Lukasz K. wrote:
>I need some help. I'm considering for some time a monosylabic conlang and I
>think tones would be very suitable for it, but...
>Tones are a very exotic thing for me and I have rather vague idea of
>uttering them and almost no idea of their possible origin. How can tones
>come into being? Are there any general rules or tendencies? I assume that
>monosyllabic morphemes of my language are results of reduction of
>polisyllabic words. How can such a reduction affect tones?
>Are there any tonal languages with well known tonogenesis or evolution of
>tones? I expect Tibetan to be quite interesting, but I don't have any
>on it and I havn't found anything in the net.

There are a number of interesting refs. on google (search "tonogenesis"),
but nothing very specific.  Check jounals devoted to Sino-Tibetan for
articles by James Matisoff, who IIRC is responsible for coining the term.
Years ago I sat in on a course of his, but didn't alas pay much attention,
and don't have any notes.

Languages in the Austro-Asiatic family (Mon-Khmer, Vietnamese) would seem to
be a good place to start; some of them have developed tones, others have not
(yet); Khmer itself is sometimes said to be "in process" of developing them
with its two-register system (normal vs. breathy vowels). One might think
that Vietnamese developed tones due to extensive contact with Chinese, but
IIRC Matisoff claimed it was an independent development.

This applies I think, to tone languages of the Chinese sort; African
languages seem to have different systems.  Mayan languages are also tonal,
but I know next to nothing about them.....

But among the things I remember are:  initial voiced stops (especially) tend
to produce low tones; then if, in addition, the stops devoice you get
contrasting **p (high tone) < *p-, **p (low) < *b-.  Also, it seems, there
are tendencies final -h > high, final -? > low.  And I'd assume that stress
placement on bi- or polysyllabic forms would produce different tones.

I have the same problem with my Gwr language, but the details aren't worked
out yet.  So far it appears that, given a CVCVC proto-form, with variable
stress-- some of it morphologically conditioned-- almost every possible
sequence of V and C can produce a great many tone patterns and  vowel
changes (from a 3-vowel system to 9).

Keep us posted on your progress!!