Re: Plausibility Check

In article <1152494106.638648.24220@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>,
DJensen <i_m0nk@xxxxxxxx> writes
Some background:
I have been working on a posthuman science fiction novel for a while
now, and I've reached the point where I have to start nailing down
certain details before they hit the page. Such as the language and
writing system of the people and civilization I'm writing about. My
attention is focussed mostly on the writing system (as it will appear
in the book - I'm a sucker for appendices in fiction, and a graphic
designer, so this is all an excuse to create a set of glyphs) and
therefore the phonology, and less so on the spoken language, grammar
and vocabulary. I've spent the last two weeks mouse-deep in
Wikipedia, snatching up the barest details of phonology to combine with
inspiration drawn from Omniglot.com, so I'd thought I'd get people
who know what they're talking about to tell me how plausible it all
is, and answer some questions. I've probably swapped "phonology"
for "phonetic" or vice versa somewhere, and I'm sure I'm
covering territory and ideas that are old hat to most, so bear with me.


Origins (in theory):
I'm basing this language/phonology/writing system on two very
unscientific theories based on my own observation:

First, that asking people to learn a whole new vocabulary, grammar,
phonology, and writing system (in the fine tradition of previous IALs)
is a serious barrier to adoption, but there has been success in getting
people to adopt a writing system when it lets them write what they're
already saying to each other (Hangul, Inuktitut, Katakana).

Second, that armed with only a handful of rules for other languages,
you can pronounce (more or less) words written in languages that share
your writing system even if you don't know what they mean (English,
Spanish, French, Italian, etc). Conversely, languages that share the
same character set but not pronunciation/phonology can retain the
meaning of the words the symbols represent (Mandarin, Cantonese,
Kanji/Hancha/Honzi, and Arabic numerals).

It follows then, at least in my mind, that if an International
Auxiliary Alphabet were introduced (by the right people) and promoted
(to the right people), languages would begin to converge as they
started sharing more and more vocabulary and phrases. By 'begin' I
mean over the course of centuries. Several generations later,
metalanguages might emerge, and generations after that, a "single"
massive metalanguage would dominate. (Obviously all of this relies on
institutions and social movements that don't have such global
influence today [or just don't exist in any form] acting with
considerable conscious foresight, so handwave that element until you
accept it as future-historical fictional fact.)

I'm aware that converting certain languages to a phonetic alphabet
wouldn't work very smoothly, while other languages are written more
or less phonetically already (Spanish, Korean) or have writing systems
that offer little phonetic information. So I propose that initially it
would be introduced with numerous compromises and exceptions between
phonetics and existing spelling. Here I'm thinking mostly of English:
people all around the world, native speakers and as-a-second-language
speakers, can pronounce the same word very differently. However,
regardless of the differences in stress patterns, rhoticity, etc,
everyone agrees how to _spell_ the word. (With the exception of a few
ou/o, s/z, and or/re/er here and there.) The problem of
heteronyms-are-now-homonyms could be resolved with certain compromises
that would keep words visually distinct but similar looking. Over the
generations reforms could weed out the non-phonetic elements of the
writing system to match usage, and homo/heteronyms could be reduced
with synonyms originating in other languages. Some non-phonetic
elements would likely remain.

The language I'm writing about in the novel itself is the
generations-removed post-reforms descendent of this undertaking. It's
not meant to be a logically constructed language. For sake of ease,
I'll refer to it as Rosetta.

Origins (in practice):
I've taken what I've grasped of the phonology (according to various
IPA charts) of some of today's most widely spoken and influential
languages, figured out the significant overlaps, and distilled the
phonology of Rosetta from that. The seed languages were English,
Spanish, French, Russian, Mandarin, Cantonese, Hindi, Arabic, Japanese,
and Bahasa Indonesian.

I assumed that given their global dominance, Rosetta would be most
influenced by them on all fronts. Even after centuries and reforms,
their phonological footprint would remain. The sounds of the seed
languages are no longer fully represented in Rosetta, because I've
trimmed away the sounds that were the least common.

Phonology:
Consonants (36): m n n^ N p b t d t. c k g q ? f v T D s z s. z. x h h?
r r<trl> r" l j w<vls> w dz ts ts. dz.
Vowels (16): i y u I U e @<umd> @ o E O" O & &. a a.
(based on sci.lang's FAQ's guideline, as best as I could decipher)

Which sounds are most likely to collapse into others, near enough that
they might be represented by the same glyph? Particularly vowels. I
have my own guesses, but they're only guesses. I plan on having some
featural elements in the letters, for certain sounds that are likely to
remain distinct enough to warrant it. I have no qualms about a 30+ or
40+ character alphabet, the more the better.

Diphthongs and triphthongs - I'd rather not have any, would this be
doable with some well-placed and carefully considered
diacritics/features?

Given the above seed languages (but remembering that they aren't the
only sources from which Rosetta will inherit), any speculation as to
the characteristics of Rosetta's grammar and vocabulary?

Any other input is welcome.

Much thanks to any and all who take the time to get through all that.


-
DJensen


I am not a linguist, just someone fascinated by how language developed
and spread. One thought occurs to me: how widely (geographically)
spread are the groups of people who are to speak your fictional
language, and how effective is the routine communication between them?
Reason I ask is that when people are separated by great distances and
have little routine communication, then a language that starts out as
common will be altered in each sub-population group. This appears to
have happened with English, where some commentators (possibly of the PC
persuasion) are now saying that the many dialects and pidgins that have
developed around the world should be regarded as valid variants of
English in their own right, and not corrupted forms of an original, and
still nominally 'correct', English.

So, your fictional scenario might have to cope with this effect.

Cheers,

Dave

--
Dave Eadsforth
.

Relevant Pages

  • Plausibility Check
    ... writing system of the people and civilization I'm writing about. ... therefore the phonology, and less so on the spoken language, grammar ... Second, that armed with only a handful of rules for other languages, ... I'll refer to it as Rosetta. ...
    (sci.lang)
  • Re: lsort -dictionary
    ... >> certainly not one that spans languages. ... >It's been well known for a long time that string collation is ... writing system to writing system, and that some other principles vary ... variation is not a function of locale. ...
    (comp.lang.tcl)
  • Re: Plausibility Check
    ... with synonyms originating in other languages. ... parents and others in their early childhood environment. ... The first generation wouldn't be taught pronunciation, ... they'd be taught a writing system that represents what they've learned ...
    (sci.lang)
  • Re: Plausibility Check
    ... with synonyms originating in other languages. ... People don't learn the pronunciation of their native language from ... The first generation wouldn't be taught pronunciation, ... they'd be taught a writing system that represents what they've learned ...
    (sci.lang)
  • Re: Historical Linguistics to the Max
    ... Nathan Sanders wrote: ... >> no graphemes for tones in any of these languages, ... > A writing system does not need to encode every possible phonemic ... > English spelling doesn't even completely distinguish between the much ...
    (sci.lang)
Quantcast