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

192060Re: the gripping language on a string instrument?

Expand Messages
  • Leland Kusmer
    Nov 3, 2012
      > --- On Wed, 10/31/12, Sam Stutter <samjjs89@...> wrote:
      > > From a (not very good) violist's point of view:

      Another violist weighing in, here. (Viola power!)

      > I would say: not a bowed instrument (unfortunately).

      Agreed. Yes, it's true that a flatter bridge will allow chords, but not
      generally all four strings at once. Uke is probably a good bet; mandolin is
      also a possibility. (Since it's strings are tuned the same as a fiddle's,
      it's a lot more intuitive for me, so when I give examples below they're all
      assuming mando.)

      Here's the simplest possibility that I can think of:

      - Map chord vs. arpeggiation to knuckledness, as suggested above
      - Map strings to fingers
      - Map thumb position to fingering: We have five possible fingers (0-4 where
      0 is open and 4 is pinky). Put the finger corresponding to the thumb
      position on the highest string; put the next highest finger on the next
      highest string; repeat, appending open strings if necessary.

      Note that this doesn't say exactly where to put the fingers on the strings;
      for that purpose, pick a key area (if you're on mando then G Major / E
      Minor is probably easiest) and always use appropriate fingers in that key.

      This doesn't even come close to achieving musicality, though I expect that
      it would actually be fairly easy to interpret. There's going to be a strong
      tension between preserving the structure of the gripping language and
      achieving musicality in any mapping.

      I'd submit that some degree of musicality is worth achieving, though, not
      just for aesthetic reasons: Small differences in the neighborhood of
      familiar musical structures (triads, etc.) will be much more perceptible
      than the same differences in the nowhere-land of unfamiliar pitch
      collections. For instance, untrainted musicians with passive familiarity
      with Western-style music will easily hear the difference between a major
      and minor triad (which differ only by a half-step in one pitch), but would
      have difficulty distinguishing between, say, C E D and C F D (with the D in
      a higher octave in both cases). So creating a mapping that keeps gripping
      language syllables close to familiar pitch collections will greatly aid in

      One of the other problems you're going to run into is that performers will
      need to reset all their fingers between each chord. The requirement that
      successive syllables share fingers doesn't help in this domain at all, when
      those fingers are mapped to strings.

      Another possibility, with slightly more musicality built in:

      - Each thumb position is assigned to triad built on a successive scale
      degree. Pick a key (again, for mandolin G Major probably makes the most
      sense); then thumb position 1 is a G Major triad; 2 is A Minor; 3 is B
      Minor; 4 is C Major; 5 is D Major.
      - Map fingerings to strings. On each string, play the lowest available
      chord tone in the appropriate triad (unless the open string is in the
      triad, in which case play the open string; or unless this results in a
      fingered fifth across two adjacent strings, in which case pick the next
      best option).
      - Again, map knuckledness to arpeggiation / chordal articulation.

      The problem here is that you start to rely on absolute pitch. Someone with
      some training, given a starting pitch (i.e. just play an open G string
      before starting your utterance) could work out the rest relatively. See
      above re: adding a drone, however.

      This doesn't necessarily produce things with musical interest, but at least
      it will all be "pretty" sounds in the sense of (fragments of) nice
      consonant triads drawn from the same key.

      A counter-proposal: Rather than mapping gripping language fingers to
      instrumental strings, map fingers to fingers. This allows a simple mapping
      of each syllable to two pitches (which may not be distinct) – the starting
      and ending point of the block of pressed fingers. The trick, then, is how
      to map thumb positions. If we take individual strings and adjacent pairs of
      strings, we end up with 7 positions, which accounts for all the basic thumb
      positions plus the two rubs.

      Then, do the following:
      - To articulate a pair of fingers on one string: If arpeggiated, play all
      intervening diatonic notes (within a chosen key); otherwise, play just the
      - To articulate a pair of fingers on two strings: Arpeggiate or play as a
      chord, as expected.

      It's not a great fit, but it would be much easier for performers, probably
      rather easier on listeners, potentially more musical (or easier to make
      musical in performance), and (best of all) playable on bowed instruments
      like viola, allowing for sustained sound through an utterance.

      If I have time, maybe I'll mock up some samples later this week.


      > I would say the listener would need quite some
      > > grounding as a professional musician to understand each
      > > "syllable", with perhaps access to a piano. Data transfer
      > > will be slow (for humans at least).

      > Quite probably true. But as with any other language, practice makes
      > perfect!
      > Padraic
      > > Sam Stutter
      > > samjjs89@...
      > > "No e na'l cu barri"
      > >
      > > On 31 Oct 2012, at 23:54, David Peterson <dedalvs@...>
      > > wrote:
      > >
      > > > Ah, this reminds me of when I was trying to come up
      > > with a guitar language with J Goard... We didn't find an
      > > ideal way to do it (too many combinations for phonemes; too
      > > few for a minimalist language). If you think you've got your
      > > tokens, though...
      > > >
      > > > On Oct 31, 2012, at 4:21 PM, Alex Fink <000024@...>
      > > wrote:
      > > >
      > > >> Now, it seems natural to do something like this.
      > > >> * Use a four-stringed instrument (which instrument
      > > is best?), and let
      > > >> the _sets of fingers_ correspond to the sets of
      > > strings being sounded
      > > >> at any one time.
      > > >
      > > > 'Ukulele. (Note that the common banjo has 5 strings,
      > > but not on all frets.)
      > > >
      > > >> * _Thumb position_ can be which fingering is
      > > selected on the strings
      > > >> being used. Ideally one would set this up in
      > > such a way that all the
      > > >> multi-finger articulations are good chords, but the
      > > linearity of the
      > > >> five thumb positions is still visible.
      > > >
      > > > Shouldn't be too hard.
      > > >
      > > >> * _Knuckledness_ can, er, perhaps be something like
      > > one of two modes
      > > >> of strumming / plucking / bowing the strings in
      > > question? I'd like
      > > >> them to be audibly distinct, though. Or maybe
      > > something else entirely
      > > >> -- maybe major vs. minor chords? Except that
      > > then it's hard to extend
      > > >> the distinction to single fingers, isn't it?
      > > >
      > > > Arpeggio vs. strummed.
      > > >
      > > > David Peterson
      > > > LCS President
      > > > president@...
      > > > www.conlang.org
      > >
    • Show all 11 messages in this topic