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

Re: Dictionary of Sound Changes -- Day 1

Expand Messages
  • Herman Miller
    ... Could be useful, especially with the reverse lookup section. Much of what I do in trying to reconstruct the history of my languages is trying to figure out
    Message 1 of 6 , May 5, 2013
      On 5/5/2013 6:06 AM, Paul Bennett wrote:
      > I've started preliminary work on my Magnum Opus, or at least one of my
      > Magnum Opuses. It's a dictionary of the sound changes in the principal
      > Indo-European languages, which will hopefully be useful for those of you
      > who are into that sort of thing. The draft listed below is basically what I
      > could crank out in about an hour. There is (obviously) a great deal of work
      > yet to be done. Questions, comments, suggestions?
      >
      > https://docs.google.com/file/d/0B21ee_I1WzbwZEQ1SExsLU9FaGs/edit?usp=sharing
      >
      > --
      > Paul Bennett

      Could be useful, especially with the reverse lookup section. Much of
      what I do in trying to reconstruct the history of my languages is trying
      to figure out the possible options for sounds in a proto-language when I
      already know the sounds in the modern language.

      The italic font you're using has some pretty bad display issues
      depending on how much you zoom in or out. It looks like the hinting
      wasn't set up properly, with some lines too thin and others too thick.
      Could you try a different font?

      The idea of a uniform phonemic notation is an interesting one. At some
      point it requires making assumptions about the pronunciation of ancient
      words. Well, basically any time before the invention of recording
      technology, but more so the further back you go. But it would certainly
      make things easier to follow for someone who isn't acquainted with all
      the specialist notations for the different branches.

      Obviously there will be issues with deciding which phoneme of one
      language is the "same" as a phoneme of another language. English has the
      phoneme /t/ with [t] and [tʰ] (both alveolar) as allophones, but Hindi
      has /t̪/, /t̪ʰ/, /ʈ/, and /ʈʰ/ (dental and retroflex) as distinct phonemes.
    • Paul Bennett
      The font display issues seem to be mostly with Google s PDF rendering engine, which I ve had quite serious problems with in the past. The PDF renders fine (for
      Message 2 of 6 , May 6, 2013
        The font display issues seem to be mostly with Google's PDF rendering
        engine, which I've had quite serious problems with in the past.

        The PDF renders fine (for me) in Adobe Acrobat (on Windows, and Android),
        Gnome Document Viewer (on Linux), Amazon Kindle (on Android), and Libre
        Office, which is what I've been using at this stage to type in the
        boilerplate -- there will eventually be a database with a user-friendly (or
        at least me-friendly) UI that exports via XML and XSLT to something ready
        to just plop into the Libre Office text (or some other mechanism that
        becomes PDF at the click of a mouse).

        The font in question is SIL Gentium, which I've never had issues with under
        any other circumstances, and which seems to cover all the characters I'll
        be needing, as well as being easier on the eyes than (e.g.) Code2000 or
        Bitstream (or Titus) Cyberbit, and easier on the psyche than Computer
        Modern.

        As far as a "uniform phonemic notation", it'll be IPA, pure and simple, and
        make as few assumptions as possible, where possible. Where two different
        languages have two different phonemes, I'll be using two different IPA
        symbols (because that's a matter of Intergalactic Law ;-)

        I will, however, be making the controversial step of assigning concrete
        values to the PIE sounds, most controversially the laryngeals. I don't
        think it matters too much what those values actually are, because they'll
        be whisked away into vowels and fricatives and all that good stuff very
        early on. I'm thinking /ʔ/, /ʕ/, /ʕʷ/ because they're fairly easy to grasp,
        but I could go for whatever the current consensus is (e.g. /ħ/, /ʁ/, /ʁʷ/
        or whatever else is fashionable), as long as they agree more or less with
        my own working hypothesis that they had something to do with "three weird
        noises, one at a slightly different POA, and one labialized").

        Likewise, I'll be going for /p/, /pʰ/, /pˀ/ for the PIE stops series, with
        footnotes and sidenotes and whatever all else absolutely chock-full of ifs,
        ands, buts, and maybes (and other caveats).

        I have also been encouraged to include more than just "the principal PIE
        langauges". Since the text is designed for actual commercial publication in
        something like 6x9 or A5 paper soft-cover (via one of the more reputable
        self-publishing outfits, maybe lulu.com if they're still going and still
        reputable), it's going to be impractical to stretch it beyond 300 to 400
        pages, 500 at the absolute limit.

        Since I'll be building an application and UI to make the process easy on
        me, I have no problem with releasing "the principal Amerind languages",
        "the principal Afro-Asiatic languages" or whatever else, and plausibly "the
        World's major languages (by number of modern speakers)" and "the
        most-interesting sound changes regardless of family", if anyone else wants
        to contribute in exchange for co-authorship credits and a cut of any money
        made. Have your people talk to my people ;-)

        Thanks for the tips and encouragement!


        --
        Paul



        On Sun, May 5, 2013 at 11:12 AM, Herman Miller <hmiller@...> wrote:

        > On 5/5/2013 6:06 AM, Paul Bennett wrote:
        >
        >> I've started preliminary work on my Magnum Opus, or at least one of my
        >> Magnum Opuses. It's a dictionary of the sound changes in the principal
        >> Indo-European languages, which will hopefully be useful for those of you
        >> who are into that sort of thing. The draft listed below is basically what
        >> I
        >> could crank out in about an hour. There is (obviously) a great deal of
        >> work
        >> yet to be done. Questions, comments, suggestions?
        >>
        >> https://docs.google.com/file/**d/0B21ee_**I1WzbwZEQ1SExsLU9FaGs/edit?**
        >> usp=sharing<https://docs.google.com/file/d/0B21ee_I1WzbwZEQ1SExsLU9FaGs/edit?usp=sharing>
        >>
        >> --
        >> Paul Bennett
        >>
        >
        > Could be useful, especially with the reverse lookup section. Much of what
        > I do in trying to reconstruct the history of my languages is trying to
        > figure out the possible options for sounds in a proto-language when I
        > already know the sounds in the modern language.
        >
        > The italic font you're using has some pretty bad display issues depending
        > on how much you zoom in or out. It looks like the hinting wasn't set up
        > properly, with some lines too thin and others too thick. Could you try a
        > different font?
        >
        > The idea of a uniform phonemic notation is an interesting one. At some
        > point it requires making assumptions about the pronunciation of ancient
        > words. Well, basically any time before the invention of recording
        > technology, but more so the further back you go. But it would certainly
        > make things easier to follow for someone who isn't acquainted with all the
        > specialist notations for the different branches.
        >
        > Obviously there will be issues with deciding which phoneme of one language
        > is the "same" as a phoneme of another language. English has the phoneme /t/
        > with [t] and [tʰ] (both alveolar) as allophones, but Hindi has /t̪/, /t̪ʰ/,
        > /ʈ/, and /ʈʰ/ (dental and retroflex) as distinct phonemes.
        >
      • Roger Mills
        ... The font display issues seem to be mostly with Google s PDF rendering engine, which I ve had quite serious problems with in the past. The PDF renders fine
        Message 3 of 6 , May 6, 2013
          --- On Mon, 5/6/13, Paul Bennett <paul.w.bennett@...> wrote:
          The font display issues seem to be mostly with Google's PDF rendering
          engine, which I've had quite serious problems with in the past.

          The PDF renders fine (for me) in Adobe Acrobat (on Windows, and Android),
          Gnome Document Viewer (on Linux), Amazon Kindle (on Android), and Libre
          Office, which is what I've been using at this stage to type in the
          boilerplate -- there will eventually be a database with a user-friendly (or
          at least me-friendly) UI that exports via XML and XSLT to something ready
          to just plop into the Libre Office text (or some other mechanism that
          becomes PDF at the click of a mouse).

          The font in question is SIL Gentium, which I've never had issues with under
          any other circumstances, and which seems to cover all the characters I'll
          be needing, as well as being easier on the eyes than (e.g.) Code2000 or
          Bitstream (or Titus) Cyberbit, and easier on the psyche than Computer
          Modern.
          ==============================================

          With Windows Vista and later, Times New Roman has (AFAIK) full IPA coverage, and is of course well known, and was required by the journal publishers I've tried to deal with.... I've seen problems with Gentium in pdf's in the past...  Herman Miller's _Thryomanes_ is good too.
        • Christophe Grandsire-Koevoets
          ... I m personally partial to Charis SIL. Looks great everywhere. And a better alternative to Gentium IMHO unless you need Greek support. -- Christophe
          Message 4 of 6 , May 6, 2013
            On 6 May 2013 15:19, Roger Mills <romiltz@...> wrote:

            >
            > With Windows Vista and later, Times New Roman has (AFAIK) full IPA
            > coverage, and is of course well known, and was required by the journal
            > publishers I've tried to deal with.... I've seen problems with Gentium in
            > pdf's in the past... Herman Miller's _Thryomanes_ is good too.
            >

            I'm personally partial to Charis SIL. Looks great everywhere. And a better
            alternative to Gentium IMHO unless you need Greek support.
            --
            Christophe Grandsire-Koevoets.

            http://christophoronomicon.blogspot.com/
            http://www.christophoronomicon.nl/
          • David McCann
            On Mon, 6 May 2013 05:16:01 -0400 ... It looks a bit ugly in Opera, but it s fine when downloaded and viewed with Evince. ... I agree with Sally Thomason here!
            Message 5 of 6 , May 6, 2013
              On Mon, 6 May 2013 05:16:01 -0400
              Paul Bennett <paul.w.bennett@...> wrote:

              > The font display issues seem to be mostly with Google's PDF rendering
              > engine, which I've had quite serious problems with in the past.

              It looks a bit ugly in Opera, but it's fine when downloaded and
              viewed with Evince.

              > As far as a "uniform phonemic notation", it'll be IPA, pure and
              > simple

              I agree with Sally Thomason here!
              http://itre.cis.upenn.edu/~myl/languagelog/archives/005287.html
              And surely it's better to use the established transcriptions?

              > I will, however, be making the controversial step of assigning
              > concrete values to the PIE sounds, most controversially the
              > laryngeals. I don't think it matters too much what those values
              > actually are, because they'll be whisked away into vowels and
              > fricatives and all that good stuff very early on. I'm
              > thinking /ʔ/, /ʕ/, /ʕʷ/ because they're fairly easy to grasp, but I
              > could go for whatever the current consensus is (e.g. /ħ/, /ʁ/, /ʁʷ/
              > or whatever else is fashionable)

              Similarly, /h₁ h₂ h₃/ are less likely to confuse, as well as allowing
              you the cover /h/ when they all come out the same way.

              > Likewise, I'll be going for /p/, /pʰ/, /pˀ/ for the PIE stops series,
              > with footnotes and sidenotes and whatever all else absolutely
              > chock-full of ifs, ands, buts, and maybes (and other caveats).

              I hope that's a typo! I use /p p’ b/ and I could live with /p b bʰ/.
            Your message has been successfully submitted and would be delivered to recipients shortly.