LINGUIST List 7.860

Mon Jun 10 1996

Sum: palatal nasals

Editor for this issue: T. Daniel Seely <dseelyemunix.emich.edu>

Directory

1. wclivax.ox.ac.uk, summary: palatal nasals

SUMMARY: palatal nasals

A while ago I asked for examples of languages with word-final palatal nasals
that contrast with alveolar and velar nasals. Here are the replies. Thanks to
Lance Eccles, Jakob Dempsey, Rod Johnson, Mark Mandel, Paul Foulkes, Geoffrey
Sampson, John Koontz, James Harris, Suzanne Kemmer, Ilona Kassai, Bruce A.
Connell, Mariana Maduell, Keith Goeringer, David Stampe, who contributed to
this summary. The languages Catalan, Vietnamese, Cambodian, Written Burmese,
Sulung, Luo (Western Nilotic), Hungarian, Munda (Austroasiatic) and Dinka were
suggested.

Wen-Chao Li
Lady Margaret Hall
Oxford University

.........................................
LUO

From: kemmerruf.rice.edu (Suzanne E Kemmer)

I am doing fieldwork on Luo, a Nilo-Saharan language
of Kenya, which has the contrast in final nasals which
you are looking for:

(~ indicates palatalization; ng represents the velar nasal,
not an n-g sequence; and c is a palatal stop)

un 'you plural'
cun~ 'heart'
cung 'stand'

(I haven't indicated vowel length or tone)

- Suzanne Kemmer
..........................................................
LUO

From: rcjmail.msen.com (Rod Johnson)

The Western Nilotic
language Luo (also known as Dholuo) has contrastive final
nasals at the bilabial, alveolar, palatal and velar places,
and used to have dental nasals as well (they have since
merged with alveolars). I can't cite any minimal pairs off
the top of my head, but Luo abounds in forms like _chuny_
'liver', _apany_ 'one-legged stool' and _piny_ 'land' (where
the _ny_ digraph is the standard spelling for the palatal
nasal).

There is also, interestingly, a series of prenasalized stops
in Luo that corresponds to the plain nasals--the two series
alternate with one another in various morphological
contexts, but they are phonemically distinct. Final voiced
segments are mostly prohibited, except in certain deverbal
nouns, so there aren't many examples, but forms like _puonj_
'teacher' are found (where _nj_ is the voiced palatal
prenasalized stop).

Oh, I can cite one minimal pair: the _chuny_ above and
_chung'_ 'to stand' (where _ng'_ is the velar nasal). I'm
sure there are others.

I hope this is useful. Luo is not terribly well known in
the literature, but I can send some references if you're
interested. I think similar phenomena occur in other
Nilotic languages: for instance, in Shilluk, _kyEN_ 'horse'
(where _E_ is IPA epsilon and _N_ is the palatal nasal), and
in Dinka-Nuer, but I'm not as conversant with them.

Rod Johnson
rcjmsen.com
...................................................
MUNDA (phylum Austroasiatic)

From: stampehawaii.edu (David Stampe)

Many of the Munda (phylum Austroasiatic) languages of India have
alveo-palatal vs alveolar, velar, and labial nasals in syllable and
word final position. Most have the corresponding non-nasal stops as
well, though occasionally there are gaps (Sora has b d j but not g
finally, though it has m n ~n and ng finally). In most of these
languages, the palatal nasal (and palatal stops, usually j [dzh]), if
it closes the syllable (i.e. if a vowel does not follow in the same
word) causes the preceding vowel to have a palatal [i-like]
transitional glide, which considerably aids in hearing the distinction
between the various lingual nasals.

If you need further information - examples, references, etc. - please
let me know by email what you need.

David Stampe <stampehawaii.edu>
Dept. of Linguistics, Univ. of Hawai`i/Manoa, Honolulu HI 96822
.....................................................................
DINKA, CATALAN

From: connellbvax.ox.ac.uk
Subject: final palatal nasals

Wenchao

Catalan I think contrasts alveolar and palatal nasals (but not velars) in
final position.
Dinka (a NIlotic language spoken in Sudan) contrasts labial, alveolar,
palatal and velar nasals finally.

A similar situation may exist in other languages of that region.

Bruce
...........................................................
CATALAN

From: maduellhawaii.edu (Mariana Maduell)

I can't think of any contrasts offhand, but Catalan has word-final palatal
nasals, represented by ny (e.g. any `year'). These came in with the dropping
of some final vowels (the Spanish cognate is obviously an~o, Italian anno).

Mariana Maduell e-mail: maduellhawaii.edu
P.O. Box 62206 http://www2.hawaii.edu/~maduell
Honolulu HI, 96839
......................................................
CATALAN

From: jharrisMIT.EDU

Catalan (though unfortunately not a Sino-Tibetan lg.) contrasts word-final
palatal nasals (spelled *ny* in the standard orthography) with alveolar, velar,
and labial (which you didn't mention) nasals --all of which appear plentifully
i
n common native words. The only qualification is that the velar nasal in
phonetic representations is from underlying Nas plus velar obstruent (the
latter deleted in word-final position but not before a V); the labial, palatal,
and alveolar nasals are all underlying.

 Examples: any 'year'
 han 'they have'
 flam 'flan - the custard'
 baN 'bank' (N=engma)

James Harris
................................................................
CATALAN, VIETNAMESE, CAMBODIAN

From: leccleslaurel.ocs.mq.edu.au (Lance Eccles)

Catalan distinguishes a word-final palatal nasal from both a velar and an
alveolar/ dental nasal:

bany = bath
ban = edict
banc [baN] = bench; bank

I believe Vietnamese makes such a distinction (at least in some
dialects), and this this is represented in spelling by -nh, -n, -ng. You
could check for details in Nguyen Dinh Hoa, Speak Vietnamese (Tuttle,
Rutland, 1966) -- there may be a later edition -- and in Laurence
Thompson, A Vietnamese Grammar (U of Washington Press, Seattle, 1965).

All three also occur in Cambodian. Check Judith Jacob, Introduction To
Cambodian (Oxford U Press, 1968), p. 22.

-Lance Eccles	
..........................................
VIETNAMESE

From: geoffscogs.susx.ac.uk (Geoffrey Sampson)

It depends how exact you are about what counts as a "palatal" sound.
If you are willing to be a little loose, Vietnamese (at least in its
Northern variety, traditionally regarded as the standard variety)
should count. It has syllable-final sounds written in the normal
Vietnamese orthography -ch, -nh, but they are not exactly lamino-alveolar,
which is what "palatal" standardly means; I describe them as "fronted
velar". They have very restricted distribution, but there would be
words in which they contrast with ordinary final velars written -c, -ng:
if the latter consonants are preceded by a vowel spelled as "a" with
a lunette (short mark) above it, the resulting syllable-final would
be identical to a final written -ach or -anh except for the precise position
of the velar closure. The -ch and -nh syllables seem to result from earlier
syllables ending with ordinary velars, to which a sound-change applied
altering both vowel and consonant.

The facts were discussed in a fair degree of phonetic detail in my first-ever
academic publication, "Hanoi dorsal finals", BSOAS vol. 32 pp. 115-34, 1969.
....

I'm reconstructing memories of things I knew about 30 years ago but have
scarcely revisited since; but broadly, what seemed to have happened in
VN on the basis of internal reconstruction was that earlier long tense
vowels before velar consonants were made short, lax, lowered, and
their frontness or rounding was spread to colour the velar consonant also,
so that a velar consonant after a front vowel became fronted, and a
velar consonant after a back rounded vowel became labialized or even
became a k^p coarticulation.

Down the years I have occasionally seen information about Southern Chinese
dialects which suggested to me that this might have been an areal sound-change
which affected them too. (As I'm sure you know, Vietnamese is a language
which genetically was not Sino-Tibetan at all but which acquired so much
material from Chinese that modern Vietnamese looks like a Chinese dialect
with a minority of vocabulary items that have no Chinese cognates.) I
always meant to look further into this, but there was never time and
my career took me in other directions.
I would say again, though, that despite the way that the romanized script
invented for Vietnamese by a 17c Jesuit missionary chooses to write them,
the "fronted velar" final consonants are definitely not phonetically
identical to initial palatal consonants (which VN has), and are not
"palatal" in the strict sense which that term usually bears in phonetics.
Since I don't know the motive of your enquiry, I don't know whether they
are likely to be relevant for your purposes or not.

Geoffrey Sampson
........................................................................
HUNGARIAN

From: kassainytud.hu (kassai)

Sorry to report on a language other than sino-tibetan which contrasts
word-final palatal nasals with word final alveolar nasals: Hungarian.
It is a member of the Ugric branch of the Finno-Ugric language family.
It is closely related to Vogul, Ostyak (West
Here are some minimal pairs where the first member contains palatal nasal
(<ny>in spelling) while the second one shows up alveolar nasal (<n>in spelling)
.
 The
' marks the length of the vowel before it:

ke'ny pleasure/ke'n sulphur
ide'ny season/ide'n this year
Uny geographical name/un be fed up
ve'ny prescription/ve'n old

And now your new question. The palatal nasal has always been a genuin phoneme
of the Hungarian sound system. It goes back to the finno-ugric palatal nasal. A
s
for velar nasal, on the other hand, it is'nt an independent phoneme, it is only
an allophonic variant.

Ilona Kassai, Linguistics Institute of the Hungarian Academy of Sciences, Budap
est.
...................................................
HUNGARIAN

From: kegviolet.berkeley.edu (Keith GOERINGER)

I can't give any Sino-Tibetan data, but Hungarian does contrast at least
the palatals and alveolar nasals word-finally. There are also word-final
velar nasals, but I'm not sure that they're entirely pure nasals -- there
might be some [g] coming through as well. I can't think of a perfect
three-way minimal triad, but some minimal pairs (assuming the velar nasals
are just nasals, unless it doesn't matter) would be:
la'ny 'girl' la'ng 'flame'

 sza'n 'pities' (verb) sza'nj 'pity' (command)

(the digraphs /ny/ and /nj/ are the palatal nasal; the digraph /ng/
represents eng; the digraph /sz/ is [s]; and the apostrophe marks historic
length which is qualit(at)ively distinguished in modern Hungarian)

A good place for finding more pairs (or possibly triads) would be a reverse
dictionary -- I can give the name of one, if you want. Also, as the second
example shows, in verbal morphology, the subjunctive (or imperative) forms
generally is marked by palatalization of the root consonant. So if you
seek out verbs whose stems end in -n, that should give you some nice
minimal pairs.

Keith Goeringer
UC Berkeley
Slavic Languages & Literatures
kegviolet.berkeley.edu
.....................................................
BURMESE, SULUNG

From: ufjakobqms5.hinet.net (jakob)

Written Burmese appears to have such finals, but this is probably an
artifact of the script, the actual pronuciations (present and historical)
are/appear to have been quite different.
Sulung (spoken at the Eastern end af the Himalayan chain -Arunachal
pradesh - marginally belongs to this type. It is found in an area where
the Sino-Tibetan languages tend to have a "Mon-Khmer" character.
Speaking of which, Diffloth's proto-Waic has contrastive palatal nasal
endings, which is based on similar finals being found in many modern Wa
(and other ?) Mon-Khmer languages.

- -Jakob Dempsey (Taiwan)
.......................................................
AUSTRALIAN LANGUAGES

From: koontzboulder.nist.gov (John E. Koontz)
Subject: Nasals

Look at Australian languages. A good starting reference might be the volume
in the Cambridge Language Series.

John E. Koontz
NIST:CAML:DCISD 888.02 Boulder, CO
john.koontznist.gov