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

Re: Not enough difference?

Expand Messages
  • George Corley
    ... Damn. That s unfortunate. I really hate to add more changes after having coined words. I guess I could pin down what I have as an earlier stage in the
    Message 1 of 10 , Mar 30, 2013
    • 0 Attachment
      On Sat, Mar 30, 2013 at 10:28 PM, Alex Fink <000024@...> wrote:

      > On Sat, 30 Mar 2013 15:57:36 -0500, George Corley <gacorley@...>
      > wrote:
      >
      > >> What sound changes have you got so far?
      >
      > [*snip*]
      >
      > Yeah, that feels scant to me for a millennium, though I guess the vowel
      > changes could bulk it up. Maybe you could add more "small" rules, that is
      > rules of relatively specific conditioning. Such small rules could either
      > result in (a) an actual change in spelling, where the "big" rules don't, or
      > (b) no change in spelling and then an opportunity for spelling
      > pronunciation to happen. Or maybe some doublets could form via some kind
      > of semi-classical reborrowing procedure, like the semi-learned reborrowings
      > like _regla_ in Spanish etc.
      >

      Damn. That's unfortunate. I really hate to add more changes after having
      coined words. I guess I could pin down what I have as an earlier stage in
      the language and then develop further. Now I need to do even _more_
      freaking derivations. (I'm deriving words by hand, because I can't find
      single damn sound change applier that can handle the kinds of sound changes
      I want, and the one I had used required too many ridiculous workaround
      rules.)


      > Though I should've asked what the initial inventory and phonotactics are,
      > as well. For instance, have you got affricates? That would make the
      > strident voicing strange.
      > Actually, it's odd that only stops take on rounding, whatever your
      > inventory is, probably.
      >

      I have no idea if this table will survive the list. If it doesn't, I'll go
      through and type the damn thing out:


      *

      Labial

      Alveolar

      Retroflex

      Palatal

      Velar

      Uvular

      Pharyngeal

      Glottal

      Voiceless Plosive

      p

      t


      k

      q

      ʔ

      Voiced Plosive

      b

      d


      g

      ɢ (ǵ)


      Nasal

      m

      n






      Trill

      r






      Fricative

      f

      s




      ħ (hh)

      h

      Approximant

      ʋ (w)

      ɻ

      j (y)




      Lateral

      l






      *

      Phonotactics are essentially (C)(C)V(C)(C) with a few extra stipulations
      (glottals and the pharyngeal do not appear in clusters, second consonant of
      the onset must be a sonorant -- as must the penultimate consonant in a
      complex coda). Initial vowels are /i a u/, expanding to a much larger nine
      vowel system through [front] and [high] harmony (once vowel harmony kicks
      in, /a/ becomes one of those odd transparent vowels.



      > I also suppose it's likely to matter for the speakers' subsequent
      > understanding of these changes whether they're allophonic or not. If all
      > [z] are <s> between vowels, fine. If there's /z/ which is now spelled <z>
      > everywhere as well as <s> intervocalically... well, that's probably still
      > survivable, but it'll be noticed.


      You'll see that I have no voiced fricatives


      > >So what type of writing system does Pahran have, and what's its history?
      > >> What other dialects are about?
      > >
      > >Pahran has an alphabet that has been used since before the old three-vowel
      > >system split, which (once I figure out how the modern form of that
      > alphabet
      > >works) will surely lead to some absolute hilarity when it gets applied to
      > >representing nine vowels plus length and nasalization. I don't actually
      > >have that writing system in front of me -- I do think I had originally
      > >written both rhotics identically, but i may change that.
      >
      > Might even be non-hilarious! If the length and nasalisation are
      > consistantly a following <r> and <n> (or nasal C), respectively, that's not
      > too bad. The qualities, I reserve judgment for now...


      The biggest issue is how a writing system meant for /i a u/ is going to
      contort itself to represent /i y e ø a ɤ o ɯ u/. I'm planning on lots of
      diacritics. Nasalization and lengthening is not going to be a huge issue,
      as you say -- just use some silent letters from history -- although there
      is some potential for confusion due to some coda nasals surviving (and not
      causing nasalization). The bigger challenge will probably be the
      romanization -- it doesn't really matter so much if your "native script" is
      balls out wacky, since that's just natural.


      > >I haven't done much dialect work, but a few ideas in my head are:
      > >-- Many dialects have merged both rhotics in favor of the trill, but the
      > >prestige dialect attempts to preserve the distinction,
      > >often hyper-correcting in the process.
      >
      > Do they contrast?


      There's a historical contrast between the retroflex approximant and
      alveolar trill. You can see them starting to merge in my tables. My idea
      is that merger goes further in some dialects, but others preserve the
      distinction. Then, literate people observe the distinction in writing,
      but occasionally end up hypercorrecting historical trills to retroflexes.


      > >-- There may be some variation in the uvulars -- the voiced uvular stop
      > >tends to be rare in this variant, but it isn't always.
      > >-- There may be small tweaks to certain rules. For instance, in the
      > >variant I'm working on, the gemination rule actually applies to nasals as
      > >well (essentially any combination YX where X is an oral stop and Y is a
      > >nasal or oral stop, it becomes XX), but that may not be the case for all
      > >dialects.
      >
      > These seem like the sort of things which would have a chance to be
      > reintroduced, then: extra instances of [G\] where it's spelled (especially
      > if that doesn't violate otherwise exceptionless phonotactics), extra
      > [nt]-type clusters where they're spelled.


      True. I will add that this language is for a fantasy world that as at a
      stage of social development where literacy isn't terribly common. As such,
      it would probably make sense for spellings to be somewhat in flux, with no
      really effective way to standardize. This society may or may not have an
      early printing press, but if they do it is a very new and very limited
      phenomenon.
    • MorphemeAddict
      I had no idea ... Tongue is also the same word. stevo
      Message 2 of 10 , Mar 31, 2013
      • 0 Attachment
        I had no idea


        On Sat, Mar 30, 2013 at 2:28 PM, H. S. Teoh <hsteoh@...> wrote:

        > On Sat, Mar 30, 2013 at 12:56:17PM -0500, George Corley wrote:
        > > Something that has been bugging me while working on Pahran -- I feel
        > > like a lot of roots are coming through my sound changes untouched. Of
        > > 68 words I have created so far (yes, I am slow with lexical
        > > development)
        >
        > Aren't we all? :)
        >
        >
        > > 15 are entirely unchanged, and many others have only developed in ways
        > > that are clearly just allophonic variation.
        > >
        > > Keep in mind that Pahran's sound changes are supposed to span
        > > 1000-2000 years.
        >
        > Maybe you want to break that up into periods of, say, 250 years or so
        > each, and at each stage, apply yet another stage of sound changes?
        >
        >
        > > I haven't really been thinking about how syntax and morphology are
        > > changing, either. Being a high-class, somewhat literary dialect, the
        > > variant of Pahran I'm working on may well make sense to be a little
        > > conservative, but it seems a bit TOO conservative at the moment.
        >
        > It seems that what happens with literary dialects much of the time is
        > that the vernacular has moved on in radical ways, but the literary
        > dialect (tries to) retain its original form. However, it's not truly
        > independent from the vernacular either; pronunciation for one thing will
        > change no matter what. What people try to do is to try to compensate for
        > it by developing a "literary pronunciation system" which is different
        > from the vernacular, but of course, that also tends to erode over time,
        > and sometimes people over-compensate, introducing supposedly
        > "conservative" forms that are actually not present in the original
        > literary language. The vernacular will influence this in both ways:
        > sometimes a colloquialism is imported into the literary language (though
        > this isn't too common), sometimes the literary language acquires new
        > anachronisms from over-compensation away from the vernacular.
        >
        > As for syntax/morphology changes, if the language is actually being used
        > actively (as opposed to being, say, a liturgical language), then it will
        > definitely experience change. The most common phrases / segments will
        > tend to shorten, contract, merge, etc.: people don't like pronouncing
        > every last syllable if everyone already knows what the entire phrase is
        > going to be anyway. Once enough has been said that determines the rest,
        > you can pretty much bet on the rest just melting into a quick slur,
        > which over time will calcify into actual word endings, for example.
        > Sometimes the same word can split into a full form and a slurred form
        > due to context-dependent simplification: e.g., the -ся passive ending of
        > Russian verbs developed from себя (oneself) in the ancestral language,
        > but себя is still used in its original sense (and without contraction to
        > *ся) today. So you can see how the original construction of verb + себя
        > came to be regarded as a unit, and over time contracted into verb + ся
        > (and even just -сь /s_j/ in some contexts), but outside of this context,
        > себя didn't erode at all.
        >
        > Another Russian example: спасибо ("thanks") is a contraction of спаси
        > Бог ("God save"), yet both standalone words still exist unmodified
        > today. (Well, "unmodified" is not quite accurate, perhaps, the
        > pronunciation *has* shifted over time, but nowhere near as much as the
        > contracted form спасибо did. IOW, the same word/phrase can undergo
        > different rates of "erosion", depending on the context! More generally,
        > different words may undergo different rates of change over time, such
        > that after a lengthy period of time, like your 1000+ years, some words
        > may be very close to the original but others have mutated beyond all
        > recognition. I'm sure you can think of all sorts of cases where this
        > could happen in your lang, creating new words or new morphemes from the
        > same original roots, and all sorts of fun stuff like that.)
        >
        > (Marginally-related note: my favorite example of mutation beyond all
        > recognition is the fact that the Russian "язык" /yazi\"k/ is cognate
        > with the English "language" /lE"NguIdZ/ (!). Now imagine if something
        > like this happened in the *same* language due to different changes being
        > applied in different contexts. Wouldn't that be cool?!?!)
        >

        "Tongue" is also the same word.

        stevo

        >
        >
        > T
        >
        > --
        > What do you get if you drop a piano down a mineshaft? A flat minor.
        >
      • Alex Fink
        ... That s what seems easiest to me; this could make a fine half-millennium or a bit more. At any rate, if down the line you do the sort of morphological
        Message 3 of 10 , Mar 31, 2013
        • 0 Attachment
          On Sat, 30 Mar 2013 22:59:07 -0500, George Corley <gacorley@...> wrote:

          >On Sat, Mar 30, 2013 at 10:28 PM, Alex Fink <000024@...> wrote:
          >
          >> On Sat, 30 Mar 2013 15:57:36 -0500, George Corley <gacorley@...>
          >> wrote:
          >>
          >> >> What sound changes have you got so far?
          >>
          >> [*snip*]
          >>
          >> Yeah, that feels scant to me for a millennium, though I guess the vowel
          >> changes could bulk it up.
          >
          >Damn. That's unfortunate. I really hate to add more changes after having
          >coined words. I guess I could pin down what I have as an earlier stage in
          >the language and then develop further.

          That's what seems easiest to me; this could make a fine half-millennium or a bit more. At any rate, if down the line you do the sort of morphological change I was talking about, with affixes becoming bleached and accreting on stems and so on, you'll need to change some words on that account as well.

          >Now I need to do even _more_
          >freaking derivations. (I'm deriving words by hand, because I can't find
          >single damn sound change applier that can handle the kinds of sound changes
          >I want, and the one I had used required too many ridiculous workaround
          >rules.)

          Eh, by hand is how I do it. I dunno how unusual this is, but myself I find holding a set of sound changes in mind and carrying them out easy enough, certainly less work than creating the lexicon to run through them in the first place, or deciding what the semantic or analogical changes will be. Hardly worth automating; probably more trouble to debug it than the work it saves.

          >Initial vowels are /i a u/, expanding to a much larger nine
          >vowel system through [front] and [high] harmony (once vowel harmony kicks
          >in, /a/ becomes one of those odd transparent vowels.

          Funky. Normally, in the frontness harmony systems I'm familiar with, front is a dominant category, but I guess you've got both [i...u] > [M] and [u...i] > [y] sort of things going at once?

          >True. I will add that this language is for a fantasy world that as at a
          >stage of social development where literacy isn't terribly common. As such,
          >it would probably make sense for spellings to be somewhat in flux, with no
          >really effective way to standardize. This society may or may not have an
          >early printing press, but if they do it is a very new and very limited
          >phenomenon.

          Ah, fair enough. Maybe an important question is then whether texts written in old Pahran continuously remained canon to the small literate segment and the development of the standard was moored to it, with imitation of the exemplars serving the sort of function the printing press might, or whether they were lost contact with (and possibly later rediscovered).

          Alex
        • George Corley
          ... Yes, that s how I got to it. My plan is to have underspecified vowels that specify [round], plus another set that is explicitly [-high] to account for
          Message 4 of 10 , Mar 31, 2013
          • 0 Attachment
            On Sun, Mar 31, 2013 at 3:43 PM, Alex Fink <000024@...> wrote:

            > On Sat, 30 Mar 2013 22:59:07 -0500, George Corley <gacorley@...>
            > wrote:
            > >Initial vowels are /i a u/, expanding to a much larger nine
            > >vowel system through [front] and [high] harmony (once vowel harmony kicks
            > >in, /a/ becomes one of those odd transparent vowels.
            >
            > Funky. Normally, in the frontness harmony systems I'm familiar with,
            > front is a dominant category, but I guess you've got both [i...u] > [M] and
            > [u...i] > [y] sort of things going at once?


            Yes, that's how I got to it. My plan is to have underspecified vowels that
            specify [round], plus another set that is explicitly [-high] to account for
            cases that would otherwise break height harmony. I wasn't aware that front
            vowels tended to be dominant -- I was thinking of Turkish as a model for
            front-back harmony.


            > >True. I will add that this language is for a fantasy world that as at a
            > >stage of social development where literacy isn't terribly common. As
            > such,
            > >it would probably make sense for spellings to be somewhat in flux, with no
            > >really effective way to standardize. This society may or may not have an
            > >early printing press, but if they do it is a very new and very limited
            > >phenomenon.
            >
            > Ah, fair enough. Maybe an important question is then whether texts
            > written in old Pahran continuously remained canon to the small literate
            > segment and the development of the standard was moored to it, with
            > imitation of the exemplars serving the sort of function the printing press
            > might, or whether they were lost contact with (and possibly later
            > rediscovered).


            In light of all this, I may need to have my Pahran alphabet developed
            somewhat later -- perhaps even after vowel harmony -- or perhaps move vowel
            harmony changes to the next stage of the language (and flesh them out
            more), so it's hard to say whether this particular stage of the language
            will have all that much written documentation. It will certainly be there
            in later stages, though.
          Your message has been successfully submitted and would be delivered to recipients shortly.