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

Re: Recreating Pakuni language from on-screen examples

Expand Messages
  • Thomas Alexander
    Hi Jim, I know it s not quite conlanger cool to blur the distinction between sounds and letters, so I ll do my best - although it s still my custom to talk
    Message 1 of 28 , Aug 3, 2009
    • 0 Attachment
      Hi Jim,

      I know it's not quite conlanger cool to blur
      the distinction between sounds and letters, so
      I'll do my best - although it's still my custom
      to talk about "the letter S" rather than "an
      unvoiced dental sibilant" or whatever they're
      calling them these days.

      > Is it clearly pronounced /ogansa/ in other
      > context by the same or other actors? Or does
      > "ogan(s|z)a" only occur in that one phrase
      > throughout the show?

      The word is used in several contexts and I've
      only noted one example where the S is unvoiced.
      My impression is that it's only spelled wth a
      Z in certain fan web pages. The official sources
      I've managed to dig up spell it with an S --
      which suggests to me that it was intended to be
      unvoiced.

      > > 1 - declare that the word really is
      > > pronounced with an unvoiced s.
      >
      > That would make sense if /ogansa/ is
      > attestested in other contexts.

      How about if in only one?

      I think it also depends on whether we imagine
      that we're creating a guide for people watching
      the original series, or a guide for the creation
      of Pakuni dialog in future works.

      > > 2 - declare that "s" can be voiced or unvoiced.
      >
      > That might make sense; if Pakuni has only
      > phonemic /z/, then [z] is a plausible,
      > probable allophone of /s/.

      I do like that idea - although it strikes me
      as odd that the allophone turns up in only one
      context.

      > > 3 - declare that "s" can be (or perhaps
      > > "always is") voiced after a nasal.
      >
      > Are there any other words where /s/ occurs in a
      > cluster with /n/, or anther voiced consonant?

      Possibly. I thought of a possible example yesterday.
      The tyranosaur in the show is sometimes called "abobo"
      sometimes "akingo". It's possible that "abobo" is a
      general word for dinosaur (since it's used to describe
      a brontosaurus), but some have suggested that "abobo"
      means tyranosaur and "akingo" is this tyranosaur's name.
      Since his name is "Grumpy" in English, "akingo" would
      mean "grumpy one" -- which would make "kingsa" an
      adjective meaning "grumpy."

      Of course, even if this dubious(*) speculation is
      correct, I'm inclined to pronounce "kingsa" with
      an unvoiced S.

      (*) if abobo doesn't mean specifically tyranosaur,
      it's unusual that it's usually sufficient to describe
      a tyranosaur, while other dino species are mentioned
      by name. If it does mean tyranosaur, it's odd to say
      that a brontosaurus is a big tyranosaur. Perhaps we
      are touching on areas where Pakuni semantic bountries
      are different from those in English.


      > > 4 - introduce the letter Z to your Pakuni
      > > alphabet and respell the word "oganza" with a Z.
      >
      > The orthography is a separate issue

      I meant, introduce a the sound "z" to your Pakuni
      phonemic inventory and treat words pronounced with
      voiced vs unvoiced sibilants as distinct words.

      > and would depend on your choice of solutions to
      > the underlying phonological question. But if [z]
      > is an allophone of /s/, I would not expect the
      > orthography to represent it with a separate letter.

      Right, that's what I meant. Number 4 would treat
      [z] and /s/ (*) as non-allophonic.

      (*) I'm copying your terms here and trusting on
      faith that we're talking about the same things -
      the sounds made by Z and S in Esperanto.

      Amike salutas,
      Thomas/Tomaso ALEXANDER.
      www.NightinGael.Net
      ---Anything below this line is not from Thomas ---
    • Mark J. Reed
      [s] and [z] refer to the sounds made by Esperanto and , yes. /s/ and /z/ refer to sounds which are distinct from each other within some language under
      Message 2 of 28 , Aug 3, 2009
      • 0 Attachment
        [s] and [z] refer to the sounds made by Esperanto <s> and <z>, yes.

        /s/ and /z/ refer to sounds which are distinct from each other within
        some language under discussion, of which [s] and [z] are somehow
        representative.

        If Pakuni has only /s/ and no /z/, it might nevertheless have [z] as
        an allophone of /s/, much like Spanish. So if [oganza] is the only
        example of a [z], and we have no instance of [ogansa] meaning
        something different, then it seems silly to postulate that Pakuni has
        a /z/ phoneme.
      • Jim Henry
        ... Se ni parolus en Esperanto, kaj parolus pri lingvo kiu havas ĝenerale nur fonemoj kiuj ankaŭ troviĝas en Esperanto, ne estus la sama danĝero de
        Message 3 of 28 , Aug 3, 2009
        • 0 Attachment
          On Mon, Aug 3, 2009 at 2:05 PM, Thomas Alexander<salivanto@...> wrote:

          > I know it's not quite conlanger cool to blur
          > the distinction between sounds and letters, so
          > I'll do my best - although it's still my custom
          > to talk about "the letter S" rather than "an
          > unvoiced dental sibilant" or whatever they're
          > calling them these days.

          Se ni parolus en Esperanto, kaj parolus pri lingvo kiu havas ĝenerale
          nur fonemoj kiuj ankaŭ troviĝas en Esperanto, ne estus la sama danĝero
          de malklareco, kiel se ni parolus en la angla pri fonemoj dirante
          "letters". .... But in English, where the letter "s" can sometimes
          represent /z/ as well as /s/ (though "z" almost always represents
          /z/), using the word "letter" when you mean to talk about phonemes can
          potentially cause confusion, especially when you might, elsewhere in
          the same conversation, want to talk about writing as well as
          pronunciation (as when you talked about how the word was spelling in
          fan sites vs. official sources). It's possible that there is some
          "conlanger cool" going on as well, but I'll try to avoid it.

          >> Is it clearly pronounced /ogansa/ in other
          >> context by the same or other actors?  Or does
          >> "ogan(s|z)a" only occur in that one phrase
          >> throughout the show?
          >
          > The word is used in several contexts and I've
          > only noted one example where the S is unvoiced.
          > My impression is that it's only spelled wth a
          > Z in certain fan web pages.  The official sources
          > I've managed to dig up spell it with an S --
          > which suggests to me that it was intended to be
          > unvoiced.
          >
          >> > 1 - declare that the word really is
          >> > pronounced with an unvoiced s.
          >>
          >> That would make sense if /ogansa/ is
          >> attestested in other contexts.
          >
          > How about if in only one?
          >
          > I think it also depends on whether we imagine
          > that we're creating a guide for people watching
          > the original series, or a guide for the creation
          > of Pakuni dialog in future works.

          Partly, yes. But even if creating a guide for future use of Pakuni,
          I'd be inclined, if the evidence on paper about the conlang Dr.
          Fromkin mean to create and the evidence from the show about the
          language the fictional Pakuni speak are in conflict, to go with the
          latter *if the latter is consistent within itself*. It seems best
          that a potential continuation of the show in whatever medium would
          have a Pakuni language that's consistent with the Pakuni language as
          attested in the original show.

          So if the actors pronounce it /oganza/ on all but one occasion, or in
          all contexts but one, then it seems that actors playing Pakuni in
          future shows should probably be told to pronounce it /oganza/ as well.
          Whether new word coinages which have /n/ followed by a fricative in
          the neighborhood of /s/ or /z/ should be pronounced with /nz/ or /ns/
          is less clear, but I'd tend to go with the former based on everything
          you've said.

          Here:

          >> > 2 - declare that "s" can be voiced or unvoiced.
          >>
          >> That might make sense; if Pakuni has only
          >> phonemic /z/, then [z] is a plausible,
          >> probable allophone of /s/.

          I had a typo: I meant to say "...if Pakuni has only phonemic
          /s/, then ...."

          > I do like that idea - although it strikes me
          > as odd that the allophone turns up in only one
          > context.

          Not necessarily odd...

          >> > 3 - declare that "s" can be (or perhaps
          >> > "always is") voiced after a nasal.
          >>
          >> Are there any other words where /s/ occurs in a
          >> cluster with /n/, or anther voiced consonant?
          >
          > Possibly.  I thought of a possible example yesterday.
          > The tyranosaur in the show is sometimes called "abobo"
          > sometimes "akingo".  It's possible that "abobo" is a
          > general word for dinosaur (since it's used to describe
          > a brontosaurus), but some have suggested that "abobo"
          > means tyranosaur and "akingo" is this tyranosaur's name.
          > Since his name is "Grumpy" in English, "akingo" would
          > mean "grumpy one" -- which would make "kingsa" an
          > adjective meaning "grumpy."

          Except that "kingsa" is not actually heard anywhere in the show's
          dialogue, right?

          In that case, if all the other known words with [s] in them have it at
          the beginning or end of a word adjacent to a vowel, or between two
          vowels, or clustered with an unvoiced consonant, then it's not
          actually surprising that the only time [z] is heard is when it's
          adjacent to /n/.

          >> > 4 - introduce the letter Z to your Pakuni
          >> > alphabet and respell the word "oganza" with a Z.
          >>
          >> The orthography is a separate issue
          >
          > I meant, introduce a the sound "z" to your Pakuni
          > phonemic inventory and treat words pronounced with
          > voiced vs unvoiced sibilants as distinct words.

          OK. No, I wouldn't necessarily recommend that, based on everything
          you've said. But how the word is spelled in materials intended to
          accurately represent the nature of the language itself, and how it's
          spelled in dialogue intended to be given to actors who might or might
          not have dialect coaches telling them that orthographic "s" is
          pronounced [s] in these contexts and [z] in this context .... that's a
          separate issue again. I think there's been some mention in previous
          threads about actors in some Star Trek episodes and/or films
          mispronouncing Klingon dialogue based on a naive intepretation of its
          romanization in terms of English orthography.

          In general, I think Mark explained things a little better than I did:

          On Mon, Aug 3, 2009 at 4:10 PM, Mark J. Reed<markjreed@...> wrote:
          > [s] and [z] refer to the sounds made by Esperanto <s> and <z>, yes.
          >
          > /s/ and /z/ refer to sounds which are distinct from each other within
          > some language under discussion, of which [s] and [z] are somehow
          > representative.
          >
          > If Pakuni has only /s/ and no /z/, it might nevertheless have [z] as
          > an allophone of /s/, much like Spanish. So if [oganza] is the only
          > example of a [z], and we have no instance of [ogansa] meaning
          > something different, then it seems silly to postulate that Pakuni has
          > a /z/ phoneme.
        • Mark J. Reed
          English has, what, about 40 phonemes, depending on how you count? But only 26 letters. To the extent that English orthography is phonemic, we make up the
          Message 4 of 28 , Aug 3, 2009
          • 0 Attachment
            English has, what, about 40 phonemes, depending on how you count? But
            only 26 letters. To the extent that English orthography is phonemic,
            we make up the difference by overloading letters and using digraphs -
            which are themselves overloaded ("ch" is the usual way to identify the
            phoneme spelled /tS/ in CXS, but it can also spell /S/ or /k/ or /x/).
            So in general it's a good idea to keep the concepts separate - after
            all, "ch" is a phoneme of English, but one would never call it a
            "letter" in that language (even though they used to do so in Spanish).
            Anyway, the question is whether or not a Paku speaker would hear
            "oganza" and "ogansa" as the same word or two different words. The
            evidence seems to be in favor of the same, which means [s] and [z] are
            allophones of a single phoneme, which we might as well follow the
            conventional orthography and call /s/.

            On 8/3/09, Jim Henry <jimhenry1973@...> wrote:
            > On Mon, Aug 3, 2009 at 2:05 PM, Thomas Alexander<salivanto@...> wrote:
            >
            >> I know it's not quite conlanger cool to blur
            >> the distinction between sounds and letters, so
            >> I'll do my best - although it's still my custom
            >> to talk about "the letter S" rather than "an
            >> unvoiced dental sibilant" or whatever they're
            >> calling them these days.
            >
            > Se ni parolus en Esperanto, kaj parolus pri lingvo kiu havas ĝenerale
            > nur fonemoj kiuj ankaŭ troviĝas en Esperanto, ne estus la sama danĝero
            > de malklareco, kiel se ni parolus en la angla pri fonemoj dirante
            > "letters". .... But in English, where the letter "s" can sometimes
            > represent /z/ as well as /s/ (though "z" almost always represents
            > /z/), using the word "letter" when you mean to talk about phonemes can
            > potentially cause confusion, especially when you might, elsewhere in
            > the same conversation, want to talk about writing as well as
            > pronunciation (as when you talked about how the word was spelling in
            > fan sites vs. official sources). It's possible that there is some
            > "conlanger cool" going on as well, but I'll try to avoid it.
            >
            >>> Is it clearly pronounced /ogansa/ in other
            >>> context by the same or other actors?  Or does
            >>> "ogan(s|z)a" only occur in that one phrase
            >>> throughout the show?
            >>
            >> The word is used in several contexts and I've
            >> only noted one example where the S is unvoiced.
            >> My impression is that it's only spelled wth a
            >> Z in certain fan web pages.  The official sources
            >> I've managed to dig up spell it with an S --
            >> which suggests to me that it was intended to be
            >> unvoiced.
            >>
            >>> > 1 - declare that the word really is
            >>> > pronounced with an unvoiced s.
            >>>
            >>> That would make sense if /ogansa/ is
            >>> attestested in other contexts.
            >>
            >> How about if in only one?
            >>
            >> I think it also depends on whether we imagine
            >> that we're creating a guide for people watching
            >> the original series, or a guide for the creation
            >> of Pakuni dialog in future works.
            >
            > Partly, yes. But even if creating a guide for future use of Pakuni,
            > I'd be inclined, if the evidence on paper about the conlang Dr.
            > Fromkin mean to create and the evidence from the show about the
            > language the fictional Pakuni speak are in conflict, to go with the
            > latter *if the latter is consistent within itself*. It seems best
            > that a potential continuation of the show in whatever medium would
            > have a Pakuni language that's consistent with the Pakuni language as
            > attested in the original show.
            >
            > So if the actors pronounce it /oganza/ on all but one occasion, or in
            > all contexts but one, then it seems that actors playing Pakuni in
            > future shows should probably be told to pronounce it /oganza/ as well.
            > Whether new word coinages which have /n/ followed by a fricative in
            > the neighborhood of /s/ or /z/ should be pronounced with /nz/ or /ns/
            > is less clear, but I'd tend to go with the former based on everything
            > you've said.
            >
            > Here:
            >
            >>> > 2 - declare that "s" can be voiced or unvoiced.
            >>>
            >>> That might make sense; if Pakuni has only
            >>> phonemic /z/, then [z] is a plausible,
            >>> probable allophone of /s/.
            >
            > I had a typo: I meant to say "...if Pakuni has only phonemic
            > /s/, then ...."
            >
            >> I do like that idea - although it strikes me
            >> as odd that the allophone turns up in only one
            >> context.
            >
            > Not necessarily odd...
            >
            >>> > 3 - declare that "s" can be (or perhaps
            >>> > "always is") voiced after a nasal.
            >>>
            >>> Are there any other words where /s/ occurs in a
            >>> cluster with /n/, or anther voiced consonant?
            >>
            >> Possibly.  I thought of a possible example yesterday.
            >> The tyranosaur in the show is sometimes called "abobo"
            >> sometimes "akingo".  It's possible that "abobo" is a
            >> general word for dinosaur (since it's used to describe
            >> a brontosaurus), but some have suggested that "abobo"
            >> means tyranosaur and "akingo" is this tyranosaur's name.
            >> Since his name is "Grumpy" in English, "akingo" would
            >> mean "grumpy one" -- which would make "kingsa" an
            >> adjective meaning "grumpy."
            >
            > Except that "kingsa" is not actually heard anywhere in the show's
            > dialogue, right?
            >
            > In that case, if all the other known words with [s] in them have it at
            > the beginning or end of a word adjacent to a vowel, or between two
            > vowels, or clustered with an unvoiced consonant, then it's not
            > actually surprising that the only time [z] is heard is when it's
            > adjacent to /n/.
            >
            >>> > 4 - introduce the letter Z to your Pakuni
            >>> > alphabet and respell the word "oganza" with a Z.
            >>>
            >>> The orthography is a separate issue
            >>
            >> I meant, introduce a the sound "z" to your Pakuni
            >> phonemic inventory and treat words pronounced with
            >> voiced vs unvoiced sibilants as distinct words.
            >
            > OK. No, I wouldn't necessarily recommend that, based on everything
            > you've said. But how the word is spelled in materials intended to
            > accurately represent the nature of the language itself, and how it's
            > spelled in dialogue intended to be given to actors who might or might
            > not have dialect coaches telling them that orthographic "s" is
            > pronounced [s] in these contexts and [z] in this context .... that's a
            > separate issue again. I think there's been some mention in previous
            > threads about actors in some Star Trek episodes and/or films
            > mispronouncing Klingon dialogue based on a naive intepretation of its
            > romanization in terms of English orthography.
            >
            > In general, I think Mark explained things a little better than I did:
            >
            > On Mon, Aug 3, 2009 at 4:10 PM, Mark J. Reed<markjreed@...> wrote:
            >> [s] and [z] refer to the sounds made by Esperanto <s> and <z>, yes.
            >>
            >> /s/ and /z/ refer to sounds which are distinct from each other within
            >> some language under discussion, of which [s] and [z] are somehow
            >> representative.
            >>
            >> If Pakuni has only /s/ and no /z/, it might nevertheless have [z] as
            >> an allophone of /s/, much like Spanish. So if [oganza] is the only
            >> example of a [z], and we have no instance of [ogansa] meaning
            >> something different, then it seems silly to postulate that Pakuni has
            >> a /z/ phoneme.
            >

            --
            Sent from my mobile device

            Mark J. Reed <markjreed@...>
          • Eric Christopherson
            ... It s possible that it s conditioned by something that shows up only once in the very small corpus of the show. It s also possible it really only occurs in
            Message 5 of 28 , Aug 3, 2009
            • 0 Attachment
              On Aug 3, 2009, at 1:05 PM, Thomas Alexander wrote:
              >> That might make sense; if Pakuni has only
              >> phonemic /z/, then [z] is a plausible,
              >> probable allophone of /s/.
              >
              > I do like that idea - although it strikes me
              > as odd that the allophone turns up in only one
              > context.

              It's possible that it's conditioned by something that shows up only
              once in the very small corpus of the show. It's also possible it
              really only occurs in one word, or a few.

              As natlang examples:

              Arabic has [l_e] (velarized or pharyngealized [l]) only in _Allah_ and
              its other forms.

              English has word-final [&(:)] in only a few words: _yeah_ most
              notably, and [w&(:)] "onomatopoeia for crying", ["d&(:)d&(:)] "daddy
              (baby talk)" and for some speakers also [n&(:)] meaning
              "no" (informal), and for some speakers [bl&(:)] "blah". Final [V(:)]
              only occurs (AFAIK) in _duh_ and _huh_. Of course, all of these are
              somewhat informal, and can be considered "expressive language", which
              commonly features sounds not found in other words (or in contexts
              where they don't occur elsewhere).
            • Lee
              For what it s worth, my sibs and I always have and still do pronounce it /oganza/, because that is what we always heard on the show. Listening to the a few
              Message 6 of 28 , Aug 3, 2009
              • 0 Attachment
                For what it's worth, my sibs and I always have and still do pronounce it /oganza/, because that is what we always heard on the show. Listening to the a few coworkers who admitted to watching the show after the movie came out, they all pronounce it the same.

                Lee

                --- On Mon, 8/3/09, Thomas Alexander <salivanto@...> wrote:

                From: Thomas Alexander <salivanto@...>
                Subject: Re: Recreating Pakuni language from on-screen examples
                To: CONLANG@...
                Date: Monday, August 3, 2009, 1:05 PM

                Hi Jim,

                I know it's not quite conlanger cool to blur
                the distinction between sounds and letters, so
                I'll do my best - although it's still my custom
                to talk about "the letter S" rather than "an
                unvoiced dental sibilant" or whatever they're
                calling them these days.

                > Is it clearly pronounced /ogansa/ in other
                > context by the same or other actors?  Or does
                > "ogan(s|z)a" only occur in that one phrase
                > throughout the show?

                The word is used in several contexts and I've
                only noted one example where the S is unvoiced.
                My impression is that it's only spelled wth a
                Z in certain fan web pages.  The official sources
                I've managed to dig up spell it with an S --
                which suggests to me that it was intended to be
                unvoiced.

                > > 1 - declare that the word really is
                > > pronounced with an unvoiced s.
                >
                > That would make sense if /ogansa/ is
                > attestested in other contexts.

                How about if in only one?

                I think it also depends on whether we imagine
                that we're creating a guide for people watching
                the original series, or a guide for the creation
                of Pakuni dialog in future works.

                > > 2 - declare that "s" can be voiced or unvoiced.
                >
                > That might make sense; if Pakuni has only
                > phonemic /z/, then [z] is a plausible,
                > probable allophone of /s/.

                I do like that idea - although it strikes me
                as odd that the allophone turns up in only one
                context.

                > > 3 - declare that "s" can be (or perhaps
                > > "always is") voiced after a nasal.
                >
                > Are there any other words where /s/ occurs in a
                > cluster with /n/, or anther voiced consonant?

                Possibly.  I thought of a possible example yesterday.
                The tyranosaur in the show is sometimes called "abobo"
                sometimes "akingo".  It's possible that "abobo" is a
                general word for dinosaur (since it's used to describe
                a brontosaurus), but some have suggested that "abobo"
                means tyranosaur and "akingo" is this tyranosaur's name.
                Since his name is "Grumpy" in English, "akingo" would
                mean "grumpy one" -- which would make "kingsa" an
                adjective meaning "grumpy."

                Of course, even if this dubious(*) speculation is
                correct, I'm inclined to pronounce "kingsa" with
                an unvoiced S.

                (*) if abobo doesn't mean specifically tyranosaur,
                it's unusual that it's usually sufficient to describe
                a tyranosaur, while other dino species are mentioned
                by name.  If it does mean tyranosaur, it's odd to say
                that a brontosaurus is a big tyranosaur.  Perhaps we
                are touching on areas where Pakuni semantic bountries
                are different from those in English.


                > > 4 - introduce the letter Z to your Pakuni
                > > alphabet and respell the word "oganza" with a Z.
                >
                > The orthography is a separate issue

                I meant, introduce a the sound "z" to your Pakuni
                phonemic inventory and treat words pronounced with
                voiced vs unvoiced sibilants as distinct words.

                > and would depend on your choice of solutions to
                > the underlying phonological question.  But if [z]
                > is an allophone of /s/, I would not expect the
                > orthography to represent it with a separate letter.

                Right, that's what I meant.  Number 4 would treat
                [z] and /s/ (*) as non-allophonic.

                (*) I'm copying your terms here and trusting on
                faith that we're talking about the same things -
                the sounds made by Z and S in Esperanto.

                Amike salutas,
                Thomas/Tomaso ALEXANDER.
                www.NightinGael.Net
                ---Anything below this line is not from Thomas ---


                     
              • Alex Bicksler
                ... And possibly _wha?_, [wV:_R], but that might be even more informal than your examples. -_^
                Message 7 of 28 , Aug 3, 2009
                • 0 Attachment
                  >
                  > Final [V(:)] only occurs (AFAIK) in _duh_ and _huh_.

                  And possibly _wha?_,
                  [wV:_R], but that might be even more informal than your examples. -_^
                • Mark J. Reed
                  wha? , although nominally a shortened form of what? , can be conveniently lumped in with several variants of huh? with different onsets: uh? , bwuh? ,
                  Message 8 of 28 , Aug 4, 2009
                  • 0 Attachment
                    "wha?", although nominally a shortened form of "what?", can be
                    conveniently lumped in with several variants of "huh?" with different
                    onsets: "uh?", "bwuh?", etc.

                    I think "uh-uh" ("no") is another example of final /V/, however.

                    On Tue, Aug 4, 2009 at 1:44 AM, Alex Bicksler<burtonlang@...> wrote:
                    >>
                    >> Final [V(:)] only occurs (AFAIK) in _duh_ and _huh_.
                    >
                    > And possibly _wha?_,
                    > [wV:_R], but that might be even more informal than your examples. -_^
                    >



                    --
                    Mark J. Reed <markjreed@...>
                  • Paul Hartzer
                    For me, these words end in the same vowel as gladiola and victrola, to name two that come to mind. -- Paul
                    Message 9 of 28 , Aug 4, 2009
                    • 0 Attachment
                      For me, these words end in the same vowel as "gladiola" and "victrola," to name two that come to mind.

                      -- Paul



                      ----- Original Message ----
                      > From: Mark J. Reed <markjreed@...>
                      > To: CONLANG@...
                      > Sent: Tuesday, August 4, 2009 8:03:45 AM
                      > Subject: Re: Recreating Pakuni language from on-screen examples
                      >
                      > "wha?", although nominally a shortened form of "what?", can be
                      > conveniently lumped in with several variants of "huh?" with different
                      > onsets: "uh?", "bwuh?", etc.
                      >
                      > I think "uh-uh" ("no") is another example of final /V/, however.
                      >
                      > On Tue, Aug 4, 2009 at 1:44 AM, Alex Bickslerwrote:
                      > >>
                      > >> Final [V(:)] only occurs (AFAIK) in _duh_ and _huh_.
                      > >
                      > > And possibly _wha?_,
                      > > [wV:_R], but that might be even more informal than your examples. -_^
                      > >
                      >
                      >
                      >
                      > --
                      > Mark J. Reed
                    • Mark J. Reed
                      ... But as it s not stressed in such words, it may be analyzed as /@/. It is at least sometimes stressed in the utterances in question, and so must (in the
                      Message 10 of 28 , Aug 4, 2009
                      • 0 Attachment
                        On Tue, Aug 4, 2009 at 11:18 AM, Paul Hartzer<paulhartzer@...> wrote:
                        > For me, these words end in the same vowel as "gladiola" and "victrola," to name two that come to mind.

                        But as it's not stressed in such words, it may be analyzed as /@/.
                        It is at least sometimes stressed in the utterances in question, and
                        so must (in the usual analysis, where /@/ is definitionally
                        unstressed) be analyzed as /V/.



                        >
                        > -- Paul
                        >
                        >
                        >
                        > ----- Original Message ----
                        >> From: Mark J. Reed <markjreed@...>
                        >> To: CONLANG@...
                        >> Sent: Tuesday, August 4, 2009 8:03:45 AM
                        >> Subject: Re: Recreating Pakuni language from on-screen examples
                        >>
                        >> "wha?", although nominally a shortened form of "what?",  can be
                        >> conveniently lumped in with several variants of "huh?" with different
                        >> onsets: "uh?", "bwuh?", etc.
                        >>
                        >> I think "uh-uh" ("no") is another example of final /V/, however.
                        >>
                        >> On Tue, Aug 4, 2009 at 1:44 AM, Alex Bickslerwrote:
                        >> >>
                        >> >> Final [V(:)] only occurs (AFAIK) in _duh_ and _huh_.
                        >> >
                        >> > And possibly _wha?_,
                        >> > [wV:_R], but that might be even more informal than your examples. -_^
                        >> >
                        >>
                        >>
                        >>
                        >> --
                        >> Mark J. Reed
                        >
                        >
                        >
                        >
                        >



                        --
                        Mark J. Reed <markjreed@...>
                      • Paul Hartzer
                        It *may* be, but why not analyze it as /V/ given that /V/ exists elsewhere? I d prefer to avoid /@/ altogether, and have [@] always an allophone of some other
                        Message 11 of 28 , Aug 4, 2009
                        • 0 Attachment
                          It *may* be, but why not analyze it as /V/ given that /V/ exists elsewhere? I'd prefer to avoid /@/ altogether, and have [@] always an allophone of some other phonemic vowel. /V/ -> [@] when unstressed seems like a reasonable enough rule, given that /vowel/ in general -> [@] when unstressed (with exceptions).

                          Also, the traditional use of V and C for vowel and consonant in IPA rules illustrates why V and C are bad choices for specific phones in X-SAMPA and derivatives.

                          -- Paul



                          ----- Original Message ----
                          > From: Mark J. Reed <markjreed@...>
                          > To: CONLANG@...
                          > Sent: Tuesday, August 4, 2009 12:03:26 PM
                          > Subject: Re: Recreating Pakuni language from on-screen examples
                          >
                          > On Tue, Aug 4, 2009 at 11:18 AM, Paul Hartzerwrote:
                          > > For me, these words end in the same vowel as "gladiola" and "victrola," to
                          > name two that come to mind.
                          >
                          > But as it's not stressed in such words, it may be analyzed as /@/.
                          > It is at least sometimes stressed in the utterances in question, and
                          > so must (in the usual analysis, where /@/ is definitionally
                          > unstressed) be analyzed as /V/.
                          >
                          >
                          >
                          > >
                          > > -- Paul
                          > >
                          > >
                          > >
                          > > ----- Original Message ----
                          > >> From: Mark J. Reed
                          > >> To: CONLANG@...
                          > >> Sent: Tuesday, August 4, 2009 8:03:45 AM
                          > >> Subject: Re: Recreating Pakuni language from on-screen examples
                          > >>
                          > >> "wha?", although nominally a shortened form of "what?", can be
                          > >> conveniently lumped in with several variants of "huh?" with different
                          > >> onsets: "uh?", "bwuh?", etc.
                          > >>
                          > >> I think "uh-uh" ("no") is another example of final /V/, however.
                          > >>
                          > >> On Tue, Aug 4, 2009 at 1:44 AM, Alex Bickslerwrote:
                          > >> >>
                          > >> >> Final [V(:)] only occurs (AFAIK) in _duh_ and _huh_.
                          > >> >
                          > >> > And possibly _wha?_,
                          > >> > [wV:_R], but that might be even more informal than your examples. -_^
                          > >> >
                          > >>
                          > >>
                          > >>
                          > >> --
                          > >> Mark J. Reed
                          > >
                          > >
                          > >
                          > >
                          > >
                          >
                          >
                          >
                          > --
                          > Mark J. Reed
                        • Thomas Alexander
                          Mark J. Reed ... Okay, so now I m confused. I thought I d be a good newbie and follow along on the chart when I read your note, but I
                          Message 12 of 28 , Aug 4, 2009
                          • 0 Attachment
                            "Mark J. Reed" <markjreed@...>
                            > [s] and [z] refer to the sounds made by Esperanto
                            > <s> and <z>, yes.
                            >
                            > /s/ and /z/ refer to sounds which are distinct
                            > from each other within some language under
                            > discussion, of which [s] and [z] are somehow
                            > representative.

                            Okay, so now I'm confused. I thought I'd be a good
                            newbie and follow along on the chart when I read your
                            note, but I find that the links from MYCONLANGLINKS.TK
                            which seem most promising are either 404 or don't
                            include the symbols you're using.

                            Reading through this and Jim Henry's reply, I'm
                            ready to guess that [s] indicates the actual sound
                            made, /s/ indicates a native speakers conception of
                            the sound which can vary in context with other sounds,
                            and <s> refers to the actual letter.

                            > If Pakuni has only /s/ and no /z/, it might
                            > nevertheless have [z] as an allophone of /s/,
                            > much like Spanish. So if [oganza] is the only
                            > example of a [z], and we have no instance of
                            > [ogansa] meaning something different, then it
                            > seems silly to postulate that Pakuni has
                            > a /z/ phoneme.

                            Silly, yes, but only slightly so. I could list several
                            examples where the pronunciation of CH and J are routinely
                            switched. At one point, I was ready to rewrite my
                            dictionary with one letter to represent [ch] and [j]
                            (I hope I got they symbols right), but I couldn't because
                            there are words such as CHI and JI which are obviously
                            distinct in meaning.

                            Jim Henry <jimhenry1973@...> wrote:

                            > But even if creating a guide for future use of Pakuni,
                            > I'd be inclined, if the evidence on paper about the
                            > conlang Dr. Fromkin mean to create and the evidence
                            > from the show about the language the fictional Pakuni
                            > speak are in conflict, to go with the latter *if the
                            > latter is consistent within itself*.

                            That sounds reasonable, and I think Lee's comment in
                            this thread backs it up. It's nearly always pronounced
                            /oganza/, and indeed, this is one of the most memorable
                            words in the series.

                            > Whether new word coinages which have /n/ followed by
                            > a fricative in the neighborhood of /s/ or /z/ should
                            > be pronounced with /nz/ or /ns/ is less clear, but
                            > I'd tend to go with the former based on everything
                            > you've said.

                            Do you think it matters if the <s> (letter s) is
                            part of the root (as in -gansa) or an afix (such as
                            in the hypothetical *yumansa" - humanish - built
                            from "yuman" and "sa")? In my mind it does.

                            > Except that "kingsa" is not actually heard
                            > anywhere in the show's dialogue, right?

                            Right. For that matter, it's not clear to me
                            how to treat the NG in akingo/kingsa. It's
                            pronounced /ng/+/g/ in the show (if that makes
                            sense), but if it's based on the hypothetical
                            word *kingsa*, I expect that <ng> would represent
                            one sound.

                            > In that case, if all the other known words with
                            > [s] in them have it at the beginning or end of
                            > a word adjacent to a vowel, or between two
                            > vowels, or clustered with an unvoiced consonant,
                            > then it's not actually surprising that the only
                            > time [z] is heard is when it's adjacent to /n/.

                            No, but this strikes me as an imported rule.
                            (Imported by the actors from their native language).
                            It's a rule I can live with, of course, but
                            that wasn't the question.

                            > [H]ow the word is spelled in materials intended to
                            > accurately represent the nature of the language
                            > itself, and how it's spelled in dialogue intended
                            > to be given to actors who might or might not have
                            > dialect coaches telling them that orthographic "s"
                            > is pronounced [s] in these contexts and [z] in this
                            > context .... that's a separate issue again.

                            I hadn't considered that. Of course, since the
                            Pakuni are never seen writing, I think it's
                            generally assumed that they don't have a native
                            writing system. (They do seem to understand what
                            writing is, though, but that's another point.)

                            While corresponding with another person interested
                            in Pakuni, I saw in his notes that [using Esperanto
                            spelling rules here] CXI is sometimes mispronounced
                            KAJ. My notes contained the words <chi> and <ka>
                            with the same meaning, but how CHI could becoem KAJ
                            was not at all clear, so I didn't believe it at first.
                            Then I realised that <CHI> and /kai/ are ways of
                            spelling and pronouncing the name of the Greek
                            letter X in certain parts of the world where the
                            actors (or their coaches) could have drawn their
                            influence.

                            > In general, I think Mark explained things a
                            > little better than I did:

                            [snip quote from Mark, already inlcluded above.]

                            If it makes you feel any better, I didn't start to
                            understand Mark's note till I started working my
                            way through yours.


                            Mark J. Reed" <markjreed@...>
                            > Anyway, the question is whether or not a Paku
                            > speaker would hear "oganza" and "ogansa" as
                            > the same word or two different words.

                            I'm inclined to agree with you that the
                            evidence seems to be in favor of the same...
                            but considering that the expressions "me ji"
                            and "meche" are at times pronounced the same
                            yet perceived as different has dulled my
                            confidence a little. :-)

                            Amike salutas,
                            Thomas/Tomaso ALEXANDER.
                            www.NightinGael.Net
                            ---Anything below this line is not from Thomas ---
                          • Mark J. Reed
                            ... Sorry. ... Right on all counts. The slashes in /s/ mean that it is a phoneme . Speech consists of lots of different sounds, but not all the differences
                            Message 13 of 28 , Aug 4, 2009
                            • 0 Attachment
                              On Tue, Aug 4, 2009 at 7:06 PM, Thomas Alexander<salivanto@...> wrote:
                              > Okay, so now I'm confused.

                              Sorry.

                              > Reading through this and Jim Henry's reply, I'm
                              > ready to guess that [s] indicates the actual sound
                              > made, /s/ indicates a native speakers conception of
                              > the sound which can vary in context with other sounds,
                              > and <s> refers to the actual letter.

                              Right on all counts. The slashes in /s/ mean that it is a "phoneme".
                              Speech consists of lots of different sounds, but not all the
                              differences affect what the native speakers hear; the ones that are
                              are said to be "phonemic". Different types of sound changes are
                              phonemic in different languages; Anglophones hear the [p_h] in "pin"
                              and the [p] as in "spin" as "the same sound", so we say that English
                              has one phoneme /p/ that occurs in both words. To a Hindi speaker, the
                              two sounds are quite distinct, and Hindi has both /p/ and /p_h/
                              phonemes. Japanese, on the other hand, has one phoneme /t/ that
                              usually sounds like English [t] but sometimes (before the vowel [i])
                              sounds like [tS] (English "ch").

                              Different realizations of the same phoneme (like [p] and [p_h] in
                              English, or [t] and [tS] in Japanese) are called "allophones" of that
                              phoneme.

                              It's usual to use <...> to refer to orthographic representations, but
                              since that confuses some HTML-based mailers, |...| is also used.

                              >  I could list several examples where the pronunciation of CH and J are routinely
                              > switched.
                              That happens in English, too, really; we just adjust what we hear
                              based on expectations. Clearly, /tS/ and /dZ/ ("ch" and "j"- be
                              careful here; /j/ in IPA/CXS refers to the English "y" sound) are
                              distinct phonemes in English, as indicated by lots of "minimal pairs"
                              - pairs where that's the only difference: "choose" vs "Jews", "batch"
                              vs "badge",etc. But the difference between them is not always very
                              clear in actual connected speech.

                              > At one point, I was ready to rewrite my dictionary with one letter to represent [ch] and [j]
                              > (I hope I got they symbols right), but I couldn't because
                              > there are words such as CHI and JI which are obviously
                              > distinct in meaning.

                              So "chi" and "ji" constitute a minimal pair indicating that "ch" and
                              "j" (/tS/ and /dZ/) are distinct phonemes.


                              >> Whether new word coinages which have /n/ followed by
                              >> a fricative in the neighborhood of /s/ or /z/ should
                              >> be pronounced with /nz/ or /ns/ is less clear, but
                              >> I'd tend to go with the former based on everything
                              >> you've said.
                              >
                              > Do you think it matters if the <s> (letter s) is
                              > part of the root (as in -gansa) or an afix (such as
                              > in the hypothetical *yumansa" - humanish - built
                              > from "yuman" and "sa")?  In my mind it does.

                              Phonetic effects like this are not usually governed by morphosyntactic
                              considerations - the /s/ will be voiced in the vicinity of voiced
                              sounds, unvoiced otherwise, regardless of how the word splits into
                              morphemes. But there are exceptions, so it could go either way; I'd
                              base the analysis on how things actually sound on the show.

                              > Right.  For that matter, it's not clear to me
                              > how to treat the NG in akingo/kingsa.  It's
                              > pronounced /ng/+/g/ in the show (if that makes
                              > sense), but if it's based on the hypothetical
                              > word *kingsa*, I expect that <ng> would represent
                              > one sound.

                              The "ng" in English "sing" is a single sound (spelled /N/ in CXS, /ŋ/
                              in the IPA), while the "ng" in English "finger" is two sounds, /N/
                              followed by /g/.

                              So if <akingo> is [@kiNgo], the question is whether <kingsa> would be
                              [kiNsa], [kiNza], [kiNgsa], or [kiNgza]. I'd guess that /ŋ/ would be
                              just [N] when followed by a consonant, [Ng] when followed by a vowel.
                              It's likely that if [n] triggers voicing of /s/ to [z], then so would
                              [N] (and [m] for that matter), but not necessarily.


                              > While corresponding with another person interested
                              > in Pakuni, I saw in his notes that [using Esperanto
                              > spelling rules here] CXI is sometimes mispronounced
                              > KAJ.

                              Both of those are English pronunciations of the written word <chi>; in
                              fact, both pronunciations exist in English with separate meanings.
                              The Greek letter is usually /kaj/ (for Americans, anyway), while the
                              spiritual essence is usually /tSi/ (a close Anglicization of the
                              Mandarin).

                              --
                              Mark J. Reed <markjreed@...>
                            • Philip Newton
                              ... I ve usually seen the spiritual essence spelled the Pinyin way, though (_qi_). Cheers, Philip -- Philip Newton
                              Message 14 of 28 , Aug 5, 2009
                              • 0 Attachment
                                On Wed, Aug 5, 2009 at 02:47, Mark J. Reed<markjreed@...> wrote:
                                > Both of those are English pronunciations of the written word <chi>; in
                                > fact, both pronunciations exist in English with separate meanings.
                                > The Greek letter is usually /kaj/ (for Americans, anyway), while the
                                > spiritual essence is usually /tSi/ (a close Anglicization of the
                                > Mandarin).

                                I've usually seen the spiritual essence spelled the Pinyin way, though (_qi_).

                                Cheers,
                                Philip
                                --
                                Philip Newton <philip.newton@...>
                              • Mark J. Reed
                                ... Not me. In martial arts comic books, movie subtitles, and the documentation from my short-lived stints in Tae Kwon Do and Hapkido classes, it was always
                                Message 15 of 28 , Aug 5, 2009
                                • 0 Attachment
                                  On Wed, Aug 5, 2009 at 10:27 AM, Philip Newton<philip.newton@...> wrote:
                                  > I've usually seen the spiritual essence spelled the Pinyin way, though (_qi_).

                                  Not me. In martial arts comic books, movie subtitles, and the
                                  documentation from my short-lived stints in Tae Kwon Do and Hapkido
                                  classes, it was always "chi".

                                  --
                                  Mark J. Reed <markjreed@...>
                                • Garth Wallace
                                  ... Chinese martial arts classes typically use qi these days AIUI. Outside of that, most people use the older, more entrenched spelling chi .
                                  Message 16 of 28 , Aug 5, 2009
                                  • 0 Attachment
                                    On Wed, Aug 5, 2009 at 7:39 AM, Mark J. Reed<markjreed@...> wrote:
                                    > On Wed, Aug 5, 2009 at 10:27 AM, Philip Newton<philip.newton@...> wrote:
                                    >> I've usually seen the spiritual essence spelled the Pinyin way, though (_qi_).
                                    >
                                    > Not me.  In martial arts comic books, movie subtitles, and the
                                    > documentation from my short-lived stints in Tae Kwon Do and Hapkido
                                    > classes, it was always "chi".

                                    Chinese martial arts classes typically use "qi" these days AIUI.
                                    Outside of that, most people use the older, more entrenched spelling
                                    "chi".
                                  • Mark J. Reed
                                    ... Well, there you go. I didn t take any Chinese martial arts, only Korean ones. :) -- Mark J. Reed
                                    Message 17 of 28 , Aug 5, 2009
                                    • 0 Attachment
                                      On Wed, Aug 5, 2009 at 4:22 PM, Garth Wallace<gwalla@...> wrote:
                                      > Chinese martial arts classes typically use "qi" these days AIUI.
                                      > Outside of that, most people use the older, more entrenched spelling
                                      > "chi".

                                      Well, there you go. I didn't take any Chinese martial arts, only
                                      Korean ones. :)

                                      --
                                      Mark J. Reed <markjreed@...>
                                    • Garth Wallace
                                      ... I only do Japanese ones, but I used to hang out on rec.martial-arts ;) (Shouldn t the Korean ones call it gi , anyway?)
                                      Message 18 of 28 , Aug 5, 2009
                                      • 0 Attachment
                                        On Wed, Aug 5, 2009 at 1:25 PM, Mark J. Reed<markjreed@...> wrote:
                                        > On Wed, Aug 5, 2009 at 4:22 PM, Garth Wallace<gwalla@...> wrote:
                                        >> Chinese martial arts classes typically use "qi" these days AIUI.
                                        >> Outside of that, most people use the older, more entrenched spelling
                                        >> "chi".
                                        >
                                        > Well, there you go.  I didn't take any Chinese martial arts, only
                                        > Korean ones. :)

                                        I only do Japanese ones, but I used to hang out on rec.martial-arts ;)

                                        (Shouldn't the Korean ones call it "gi", anyway?)
                                      • Mark J. Reed
                                        ... Well, we called the uniform a gi , so... -- Mark J. Reed
                                        Message 19 of 28 , Aug 5, 2009
                                        • 0 Attachment
                                          On Wed, Aug 5, 2009 at 4:38 PM, Garth Wallace<gwalla@...> wrote:
                                          > I only do Japanese ones, but I used to hang out on rec.martial-arts ;)
                                          >
                                          > (Shouldn't the Korean ones call it "gi", anyway?)

                                          Well, we called the uniform a "gi", so...

                                          --
                                          Mark J. Reed <markjreed@...>
                                        • Garth Wallace
                                          ... I guess your classes just used the most common terms in English. Gi for uniform is from the Japanese dōgi . The Korean term, IIRC, is dobok .
                                          Message 20 of 28 , Aug 5, 2009
                                          • 0 Attachment
                                            On Wed, Aug 5, 2009 at 1:51 PM, Mark J. Reed<markjreed@...> wrote:
                                            > On Wed, Aug 5, 2009 at 4:38 PM, Garth Wallace<gwalla@...> wrote:
                                            >> I only do Japanese ones, but I used to hang out on rec.martial-arts ;)
                                            >>
                                            >> (Shouldn't the Korean ones call it "gi", anyway?)
                                            >
                                            > Well, we called the uniform a "gi", so...

                                            I guess your classes just used the most common terms in English. "Gi"
                                            for uniform is from the Japanese "dōgi". The Korean term, IIRC, is
                                            "dobok".
                                          • Mark J. Reed
                                            Yeah, I know. In both cases the teachers were Americans, which probably has something to do with it. Come to think of it, our hapkido instructor might have
                                            Message 21 of 28 , Aug 5, 2009
                                            • 0 Attachment
                                              Yeah, I know. In both cases the teachers were Americans, which
                                              probably has something to do with it.

                                              Come to think of it, our hapkido instructor might have told us
                                              explicitly not to call it a "gi". But that's what we called it in TKD.
                                              (And in my son's karate class, but since that was actual karate, you'd
                                              expect the Japanese term.)


                                              On 8/5/09, Garth Wallace <gwalla@...> wrote:
                                              > On Wed, Aug 5, 2009 at 1:51 PM, Mark J. Reed<markjreed@...> wrote:
                                              >> On Wed, Aug 5, 2009 at 4:38 PM, Garth Wallace<gwalla@...> wrote:
                                              >>> I only do Japanese ones, but I used to hang out on rec.martial-arts ;)
                                              >>>
                                              >>> (Shouldn't the Korean ones call it "gi", anyway?)
                                              >>
                                              >> Well, we called the uniform a "gi", so...
                                              >
                                              > I guess your classes just used the most common terms in English. "Gi"
                                              > for uniform is from the Japanese "dōgi". The Korean term, IIRC, is
                                              > "dobok".
                                              >

                                              --
                                              Sent from my mobile device

                                              Mark J. Reed <markjreed@...>
                                            • Thomas Alexander
                                              Hey Mark, Most of what you re saying is indeed familiar. I m just not used to needing to be so precise. ... There are a few instances in the show where it
                                              Message 22 of 28 , Aug 5, 2009
                                              • 0 Attachment
                                                Hey Mark,

                                                Most of what you're saying is indeed familiar. I'm just not
                                                used to needing to be so precise.

                                                "Mark J. Reed" <markjreed@...> wrote:

                                                > > I could list several examples where the pronunciation
                                                > > of CH and J are routinely switched.

                                                > That happens in English, too, really;

                                                There are a few instances in the show where it could be
                                                a case of the sounds "not always being very clear in actual
                                                connected speech" (or while wearing false Pakuni teeth),
                                                but I think it's more a case of the actors mis-learning the
                                                language as they went - perhaps copying each other's
                                                mispronunciations. A more drastic example is the published
                                                word "tiri" (indeed used in the show) which eventually
                                                became "chiri".

                                                > we just adjust what we hear based on expectations.

                                                Ain't that the truth! I transcribed season 2 first and then
                                                found out that there are already transcriptions on the
                                                Internet. For season 1 my method was to do my own
                                                transcription, compare mine to the existing one, then go
                                                back to the show to confirm who is correct. In some cases
                                                the sounds I hear change with the words I'm looking at.

                                                > So "chi" and "ji" constitute a minimal pair indicating
                                                > that "ch" and "j" (/tS/ and /dZ/) are distinct phonemes.

                                                Exactly what I would have said - except that I was not
                                                familiar with the term "minimal pair."

                                                To kind of bring things back to the original topic, one
                                                thing I'm trying to guard against is the temptation to
                                                try too hard to connect all the dots. Given any number
                                                of data points, it's possible to draw a curve to connect
                                                them all. this doesn't mean that the curve is at all
                                                parsimonious, nor that it would explain any new data.
                                                I've seen this in a lot of the fan disucssions of Land
                                                of the Lost where incidental details of how things are
                                                portrayed (such as the color and state of an old fern)
                                                are used to sleuth deep details about the story. It's
                                                possible it was the easiest color fake fern available
                                                the day they made the dummy.

                                                One person concluded that Pakuni are not native to
                                                the land of the lost because their word for Sleestak
                                                doesn't have a vowel prefix like native nouns have -
                                                showing that this word is a recent borrowing. That's
                                                fine, except the same thing is true of the Pakuni word
                                                for Paku. It's seems that the pakuni are not native
                                                even to the world they came from. The more parsimonious
                                                answer is that "sleestak" and "paku" were coined by
                                                the Kroffts or the story writers without consulting
                                                Professor Fromkin.

                                                Along these lines are whether "agomba" can mean both
                                                "tyranosaur" and "dinosaur. I'm about ready to conclude
                                                that it can -- in the same way that "cat" means housecat
                                                and lion. My curve goes through all the data points and
                                                is still somewhat parsimonious.

                                                Pakuni has two words glossed as "to do" in fan dictionaries.
                                                I'm starting to think that one of them ("shi" - also
                                                mistakenly glossed as "it" in some fan dictionaries)
                                                really means "to attempt to do" or "to undertake a task."
                                                It's tempting to say that this word "shi" has a meaning
                                                close to, but not quite the same as "to try", in that
                                                it suggests a greater surity of success. By this time
                                                I wonder if I'm just drawing a wigly line through all
                                                my data points.

                                                > Phonetic effects like this are not usually governed by
                                                > morphosyntactic considerations - <...>
                                                > But there are exceptions, so it could go either way;
                                                > I'd base the analysis on how things actually sound
                                                > on the show.

                                                There are no examples in the show. That's the problem.
                                                I suppose one could claim that the ending <sa> is always
                                                pronounced /sa/ (or is it [sa]) when the word is an adjective.

                                                > So if <akingo> is [@kiNgo],

                                                Well, that's the first "if." The actors do pronounce it [akiNgo]
                                                (won't swear whether it's [a] or [@]) but I'm not convinced
                                                it's corect. It's all I have to go on, though.


                                                > the question is whether <kingsa> would be
                                                > [kiNsa], [kiNza], [kiNgsa], or [kiNgza].

                                                Right ... or [kiNksa], I suppose. My speculation was that
                                                the root word is /kiN/ (grumpy) and that the word <akingo>
                                                (the grumpy one) is a derrivation meant to be pronounced
                                                /akiNo/ and that the /g/ was added by the English speaking
                                                actors who saw <akingo> in the script.

                                                > It's likely that if [n] triggers voicing of /s/ to [z],
                                                > then so would [N] (and [m] for that matter), but not necessarily.

                                                If we underline "if", this is my point exactly.


                                                > Both of those are English pronunciations of the written word <chi>;

                                                It's just amazing to me that both apparently made it into the show.

                                                Amike salutas,
                                                Thomas/Tomaso ALEXANDER.
                                                www.NightinGael.Net
                                                ---Anything below this line is not from Thomas ---
                                              Your message has been successfully submitted and would be delivered to recipients shortly.