Spontaneous nasalisation and the
development of nasals from non-nasals are perfectly possible, though not very
common. I'll post some examples shortly.
Piotr
----- Original Message -----
Sent: Thursday, July 04, 2002 11:29 AM
Subject: [phoNet] Nasals from Non-nasals
Dear Linguists,
I remember reading a comment a long
time ago that there seemed to be no way
for a language without nasals to
develop nasals. The implication of this
comment was that nasal segments
could only originate from or under the
influence of nasals. How true is
this claim?
There appear to be some counterexamples to the implication,
but perhaps they
are all neutralisations of nasal-nonnasal
contrasts:
(a) Change of /y/ to /?/ (palatal nasal) in the development of
Lao from
Proto-Tai (whereas Siamese shows the opposite change /?/ to /y/),
and both
simplified /?y/ to plain /y/). (My source is Fang-Kuei Li's
'Handbook of
Comparative Tai'.)
(b) Change of /?b/ to /m/ in both Shan
(SW Tai) and in the Po-ai (SE corner
of Yunnan province of China) dialect of
Northern Tai. Li reconstructs
Proto-Tai as having preglottalised labial
and dental plosives but not
preglottalised nasals, whereas some related
languages do contrast
preglottalised plosives and preglottalised
nasals. /?d/ yielded /n/ in
Po-ai but /l/ in Shan. In most Tai
dialects, the preglottalised plosives
(/?b/, /?d/) have become (occasionally
implosive) voiced plosives. The
Northern Tai dialect of Wuming retains
preglottalised plosives.
This may have been a difficult
reconstruction. Sui, a relative of the Tai
languages, has /?b/
corresponding to Proto-Tai */?b/ and both /?m/ and /hm/
corresponding to
Proto-Tai */hm/. Correspondences include ?baan = *?baan
'house' and
?maa = *hmaa 'dog'. However, Tai loans into Khmu?, which is
spoken in
Thailand and Laos, show ?m- from *?b-.
(c) Replacement of final /l/, /r/
by /n/ in languages lacking final
laterals. Possibly also when final
laterals are eliminated form the
phonological system? Li reports that
all Tai dialects but the 'aberrant
Northern Tai dialect' Saek have
merged final /l/ with /n/.
I don't know how to reckon the intrusion of
'n' into English 'passenger' and
'messenger'. (Is there a Greek word
for this peculiar intrusion?)
I don't include assimilatory nasalisations
in this category. Dissimilations
I would regard as only a minor
counter-example - not that I can think of any
at the moment. (I have a
dim recollection of some English word
dissimilating ..r..r.. to ..r..n..,
possibly in Old French.)
I have seen a claim (at http://groups.yahoo.com/group/nostratic/message/450)
that
Eskimo-Aleut voiceless plosives becomes nasal when word final, but
that
source is at best secondary. If this is correct, and not the
result of say,
analogy arising after an intervocalic change /nt/ to /tt/, it
would be a
clear counter example to the rule that nasals only derive from
nasals.
Richard.