Re: Some conlang questions

Quoting Michael Fors <micke@...>:

> Hello everybody!
>
> I'm new to this group, so I thought that I should introduce myself. My name
> is Michael Fors, I'm 21 and I live in Sweden. I have been interested in
> languages and language construction for some time, but I just recently
> started trying to create a language of my own.
>
> To my questions:
> 1. I am having problems with the vowels of my conlang. Since I want to make a
> naturalistic language, I want a naturalistic vowel system. The only problem
> is that I don't know what that is. I mean, is it natural for a language to
> have many front vowels, rounded and unrounded, and two rounded back vowels?
> That's one of the questions that keeps me awake at night. =)
The short answer is "no", but don't let that worry you too much.
Basically, roundedness is an acoustic marker of backness in a vowel,
which means that front rounded vowels and back unrounded vowels are
both structurally marked classes and so tend not to show up as often in the
inventories of vowels systems as their oppositional counterparts. But
it really depends;  /u/ is one of the most unmarked vowels there is,
yet at least one language I know (Meskwaki) lacks it entirely, despite
having both long and short /i e o a/. So natural languages have
considerable "fudge" room.

> 2. I also want to make "daughter"-languages to my language. I have been
> reading some about sound change, but I wonder if there is some sort of a
> table for sound changes? So I can see which changes that are likely to occur
> and which that are less likely.
I don't know of a "table" of such changes, but the best way to
go in my experience is to think about it terms of abstract
structure:  don't ask which particular sounds are good/bad, but
where are the sounds are likely to occur, or whether a feature
of a set of sounds favors/disfavors them.  So, for example, codas
(consonants that close a syllable) are marked, and so tend to get
lost in languages.  Onsets (consonants that open a syllable) are
unmarked, and so tend to get added even where they didn't exist
previously.  There are various levels of structure to look at, so
you're not terribly limited by this viewpoint.  Sounds that are
both e.g. fricative and voiced are marked, and so tend to get lost/get
changed over time to some other class of sounds less marked. These
kinds of considerations just brush the surface.

 =========================================================================
Thomas Wier            "I find it useful to meet my subjects personally,
Dept. of Linguistics    because our secret police don't get it right
University of Chicago   half the time." -- octogenarian Sheikh Zayed of
1010 E. 59th Street     Abu Dhabi, to a French reporter.
Chicago, IL 60637