On Saturday, April 10, 2004, at 08:50 PM, Philippe Caquant wrote:

> I never said that the study of speech is irrelevant, I
> said it's peripheral when building a language.

This is simply not true - and there are actual examples to show this.


> If you use "purely graphical languages", that you
> don't use "phonetics", but writing, or drawing.

Yes, but graphics are merely the outward form of the language in the same
way that phonetics are.

> "Phone" (phi, omega, nu, eta) means "voice" in Greek

..among other things - it actually has quite a wide range of meaning.

> Only if you think in a metaphorical way you can call
> writing, "phonetics". But with metaphors, everything
> is possible, of course. I was thinking of Chinese
> ideograms that the Chinese pronounce in their own
> way(s), and the Japanese, a completely different way,
> and yet it's the same concept. If a Chinese hears the
> word pronounced by a Japanese, he will not understand
> it, but if the Japanese writes it to him, then he will
> (I never experienced this, but I was told it is so).

Yes, but this works basically for individual morphograms. The syntax of
Japanese & Chinese are rather different. But 'ideograms' are only one part
of Chinese writing. The Chinese linguist Yuen Ren Chao says that in
traditional classification of the Chinese writing system there are six
classes of characters. One called _chuanchu_ is obscure in nature and
there are, he says, few established cases of characters clearly falling
into that class. The other five classes are:
1. Pictographs: characters that originated from pictures of objects.
2. Ideographs: diagrammatic indications of ideas, e.g. the symbols for 1,
2 & 3 (one, two 7 three horizontal lines respectively), symbols for up &
down (which I can't show in ASCII but one is the inverted form of the
3. Compound ideographs: characters in which the meaning of the whole is a
combination of the meanings of its parts - stick examples are
'military' <-- 'to stop' + 'arms' (cf. the idea of "war to end all wars"),
'honest' <-- 'person' + 'word'; 'bright' <-- 'sun' + 'moon'.
4. Loan characters: this does not mean a character borrowed from a foreign
language. I means one used for its *phonetic* value, to represent a
morpheme which is homophonous with it. E.g. the character for 'lai2',
originally a pictograph for '(a type of) grain' came to be used (and is
used) for the homophonous morpheme meaning 'come'.
5. Phonetic compounds: "by far the most common type of Chinese characters"
, according to Y.R. Chao.

To ignore the phonetic element in the Chinese script is unwarranted. Nor
is it surprising; humans were communication by sound (i.e. speeking) for
millennia before anyone used any graphical way to represent those spoken

> I see semantics and syntax a little alike a computer
> application. When you have such a project, first you
> writing and phonetics, and defining the actual words.
> It's a little like writing the final programs. This
> comes nearly at the end.

This is clearly the way you do things but you should not assume that it's
the way everyone else does things, still less should you define it as
_the_ way of creating languages. All it does, with respect, is to show
your particular interest in language creation.

We are all different & those of us that have been on the list for a few
years realize that different people have very different reasons for
conlanging and there is and cannot be just one way of doing things. What
is more important or less important depends, surely, on WHY a person is
creating a language.

To some, semantics is of low importance. Obviously for Tom Breton in
devising AllNoun, both semantics and phonology were of little or no
importance since he merely used the semantics & phonology of English. What
was important was the syntax, in his effort to show that communication was
possible just using nouns.

Mark Line wanted, among other things, to create a language which would be
as unambiguous as possible with his Classical Yiklamu. Did he spend ages
on semantics. Well, er, no - he simply used what WordNet had already done.
  But he did need to define his phontactics *before* he was able to give
his language a vocabulary. Having defined the phonotactics, he was able to
write a computer program that randomly mapped all the WordNet entries into
unique patterns that were acceptable according to CY phontactics.

I can well imagine that a creator of an _artlang_ might be very concerned
by the phonology & phontactics of her/his language as, by definition, it
is a work of art and is probably aiming at an aesthetic outer form to the

> If you do it first, you're
> almost certain to have to do the whole thing again,
> because it won't work.

Why not? I can think of counter examples.

> And it's the same if you want
> to build a car, a bridge, a house, or whatever.

Not relevant. There is basically one reason for building a car - to get
from A to B; one reason for a house - a place to live. There is not one
reason for creating a conlang - there are many different reasons and on
this list it has been normal to respect different reasaons and different

If semantics is the most important thing for you, then by all means start
a thread on semantics, but please don't tell us that this branch or that
branch of linguistics is more or less important than another.

> --- David Peterson <[log in to unmask]> wrote:
>> Philippe wrote:
>> <<So it's just the same for a natlang, or for a
>> conlang.
>> It is possible to exchange without bothering about
>> talking and pronunciation. It's the same for Chinese
>> ideograms too, AFAIK. This clearly proves that
>> phonology is a peripheral question, not a central
>> one.>>
>> I don't think this follows.

It certainly doesn't and, as I've hinted above, shows faulty understanding
of Chinese writing.

>> Seems to be that if it's possible to have a purely
>> written language (and it
>> is), then the study of speech must be irrelevant to
>> *all* languages, even
>> spoken ones.   I don't understand how that follows,
>> either.

Nor do I understand how it follows either. To me it seems an obvious
non_sequitur, and a pretty barmy one too.

Can't we have more positive threads? If someone's not interested in
phonology they're not obliged to read those mails that deal with. But
please don't whinge and dogmatically state that this or that is the most
important area of linguistics & one must do things in such and such a way.
  Be positive and start a thread on semantics or whatever - and remember
some of us on this list have quite different aims in creating conlangs.

Most of us do it for fun  :)

[log in to unmask]    (home)
[log in to unmask]   (work)