx nxtxr> skrev:
                    > I'm just curious about any conlangs out there that have vowel harmony
                    > and how it's used. I'm considering it for a conlang I've been
                    > contemplating for a few days now.
                    >

                    AFMOC Classical Sohlob has vowel height harmony.

                    The vowel system looks like this:

                    high: i i\ u
                    low: & a Q

                    A word can have only all high or all low vowels.
                    The related languages have similar systems. Kidilib
                    has merged /i/ and /i\/ but is otherwise identical.
                    Heleb has front rounded vowels:

                    i y M u
                    & 9 A Q

                    and the vowels of every word must agree
                    in height AND frontness, but not in rounding.

                    There is an ANADEW for vowel height harmony, namely
                    in Korean, although it's vestigal in the modern language:

                    <http://www.cog.jhu.edu/grad-students/finley/isokl-paper.pdf>

                    /BP 8^)>
                    --
                    Benct Philip Jonsson -- melroch atte melroch dotte se
                    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
                    "C'est en vain que nos Josués littéraires crient
                    à la langue de s'arrêter; les langues ni le soleil
                    ne s'arrêtent plus. Le jour où elles se *fixent*,
                    c'est qu'elles meurent." (Victor Hugo)
                  • John Vertical
                    ... Umlaut, basically? Except maybe for the part where each feature can only spred one vowel leftwards (eg. /katiru/ being [ketyru] rather than [k2tyru]). If
                    Message 9 of 22 , Sep 7, 2009
                      >Another thing I like the idea of (I think I tried it out in the Conlang
                      >Evolotion Experiment, but talideon wiki is down at the moment) is what I
                      >call a "vowel melody" system, mainly on the ground that I don't know a
                      >better term for it. There's an underlying three-vowel system, but each
                      >vowel assimilates towards the following one, eg
                      >katiru > ketyru
                      >surami > soremi
                      >So we have a total of six vowels
                      >i y u
                      > e o
                      > a
                      >
                      >where subsequent syllables must contain vowels which are either
                      >identical or adjecent in the triangle, and the final vowel of a word
                      >must be at an apex of the triangle.
                      >
                      >Any real-life examples of such a system, and what would it be called?
                      >
                      >Pete

                      Umlaut, basically?

                      Except maybe for the part where each feature can only spred one vowel
                      leftwards (eg. /katiru/ being [ketyru] rather than [k2tyru]). If you did
                      that, all vowels from the antepenult left would probably be the same, and
                      also likely [2], so I think that's aesthetically very justified.

                      AFMCL, haven't done much with harmony, but one language sketch spontaneously
                      develop'd a height harmony of sorts, with /e a o/ in one group and /E Q O/
                      in another. By "spontaneously develop'd" I mean I was looking over the
                      corpus (at a stage of just a few dozen words) and spotted the pattern.
                      Funny, that.

                      For another example, Proto-Sundan' has a system of total harmony - every
                      vowel in a word must be the same (inventory: /i u\ u e 8 o a/). As an
                      exception, /a/ is allowed as a final vowel. The interesting part, however,
                      is probably how this then breiks down in the descendants… but I've mostly
                      not work'd that out yet.

                      (IIRC there are some varieties of Mongolian that have total harmony)

                      John Vertical
                    • Eric Christopherson
                      ... Does that operate on the vowels in roots, or on roots+affixes? I ve been googling total harmony and, although I haven t found the Mongolian varieties you
                      Message 10 of 22 , Sep 7, 2009
                        On Sep 7, 2009, at 1:15 PM, John Vertical wrote:

                        > For another example, Proto-Sundan' has a system of total harmony -
                        > every
                        > vowel in a word must be the same (inventory: /i u\ u e 8 o a/). As an
                        > exception, /a/ is allowed as a final vowel. The interesting part,
                        > however,
                        > is probably how this then breiks down in the descendants… but I've
                        > mostly
                        > not work'd that out yet.
                        >
                        > (IIRC there are some varieties of Mongolian that have total harmony)

                        Does that operate on the vowels in roots, or on roots+affixes? I've
                        been googling total harmony and, although I haven't found the
                        Mongolian varieties you mention, it's been very interesting. It
                        appears that total harmony is usually fairly limited (e.g. just to
                        certain affixes). A book I found on Google Book Search[1] says,
                        speaking of certain suffixes in Yucatec Maya and Ainu being total
                        copies of the preceding vowel, "One might argue... whether these
                        patterns fall under the category of vowel harmony or should rather be
                        treated as reduplication, since harmony applies only between the root
                        vowel and a few affix vowels and the process is not iterative. Given
                        that this type of complete harmony affects maximally one vowel in a
                        word it could be regarded as separate from vowel harmony, just as
                        umlaut." Indeed, I've often wondered about the relations between
                        reduplication, vowel copying, vowel harmony, umlaut, and ablaut.

                        [1] _Vowel harmony and correspondence theory_
                        <http://books.google.com/books?id=OuEGsxKhFG4C&pg=PA16&lpg=PA16&dq=vowel+harmony+target&source=bl&ots=yb4W3n_b-J&sig=2QfWxkDtjwILokl251zhQVFkNuo&hl=en&ei=b3GlStujNYrbnAe29oHuBA&sa=X&oi=book_result&ct=result&resnum=4#v=onepage&q=vowel%20harmony%20target&f=false
                        >
                      • Roger Mills
                        Pete (Bleackley?) wrote: ... My Prevli is similar, but the word-bases are only 2-syl. with penult stress, and the final V is affected in certain forms of the
                        Message 11 of 22 , Sep 10, 2009
                          Pete (Bleackley?) wrote:
                          ....what I
                          > >call a "vowel melody" system, mainly on the ground that
                          > I don't know a
                          > >better term for it. There's an underlying three-vowel
                          > system, but each
                          > >vowel assimilates towards the following one, eg
                          > >katiru > ketyru
                          > >surami > soremi
                          > >So we have a total of six vowels
                          > >i y u
                          > >  e o
                          > >   a
                          > >
                          > >where subsequent syllables must contain vowels which
                          > are either
                          > >identical or adjecent in the triangle, and the final
                          > vowel of a word
                          > >must be at an apex of the triangle.
                          > >
                          > >Any real-life examples of such a system, and what would
                          > it be called?

                          My Prevli is similar, but the word-bases are only 2-syl. with penult stress, and the final V is affected in certain forms of the word, but in others forms, the underlying vowels appear. Thus the modified final vowels are not really phonemic. FWIW, it has a 5-vowel system /i e a u o/, only /i a u/ occur in final syl. and modify to [e, y, o, &, O]. (IIRC)

                          Your system seems to involve a recusive umlaut rule-- V1 is affected by V2, then V2 by V3. You refer to the underlying /a i u/, but do these vowels ever actually occur in 1st or 2nd syllables? Where is the stress (perhaps word-final?), and does it matter?

                          It's a rather intresting system :-))))
                        • Peter Bleackley
                          ... It s certainly possible for apical vowels to occur in first sylables, for example kafanu kafonu As for position of stress, I haven t consisdered it. I
                          Message 12 of 22 , Sep 11, 2009
                            staving Roger Mills:
                            > Pete (Bleackley?) wrote: ....what I
                            >>> call a "vowel melody" system, mainly on the ground that
                            >> I don't know a
                            >>> better term for it. There's an underlying three-vowel
                            >> system, but each
                            >>> vowel assimilates towards the following one, eg katiru > ketyru
                            >>> surami > soremi So we have a total of six vowels i y u � e o ���a
                            >>>
                            >>>
                            >>> where subsequent syllables must contain vowels which
                            >> are either
                            >>> identical or adjecent in the triangle, and the final
                            >> vowel of a word
                            >>> must be at an apex of the triangle.
                            >>>
                            >>> Any real-life examples of such a system, and what would
                            >> it be called?
                            >
                            > My Prevli is similar, but the word-bases are only 2-syl. with penult
                            > stress, and the final V is affected in certain forms of the word, but
                            > in others forms, the underlying vowels appear. Thus the modified
                            > final vowels are not really phonemic. FWIW, it has a 5-vowel system
                            > /i e a u o/, only /i a u/ occur in final syl. and modify to [e, y, o,
                            > &, O]. (IIRC)
                            >
                            > Your system seems to involve a recusive umlaut rule-- V1 is affected
                            > by V2, then V2 by V3. You refer to the underlying /a i u/, but do
                            > these vowels ever actually occur in 1st or 2nd syllables? Where is
                            > the stress (perhaps word-final?), and does it matter?
                            >

                            It's certainly possible for apical vowels to occur in first sylables,
                            for example

                            kafanu > kafonu

                            As for position of stress, I haven't consisdered it.

                            I think part of the inspiration is Grey Codes - if you think of it in
                            terms of distinctive features
                            f l r
                            i 1 0 0
                            e 1 1 0
                            a 0 1 0
                            o 0 1 1
                            u 0 0 1
                            y 1 0 1

                            Pete
                          • Roger Mills
                            ... Aha. I had nt thought that through. I suppose kafana, kifini, kifynu, kufunu, kufona etc. are possible too? ... My thought was that it must be word final,
                            Message 13 of 22 , Sep 11, 2009
                              --- On Fri, 9/11/09, Peter Bleackley <Peter.Bleackley@...> wrote:

                              > staving Roger Mills:

                              > > Your system seems to involve a recusive umlaut rule--
                              > V1 is affected
                              > > by V2, then V2 by V3. You refer to the underlying /a i
                              > u/, but do
                              > > these vowels ever actually occur in 1st or 2nd
                              > syllables? Where is
                              > > the stress (perhaps word-final?), and does it matter?
                              > >
                              >
                              > It's certainly possible for apical vowels to occur in first
                              > sylables, for example
                              >
                              > kafanu > kafonu

                              Aha. I had'nt thought that through. I suppose kafana, kifini, kifynu, kufunu, kufona etc. are possible too?
                              >
                              > As for position of stress, I haven't consisdered it.

                              My thought was that it must be word final, but maybe that's not necessary.
                            • Peter Bleackley
                              ... Yes, they re all possible. Imagine that a recent ancestor had a 3-vowel system, and feature-spreading occured. Pete
                              Message 14 of 22 , Sep 11, 2009
                                staving Roger Mills:
                                > --- On Fri, 9/11/09, Peter Bleackley <Peter.Bleackley@...>
                                > wrote:
                                >
                                >> staving Roger Mills:
                                >
                                >>> Your system seems to involve a recusive umlaut rule--
                                >> V1 is affected
                                >>> by V2, then V2 by V3. You refer to the underlying /a i
                                >> u/, but do
                                >>> these vowels ever actually occur in 1st or 2nd
                                >> syllables? Where is
                                >>> the stress (perhaps word-final?), and does it matter?
                                >>>
                                >> It's certainly possible for apical vowels to occur in first
                                >> sylables, for example
                                >>
                                >> kafanu > kafonu
                                >
                                > Aha. I had'nt thought that through. I suppose kafana, kifini, kifynu,
                                > kufunu, kufona etc. are possible too?

                                Yes, they're all possible. Imagine that a recent ancestor had a 3-vowel
                                system, and feature-spreading occured.

                                Pete
                              Your message has been successfully submitted and would be delivered to recipients shortly.