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

English lexical sets in Teascript

Expand Messages
  • Herman Miller
    I m continuing to work on my new alphabet project, Teascript. Since it s hard to talk about it without an illustration, I put up an image of one way Teascript
    Message 1 of 7 , Nov 2, 2011
    • 0 Attachment
      I'm continuing to work on my new alphabet project, Teascript. Since it's
      hard to talk about it without an illustration, I put up an image of one
      way Teascript might be used to write English vowels.

      http://www.prismnet.com/~hmiller/png/teascript-lexical-sets.png

      I'm still working out the details for the shapes of the letters, but
      I've got a rough plan that seems to be working. Basically I'm looking
      for a featural script that's easy to write by hand and read, and takes
      up about the same amount of space on a page as more traditional
      alphabets. Some of the letters may look a bit like Tengwar, but the
      effect of the whole script looks a bit different.

      The general plan is that sounds in the front of the mouth have a
      vertical stroke on the left side, and sounds in the back of the mouth
      have one on the right side. Ascenders and descenders are added to
      distinguish different points of articulation according to the language.
      In English (and most languages), bilabial and labiodental sounds have a
      descender on the left. An ascender on the left can have different
      meanings in different languages; I'm using it for dental fricatives in
      English and alveolar affricates in Tirelat. Hindi could use it for
      retroflex consonants.

      In English, voiced stops have a single horizontal stroke attached to the
      vertical stroke, so the letter for /d/ looks a bit like "r" and the
      letter for /b/ like "ɼ". Voiceless stops continue the horizontal stroke
      into a vertical stroke (a single stroke like "7"), so that the letter
      /t/ looks like "n". Voiceless fricatives have a curved stroke with a
      corner that looks a bit like the number "2". But a different language
      might have a different mapping for these sets of letters. Basically I
      have four sets for stops, one for nasals, and two for fricatives.

      Vowels are basically in 3 sets: front, central, and back, with 3 heights
      in each set (a circle of 8 vowels with one in the middle). There are
      also 3 extra vowel letters which may be used as needed, and a length mark.

      In this way I can keep the number of letters to a minimum while still
      keeping the featural aspects of the script. I've actually managed to
      assign most of the common letters to the alphabetic keys on the keyboard
      (with or without the shift key), but I've had to use a few punctuation
      keys for some less common letters (like the ones I'm using for English
      /θ/ and /ð/).

      Teascript also has a small set of "miscellaneous" letters that can be
      used for liquids, approximants, or other sounds that don't fit into the
      overall table of phonemes for a language. These are represented by
      letters with curved strokes without a vertical stroke on the right or
      left side (looking a bit like the numbers "2", "3", and "6" and rotated
      versions of the same).
    • Peter Cyrus
      It looks nice. How do you show voiced fricatives? How about h?
      Message 2 of 7 , Nov 3, 2011
      • 0 Attachment
        It looks nice.

        How do you show voiced fricatives? How about h?

        On Thu, Nov 3, 2011 at 3:20 AM, Herman Miller <hmiller@...> wrote:

        > I'm continuing to work on my new alphabet project, Teascript. Since it's
        > hard to talk about it without an illustration, I put up an image of one way
        > Teascript might be used to write English vowels.
        >
        > http://www.prismnet.com/~**hmiller/png/teascript-lexical-**sets.png<http://www.prismnet.com/~hmiller/png/teascript-lexical-sets.png>
        >
        > I'm still working out the details for the shapes of the letters, but I've
        > got a rough plan that seems to be working. Basically I'm looking for a
        > featural script that's easy to write by hand and read, and takes up about
        > the same amount of space on a page as more traditional alphabets. Some of
        > the letters may look a bit like Tengwar, but the effect of the whole script
        > looks a bit different.
        >
        > The general plan is that sounds in the front of the mouth have a vertical
        > stroke on the left side, and sounds in the back of the mouth have one on
        > the right side. Ascenders and descenders are added to distinguish different
        > points of articulation according to the language. In English (and most
        > languages), bilabial and labiodental sounds have a descender on the left.
        > An ascender on the left can have different meanings in different languages;
        > I'm using it for dental fricatives in English and alveolar affricates in
        > Tirelat. Hindi could use it for retroflex consonants.
        >
        > In English, voiced stops have a single horizontal stroke attached to the
        > vertical stroke, so the letter for /d/ looks a bit like "r" and the letter
        > for /b/ like "ɼ". Voiceless stops continue the horizontal stroke into a
        > vertical stroke (a single stroke like "7"), so that the letter /t/ looks
        > like "n". Voiceless fricatives have a curved stroke with a corner that
        > looks a bit like the number "2". But a different language might have a
        > different mapping for these sets of letters. Basically I have four sets for
        > stops, one for nasals, and two for fricatives.
        >
        > Vowels are basically in 3 sets: front, central, and back, with 3 heights
        > in each set (a circle of 8 vowels with one in the middle). There are also 3
        > extra vowel letters which may be used as needed, and a length mark.
        >
        > In this way I can keep the number of letters to a minimum while still
        > keeping the featural aspects of the script. I've actually managed to assign
        > most of the common letters to the alphabetic keys on the keyboard (with or
        > without the shift key), but I've had to use a few punctuation keys for some
        > less common letters (like the ones I'm using for English /θ/ and /ð/).
        >
        > Teascript also has a small set of "miscellaneous" letters that can be used
        > for liquids, approximants, or other sounds that don't fit into the overall
        > table of phonemes for a language. These are represented by letters with
        > curved strokes without a vertical stroke on the right or left side (looking
        > a bit like the numbers "2", "3", and "6" and rotated versions of the same).
        >
      • Herman Miller
        ... The set that I m using for voiced fricatives in languages like English and Jarda has a horizontal stroke connected to the vertical stroke, and a curved
        Message 3 of 7 , Nov 3, 2011
        • 0 Attachment
          On 11/3/2011 11:35 AM, Peter Cyrus wrote:
          > It looks nice.
          >
          > How do you show voiced fricatives? How about h?

          The set that I'm using for voiced fricatives in languages like English
          and Jarda has a horizontal stroke connected to the vertical stroke, and
          a curved stroke attached to the end of the horizontal stroke. The 4th
          and 6th characters of the first word in the Relay 18 text in Jarda
          (kɛlzevo) are voiced fricatives. But in other languages, that set might
          be used for aspirated or ejective fricatives, or implosive stops, or
          anything else according to the needs of the language.

          http://www.prismnet.com/~hmiller/png/teascript-relay-18-jarda.png

          The upper half shows the original Njörr (or Ljörr) script, which was one
          of my first attempts at a featural script that actually wasn't half bad.
          It doesn't look as nice as Teascript, but I've had some practice since
          then in designing alphabets and making fonts for them.

          The text in IPA:

          kɛlzevo ɟo kipvɔ reɬvod diɹ.
          liɹvi runi ɡru, ɲazɤs kraɣ reɬ.
          ɣelra vɤɹa zɛl ʑɹeɲna nɤ raɕ au keɡ au pin,
          ʑe pinsicaɡ loxom, cødadɔl zulɹema lɔɡe jɛz.
          zulɹemvɔ ste wɛl ɲer zabe nev
          nɤ ɮizɤs kez ni ʎɛv neslø baɹ ɮan,
          ni ɹɛl ʑɯɣ au mavlø,
          ni meɮ syn stonta ɟyn ʑe tɹaz.
          ʑin ɣøɹ ɕlw pɹapɹinvɔ, au fømmas vɤ jo.
          plen ɹɔxe lul kezɹa ni vɔɹ faɹvu ka ʑɛv ɬim deɡɔl vɤ.

          For English /h/, I'm using a voiceless fricative symbol that looks a bit
          like the symbol for Jupiter (♃). It's the character in the S-row with a
          descender on the right.

          I'll need an easier way to refer to these letters... Say the characters
          in the D-row (the ones I use for voiced stops in English) are labeled
          D1-D6, where D1-D3 are the ones with the vertical stroke on the left (1
          with a descender, 3 with an ascender) and D4-D6 have the vertical stroke
          on the right (4 with an ascender and 6 with a descender). So the
          character for /h/ in English is S6.

          So far the rows I'm using are the D-row (D2 looks like "r"), T-row (T2
          looks like "n"), Dh-row (Dh2 looks like a sort of "rr" ligature), Th-row
          (Th2 looks like "m"), N-row (N1 looks like "p", N3 looks like "b"),
          S-row (S2 looks like a "12" ligature), and Z-row (Z2 looks a bit like
          "TC" joined together). Probably not many languages will use the Dh and
          Th rows, but they're useful for aspirated stops, and potentially for
          palatalized or velarized sounds in some languages. Lindiga could use
          them for its lateral fricatives.
        • BPJ
          ... If the letters can easily be arranged in a grid alla Tengwar it makes sense to label columns with letters ond rows with numbers like longitudes and
          Message 4 of 7 , Nov 4, 2011
          • 0 Attachment
            On 2011-11-04 04:18, Herman Miller wrote:
            > I'll need an easier way to refer to these letters... Say the
            > characters in the D-row (the ones I use for voiced stops in
            > English) are labeled D1-D6, where D1-D3 are the ones with the
            > vertical stroke on the left (1 with a descender, 3 with an
            > ascender) and D4-D6 have the vertical stroke on the right (4 with
            > an ascender and 6 with a descender). So the character for /h/ in
            > English is S6.

            If the letters can easily be arranged in a grid alla
            Tengwar it makes sense to label columns with letters
            ond rows with numbers like longitudes and latitudes on
            a map and refer to them with a0, a1, a2... Thus if you
            have a chart you can even use the codes as an input
            method with a vim keymap if you are of that bent: each
            /[a-z]\+[0-9]\+/ then corresponds unambiguously to one
            character, regardless of how many columns/rows you
            have: a0, z12 and bb4 work equally well, since each
            instance of a letter following a number (or whitespace)
            marks the beginning of a new code.

            /bpj
          • Herman Miller
            ... The main part of the consonant chart does fit nicely onto a grid like Tengwar. http://www.prismnet.com/~hmiller/png/teascript-chart.png I could just label
            Message 5 of 7 , Nov 4, 2011
            • 0 Attachment
              On 11/4/2011 7:29 PM, BPJ wrote:
              > On 2011-11-04 04:18, Herman Miller wrote:
              >> I'll need an easier way to refer to these letters... Say the
              >> characters in the D-row (the ones I use for voiced stops in
              >> English) are labeled D1-D6, where D1-D3 are the ones with the
              >> vertical stroke on the left (1 with a descender, 3 with an
              >> ascender) and D4-D6 have the vertical stroke on the right (4 with
              >> an ascender and 6 with a descender). So the character for /h/ in
              >> English is S6.
              >
              > If the letters can easily be arranged in a grid alla
              > Tengwar it makes sense to label columns with letters
              > ond rows with numbers like longitudes and latitudes on
              > a map and refer to them with a0, a1, a2... Thus if you
              > have a chart you can even use the codes as an input
              > method with a vim keymap if you are of that bent: each
              > /[a-z]\+[0-9]\+/ then corresponds unambiguously to one
              > character, regardless of how many columns/rows you
              > have: a0, z12 and bb4 work equally well, since each
              > instance of a letter following a number (or whitespace)
              > marks the beginning of a new code.
              >
              > /bpj

              The main part of the consonant chart does fit nicely onto a grid like
              Tengwar.

              http://www.prismnet.com/~hmiller/png/teascript-chart.png

              I could just label the rows "A, B, C, D, E, F, G" instead of "D, T, Dh,
              Th, N, S, Z". The miscellaneous consonants could keep their "L, R, W"
              labels or I could continue with "H, I, J".

              I think it's easier to remember "D-row" for languages where D2
              represents /d/ (i.e., most languages). But it's /t/ in Eastern Armenian
              vs. /d/ in Western Armenian. In some other languages D2 might be /t/
              while T2 is some other t-related sound (/tʰ/ or /tʼ/).
            • BPJ
              ... Yes, you are absolutely right of course! So in my Tengwar keymap I might rename the consonant columns (t,p,c,k) -- good thing the map is generated by a
              Message 6 of 7 , Nov 5, 2011
              • 0 Attachment
                On 2011-11-05 02:49, Herman Miller wrote:
                > On 11/4/2011 7:29 PM, BPJ wrote:
                >> On 2011-11-04 04:18, Herman Miller wrote:
                >>> I'll need an easier way to refer to these letters... Say the
                >>> characters in the D-row (the ones I use for voiced stops in
                >>> English) are labeled D1-D6, where D1-D3 are the ones with the
                >>> vertical stroke on the left (1 with a descender, 3 with an
                >>> ascender) and D4-D6 have the vertical stroke on the right (4 with
                >>> an ascender and 6 with a descender). So the character for /h/ in
                >>> English is S6.
                >>
                >> If the letters can easily be arranged in a grid alla
                >> Tengwar it makes sense to label columns with letters
                >> ond rows with numbers like longitudes and latitudes on
                >> a map and refer to them with a0, a1, a2... Thus if you
                >> have a chart you can even use the codes as an input
                >> method with a vim keymap if you are of that bent: each
                >> /[a-z]\+[0-9]\+/ then corresponds unambiguously to one
                >> character, regardless of how many columns/rows you
                >> have: a0, z12 and bb4 work equally well, since each
                >> instance of a letter following a number (or whitespace)
                >> marks the beginning of a new code.
                >>
                >> /bpj
                >
                > The main part of the consonant chart does fit nicely onto a grid
                > like Tengwar.
                >
                > http://www.prismnet.com/~hmiller/png/teascript-chart.png
                >
                > I could just label the rows "A, B, C, D, E, F, G" instead of "D,
                > T, Dh, Th, N, S, Z". The miscellaneous consonants could keep their
                > "L, R, W" labels or I could continue with "H, I, J".
                >
                > I think it's easier to remember "D-row" for languages where D2
                > represents /d/ (i.e., most languages). But it's /t/ in Eastern
                > Armenian vs. /d/ in Western Armenian. In some other languages D2
                > might be /t/ while T2 is some other t-related sound (/tʰ/ or /tʼ/).
                >

                Yes, you are absolutely right of course! So in my Tengwar keymap
                I might rename the consonant columns (t,p,c,k) -- good thing the
                map is generated by a Perl script... ;-)

                /bpj
              • Charles W Brickner
                ... From: Constructed Languages List [mailto:CONLANG@listserv.brown.edu] On Behalf Of BPJ Sent: Saturday, November 05, 2011 8:07 AM To:
                Message 7 of 7 , Nov 5, 2011
                • 0 Attachment
                  -----Original Message-----
                  From: Constructed Languages List [mailto:CONLANG@...] On Behalf Of BPJ
                  Sent: Saturday, November 05, 2011 8:07 AM
                  To: CONLANG@...
                  Subject: Re: English lexical sets in Teascript

                  Yes, you are absolutely right of course! So in my Tengwar keymap I might rename the consonant columns (t,p,c,k) -- good thing the map is generated by a Perl script... ;-)
                  /bpj


                  I have a Tengwar-type table for the Senjecan consonants. The order, however is p,t,c,k, the order of POA from front to back: labial, dental, alveolar, palatal.

                  Speaking of Tengwar, I'd appreciate some recommendations on a site of the Tengwar symbols to download. Thanks.

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