Re: Camion Code - way off topic

From: peter_constable@sil.org
Date: Fri Aug 20 1999 - 03:00:30 EDT


>There have been many attempts and in particular Bell's Visible
       Writing and Shavian refer to similar principles as CC, BUT
       they (like the alphabet, Pitman's shorthand, and the IPA) use
       'kabbalistic', inherently 'meaningless' symbols (the top tips
       to the left, or right...) to distingush between sounds, whereas
       CC is 'decipherable' - each part of its visual code refers to
       the composite of how the sound is formed: vowel or consonant,
       ('schematic') place of articulation, voiced or not, plosive or
       fricative, or 'stretched' or 'rounded' (if vowel). It's true
       that these terms and their 'meanings' must be learned (more
       difficult if one has already learned that the 'sign': 'A', or
       'a' or'@' (cursive...) represents the vowel sound in the word
       'at', or 'law, or 'date' or 'above'...).

       Rather than "decipherable", I think a better term might be
       "iconic" for what you mean. There are lots of people who have
       little trouble deciphering IPA. Even so, I'm not sure I agree
       that CC is as completely "decipherable" (iconic) as you claim
       it to be. For example, the distinction between /l/ and /r/, as
       with the distinction between /m/ and /n/, is indicated solely
       by the weight of the line. How is that not anything but
       arbitrary? For that matter, it could be argued that the use of
       weight to distinguish voiced/voiceless pairs of plosives is
       equally artibrary. (Generally, the voiceless plosives are
       considering stronger or more forceful than the voiced plosives,
       but this would have suggested - at least to some - that the the
       voiceless plosives have the heavier line.)

       By the way, how would any other nasal consonant (e.g. velar,
       palatal) be represented? The answer can only be arbitrary,
       since there is no systematic mechanism that provides the
       answer.

       This last point raises up an interesting issue that is a
       problem for Iconic systems of representation. Any
       representation - even an iconic one - is, of necessity, an
       abstraction. An iconic representation system, however,
       minimises that abstraction sufficiently so that the
       interpretation of the formal representation is apparent. (At
       least this is true of any *successful* iconic system of
       representation.) It remains an abstraction, nontheless: the
       denotation is not the same thing as the denotee. Now, for
       iconic systems, the degree of abstraction becomes an important
       factor in the determining the range of possible
       interpretations: since the interpretation is dependent *solely*
       upon the inherent qualities of the formal representation, the
       amount of information resulting from the interpretation is less
       than or equal to the amount of information that can be embodied
       by those inherent qualities. When there is a need to provide a
       high level of detail in representations, then the inherent
       qualities of that iconic system must be adequate to meet those
       demands. The fact that the representation is an abstraction,
       however, necessarily imposes limitations.

       From what I've seen, CC falls far short of phonetic
       representation. (This probably has a lot to do with why I had
       never heard of it before.) The palatal and velar nasals are
       indicative of this. Other examples abound:

       front rounded vowels
       back unrounded vowels
       aspirated vs. unaspirated plosives
       clicks
       implosives
       lateral articulation (e.g. the "tl" in Nahuatl (Aztec)
       languages)
       laryngealised vowels
       breathy vowel
       tone

       Perhaps, though, I just missed the following warning

       (Warning: Do not attempt to represent languages of SE Asia
       using this system.)

       from the Linguist General, and several more like it. (Opps -
       now I'm getting cynical. Apologies offered.)

       There are reasons why IPA works and is accepted. First, it is
       infinitely extensible - whenever a need is identified to
       represent some detail, a new symbol can always be invented and
       conventionalised - and this is so precisely because the system
       is *not* iconic: semantics are not dependent in any way upon
       any inherent qualities in the formal language. Secondly, it has
       been possible to establish a workable balance between the size
       of the formal language that must be mastered in order to use it
       effectively and the level of detail in representation that is
       required.

       CC could be extended to support a greater level of detail to
       cover the things I listed above and more, but I don't see how
       that can be done while keeping it all iconic. There is no
       obvious way using the metaphor adopted to indicate inward air
       flow, details of the configuration of vocal chords involved in
       breathy, creaky or laryngealised vowels, glotallic air
       mechanism involved in clicks, position of the tongue body in
       velarised or palatalised plosives (as in Arabic or Russian),
       etc. It may be possible to add to the metaphor to provide ways
       to represent these things iconically, but then a problem would
       arise as to whether the formal mechanisms become so hard to
       master as to become impractical.

       (In considering the need to balance the size of the formal
       apparatus with the amount of detail that is required from the
       interpretation, I'm reminded of SignWriting, a system for
       writing signed languages that was adapted from dance notation.
       The representations involve a fairly high level of detail,
       which has made me curious to know whether it will succeed as a
       practical orthography for any Sign language, or whether the
       level of detail is enough of a hindrence to fluent writing as
       to make it fail in this function. Not that I have any opinion
       as to whether I think it ought to succeed or actually will
       succeed. From a purely academic perspective, it's a case I find
       interesting because it's at the edge of my knowledge regarding
       orthogrphies and socio- and psychological aspects of literacy.)

       Beyond that, I'd have to ask, why would I want to bother? Even
       the iconic aspects of the system are don't seem to me to be all
       that intuitive. (Marie, herself, pointed out, "It's true that
       these terms and their 'meanings' must be learned... ".) Even
       having taken a course in articulatory phonetics, which
       apparently imporant in the iconicity of this system, symbols
       derived from Roman script have, at least for me, far more
       mneumonic value for phonetic representation.

       There is, in addition to all the, the critique that has been
       mentioned already that a phonetic representation of "English"
       is, of necessity, a representation of some specific dialect of
       English. For all of the problems of English spelling, it has
       one very great benefit, which is that a single representation
       can be read by people from (almost) all over the globe and be
       interpreted very nearly uniformly. No phonetic representation
       of English could do that.

       This has been way off topic. To bring it back on topic, I am
       not aware of any evidence indicating that CC has enough user
       acceptance (I currently know of a user community consisting of
       one) to merit incorporation in any plane of Unicode other than
       plane 19. But I think I'm now repeating what's already been
       said.

       I feel a little bad having said so much to deride what a new
       member of the list holds dear, so let me close by saying,
       Marie, don't let my critique bother you: you obviously find
       this interesting, so keep enjoying it. I know that if I
       admitted to my fascination with pant pocket lint, I'd be
       laughed off this list.

       Peter



This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:51 EDT