What exactly is the Tamil /zh/ sound
M. Ranjit Mathews
general consumption:
Yahya wrote:
>I am still trying to get a fix on the Tamil voiced alveolar ...
>whatever it is.
Dkcsac (dkc...@aol.comment)>
>>Judging from the descriptions given, you put your tongue in the
position of
>>the English R, that is, with the tip curled up close to, but not
quite
>>touching the roof of the mouth, and then leaving it there, try to
pronounce
>>the Z of "zoo". The sound produced is halfway between the Z of "zoo"
and the
>>S of "pleasure".
Malayalam & Tamil /zh/ is a retroflex frictionless continuant. Where
it occurs between vowels, it is mispronounced by some Malayalis as [j]
(kozhikod is apparently pronounced by some Malayalis as Koyikod) and
mispronounced by many Tamizhans (Tamilians) as [l.]. Where it occurs
before a consonant, it is usually pronounced correctly as far as I can
remember. While it is frictionless, its pronunciation is close to the
Pinyin (Mandarin) fricative r and I think you will be understood by
Malayalis and Tamils as saying /zh/ if you pronounce this in place of
/zh/. Alternatively, pronounce it as the frictionless continuant r in
a rhotic English pronunciation of nurse.
Audio clips of Pinyin (Mandarin) r:
http://www.courses.fas.harvard.edu/~pinyin/pinyin6.htm
nanthica
Example.. Valapalam in Tamil.. Malayalam is with correct sound for the
letter Vazhapazham.
M. Ranjit Mathews <ranjit_...@yahoo.com> wrote in message
news:1d4c67e3.02050...@posting.google.com...
M. Ranjit Mathews
> I think many Malayalis still talk the correct usage of the words.
> Example.. Valapalam in Tamil.. Malayalam is with correct sound for the
> letter Vazhapazham.
As far as I've heard, Malayalis always pronounce it correctly, but I
asked someone from North Kerala on Saturday whether there are people
who mispronounce it and he said some people in North Kerala
mispronounce /zh/ as y. In Tamil, it is vaazhaipazham (not
vaazhpazham) and is pronounced correctly by some people.
> M. Ranjit Mathews <ranjit_...@yahoo.com> wrote ...
M. Ranjit Mathews
"nanthica" <nant...@accessv.com> wrote ...
> I think many Malayalis still talk the correct usage of the words.
I have never heard a Malayali mispronounce it, but I'm not intimately
familiar with the speech in North Kerala so I asked someone from
Kannur and he said that some people in North Kerala mispronounce /zh/
as y.
> Example.. Valapalam in Tamil.. Malayalam is with correct sound for the
> letter Vazhapazham.
vaazhapazham in Malayalam becomes vaazhaipazham in Tamil (i.e., the
Tamil word has an 'ai' in place of an 'a').
Nanban
If you start from Madras and go south towards Kanyakumari you can
identify this. Tamil is written in one form but spoken in many form.
It is just that one form of spoken Tamil gave birth to Malayalam.
Malayalam is spoken and written in one form for the most part.
I am from Nellai (Titunelveli) district. If we compare the way we
speak Tamil compared to the people from Nagercoil speaks Tamil, I can
say their spoken Tamil is more close Malayalam than ours.
We use AMMa (short a) but we write AMMA (long A).
Anpudan
Nanban.
M. Ranjit Mathews
umari...@yahoo.co.in (Nanban) wrote ...
> That is the Malayalam is. Malayal is nothing but a dialect of TAMIL.
It was one of many dialects of Tamil a millenium back, but it is not a
dialect of Tamil today.
> If you start from Madras and go south towards Kanyakumari you can
> identify this.
Of course, the relationship between Tamil, the mother of Malayalam,
and Malayalam is obvious.
> Tamil is written in one form but spoken in many form.
> It is just that one form of spoken Tamil gave birth to Malayalam.
> Malayalam is spoken and written in one form for the most part.
>
> I am from Nellai (Tirunelveli) district. If we compare the way we
> speak Tamil compared to the people from Nagercoil speaks Tamil, I can
> say their spoken Tamil is more close Malayalam than ours.
IMHO, the best spoken Tamil is spoken in Kanyakumari, especially the
pronunctiation of /zh/, r[(the first r), and n_ (the last letter of
Tamil)
> We use AMMa (short a) but we write AMMA (long A).
Except when asking for bhiksha (picca):-)
nanthica
Malayalis, Sri Lankan Tamils and Sinhalese have common customs and all these
three groups are differ from Tamils of Tamil Nadu!
Even the language of sri Lankan tamils are closer to Malayalam than the
Tamil of Tamil Nadu!
Nanban <umari...@yahoo.co.in> wrote in message
news:6bc56bcc.02050...@posting.google.com...
Peter T. Daniels
>
> I think you are not familiar with MALAYALAM!
> Malayalis, Sri Lankan Tamils and Sinhalese have common customs and all these
> three groups are differ from Tamils of Tamil Nadu!
> Even the language of sri Lankan tamils are closer to Malayalam than the
> Tamil of Tamil Nadu!
Common customs are totally irrelevant to linguistic relationships.
--
Peter T. Daniels gram...@att.net
M. Ranjit Mathews
> I think you are not familiar with MALAYALAM!
> Malayalis, Sri Lankan Tamils and Sinhalese have common customs and all these
> three groups are differ from Tamils of Tamil Nadu!
> Even the language of sri Lankan tamils are closer to Malayalam than the
> Tamil of Tamil Nadu!
Some aspects of ILankai Tamizh, such as its intonation, some of its
phonology and some of its vocabulary, bear similarities to Malayalam
but the language is, on the whole, closer to Tamil, especially in
lexical similarity and substantially shares Tamil phonology (even if
not intonation) including many traits such as shifts from front vowels
to back vowels (eg. pennu pronounced as ponnu) & pronunciation of /zh/
between vowels as /L/. Most of Lankan Tamil's vocabulary is shared
with Tamil; only some words uncommon in spoken Tamil are common to
both Lankan Tamil and Malayalam. In cuisine too, there are some
respects in which between lankan Tamils have more in common with
Malayalis.
> Nanban <umari...@yahoo.co.in> wrote ...
nanthica
and SL. If you read the books which were published hundred years ago, you
will know how the closeness of SL Tamils and Malayalees!
After the arrival of British only this commoness of Tamil is created but
before the British, SL Tamils used Tamil alphabets but the language was very
close to Malayalam!
Portugeuse and Dutch brought a sizable malayalee population to Jaffna and
other parts of Sri lanka in 16, 17 and 18 centuries. The traditional law of
Easter Sri lanka is still called MUKUWA LAW. Further MARUMAKKA THAYAM was
the basis for the traditional law of Jaffna and North Sri lanka!
But SL Tamils never follow any customs of TN Tamils. ex. Marriage. SL TAMILS
marry AMMAN's (AMMAVAN's) daughter who is the MURAI PENN for SL TAMILS but
TN Tamils marry their (AKKA's Magal) niece!
Further Malayalam and Tamil share a lot in common!
M. Ranjit Mathews <ranjit_...@yahoo.com> wrote in message
news:1d4c67e3.02050...@posting.google.com...
nanthica
Majority of the customs develop with language and religeous practices!
Blacks came to US and adopted Christian practices but their Fellow-Blacks in
Africa do not but American blacks are still unable to become 100%
Christianized or Anglized!
I do not know what do you mean by COMMON CUSTOMS?
Peter T. Daniels <gram...@att.net> wrote in message
news:3CD7BA...@att.net...
Peter T. Daniels
>
> That is not true!
> Majority of the customs develop with language and religeous practices!
> Blacks came to US and adopted Christian practices but their Fellow-Blacks in
> Africa do not but American blacks are still unable to become 100%
> Christianized or Anglized!
>
> I do not know what do you mean by COMMON CUSTOMS?
I meant to copy the phrase you used. What did YOU mean by it?
However:
Did you try reading what I wrote?
"American blacks" speak English, either a variety that is identifiable
with a black community or one that isn't.
That doesn't make English a member of, say, the Niger-Congo phylum, to
which belong most of the languages of their ancestors who were imported
as slaves centuries ago.
M. Ranjit Mathews
Not quite irrelevant. There are castes that have dialects pecular to
them and there are castes that are considered closely related by
virtue of having many common customs and speaking closely related
dialects.
Peter T. Daniels
Bring a linguist unidentified recordings of these languages
(sufficiently large specimens, of course), and s/he will determine the
relationships among the dialects with no reference whatsoever to the
social status, customs, etc. of the speakers.
M. Ranjit Mathews
What does that have to do with the price of Lima beans in Timbaktu?
All that was observed was that in Indian mileu, where there is
similarity of customs between subgroups of a linguistic population
group, there is also a concomitant similarity of dialect. Hence, the
question of whether two subgroups customs are similar is relevant to
the question of whether their dialects are similar.
Of course, a linguist won't need any of this information in order to
be able to determine the relationships between the speakers' dialects
if tape recordings are available, but since no tape recordings have
been posted on Usenet, whatsHisName provided other data that was
germane to the answer to the question of whether two certain
linguistic subgroups were similar.
Peter T. Daniels
It absolutely is not.
In Papua New Guinea, for instance, utterly unrelated languages can be
spoken by peoples living side by side and interacting every day whose
lifestyles are all but indistinguishable to the casual observer.
In North America, the same language is spoken by Weans and Canadians,
and no one could suggest that Canadian and US cultures are remotely
similar!
> Of course, a linguist won't need any of this information in order to
> be able to determine the relationships between the speakers' dialects
> if tape recordings are available, but since no tape recordings have
> been posted on Usenet, whatsHisName provided other data that was
> germane to the answer to the question of whether two certain
> linguistic subgroups were similar.
They certainly were not. (Germane.)
M. Ranjit Mathews
That is in Papua on the Indonesian Archipelago with its 3,000+
languages, not in Tamilnadu. The rule doesn't apply unless one knows
apriori that it applies in some particular part of the world. We know
apriori here that the two groups speak Tamil. The only question was
which dialect they speak. Which dialect they speak depends to a great
extent on which part of Tamizhaham they moved from to Sri Lanka. The
customs provided the information as to which part they originated
from. If their customs originated from a certain place the odds are
that their dialect too originated there. It is known a priori that
there isn't a wide variety of dialects in population groups that fit
this description in that part of the world; ergo ...
> In North America, the same language is spoken by Weans and Canadians,
> and no one could suggest that Canadian and US cultures are remotely
> similar!
>
> > Of course, a linguist won't need any of this information in order to
> > be able to determine the relationships between the speakers' dialects
> > if tape recordings are available, but since no tape recordings have
> > been posted on Usenet, whatsHisName provided other data that was
> > germane to the answer to the question of whether two certain
> > linguistic subgroups were similar.
>
> They certainly were not. (Germane.)
Hiopefully, you will never have the msifortune of being assigned to a
detective case in Tamizhaham where you have to track down your suspect
solely by his customs:-) That would leave you with little germane
information, what? Your junior fellow detective approaches the suspect
on a hunch, detecting a certain peculiarity in his accent. Later, at
the awards ceremony, your partner asks as you congratulate him for his
award, "How did you manage to miss his telltale accent? You could have
bagged the award, you know?". "Only his customs were noticed", you say
stiffly; "They were not germane to his accent." Now, you wonder why
all the detectives call you Patros Dumkof:-)
Suppose a lady were accosted in Montana by a bag snatcher wearing a
drooping moustache and a sombrero. You look around but only see
baseball caps and bloinde heads. You feel a sudden urge and walk
toward a bush to relieve ourself. The thief, hiding behind the bush
thinks you're coming for him and says "Mama mia; Madre de Deus!".
Would you be unable to make a connection between the sombrero, etc and
his expostulation?
Peter T. Daniels
I have no idea what you're going on about.
The matter at hand is not whether you can identify someone's birthplace
on the basis of their accent!
The question is whether similarity of customs has any necessary
connection with similarity of language.
M. Ranjit Mathews
No; it was whether one could identify one particular group of Tamil's
ancestors' place by their customs.
>
> The question is whether similarity of customs has any necessary
> connection with similarity of language.
It does not necessarily. In this particular case, it did.
M. Ranjit Mathews
> Anbu Nanbane,
> > We use AMMa (short a) but we write AMMA (long A).
> Except when asking for bhiksha (picca):-)
TWICE!
"Ammaa; taaye!":-)
For those who don't know Tamil, here is the pun explained:
Amma means mother, but is also a salutation meaning Lady.
taaye means please give (a handout), but also means mother
(accusative).
On a more serious note:
This phenomenon is common in Indian phonology; at the end of a word, a
short a is treated as a schwa [@] and a long a is approximately a
stressed a [V] in many contexts even if not all.. In Hindi, a short a
at the end of a word is often, if not invariably, treated as no vowel
at all unless it follows a consonant cluster in which case it's a
schwa. For example, the last vowel in "Krishna" is a schwa. In Nagari
Kashmiri, a short at the end of a word is always a schwa, a long a is
always long, and a stressed a is a long a followed by an udatta matra
(diacritic).
M. Ranjit Mathews
> general consumption:
>
> Yahya wrote:
> >I am still trying to get a fix on the Tamil voiced alveolar ...
> >whatever it is.
The following .au clip reproduces perfectly ONE OF the native
pronunciations of this letter (romanised to English as zh). This is an
alveolar pronunciation; the other pronunciation is retroflex.*
http://www.unil.ch/ling/phonetique/SonsAPI/Consonnes/inverted_r.au
On the IPA chart, it is represented as an upside down r - the two
upside down r's to the left of the j.
http://www2.arts.gla.ac.uk/IPA/pulmonic.html
* For brevity (and for lack of expertise), I won't comment on which
contexts find greater use of which of {alveolar, retroflex}
M. Ranjit Mathews
> > >I am still trying to get a fix on the Tamil voiced alveolar ...
> > >whatever it is.
>
> The following .au clip reproduces perfectly ONE OF the native
> pronunciations of this letter (romanised to English as zh). This is an
> alveolar pronunciation; the other pronunciation is retroflex.*
> http://www.unil.ch/ling/phonetique/SonsAPI/Consonnes/inverted_r.au
Here is the retroflex sound, which is the more typical (than the
alveolar above) pronunciation of 'zh' in Malayalam and in the
pronunciation of those who speak Tamil properly/ didactically.
http://www.unil.ch/ling/phonetique/SonsAPI/Consonnes/long_l.au
> On the IPA chart, it is represented as an upside down r - the two
> upside down r's to the left of the j.
> http://www2.arts.gla.ac.uk/IPA/pulmonic.html
Egad, it seems that I might be mistaken in assuming that the retroflex
'zh' is the letter immediately to the left of the j.
I'm surprised that a retroflex 'zh' is represented by a long l in IPA.
If this is so, how on earth is the upside down r immediately to the
left of the j pronounced (I can't seem to find a sound clip for that)?
Is it pronounced in a manner very similar to a long l or is it
noticeably different? It seems counter-intuitive to me that the IPA
symbol for a retroflex 'zh' is a long l rather than the other upside
down r to the right of what sounds to my ear like an alveolar 'zh'.
Before listening to these clips, I had thought that a long l in IPA
was the second l in Malayalam and Tamil but now, I find that the
second l is classified in IPA as alveolar rather than retroflex* and
that the Malayalam l's are [l[] and [l] rather than [l[] and [l.] as I
had thought earlier.
* I've pronounced what IPA seems to call an alveolar l as a
lamino-retroflex dorsal ever since I learnt to speak. Ask any Malayali
to say the words puLi, puLLi, kaLi and kaLLan and ask him to describe
how he modulates his tongue to render the L; I would be surprised to
hear that he pronounces it as an alveolar.
FYI (if knowing one of my other pronunciations will be helpful to you
in resolving my confusion about why a retroflex 'zh' is not the upside
down r to the immediate left of the j), my t's are apicodental rather
than alveolar in my pronunciation of many words (eg. "testy") but are
apicoalveolar in my pronunciation of other words (eg. putty).
M. Ranjit Mathews
> >I am still trying to get a fix on the Tamil voiced alveolar ...
> >whatever it is.
Here's my last word on the subject since I don't have sufficient
knowledge to say any more. With keen interest do I await responses
with insightful observations.
For the purpose of this discussion, let us call the Malayalam/ Tamil
'zh' a Y.
Let us call the r in a rhotic US pronunciation of nurse and nerd an R.
R is a VOICED frictionless continuant.
It might be a retroflex R. as in "nurse" or an alveolar R as in
"nerd".
It is not trilled in a rhotic US pronunciation.
Y is a VOICELESS frictionless continuant.
It is canonically retroflex as in [AY@m] (aazham means depth)
but can also be alveolar as in [mVY@] (mazha means rain)
These two words are the same in both Malayalam and Tamil. A Malayali
from Trivandrum (in Kerala) and a Tamilian from KoLachal (in
Tamilnadu) pronounce them identically.
Now for the pronunciations:
R. (seems too trilled in this clip to be the retroflex frictionless
continuant in "nurse")
http://hctv.humnet.ucla.edu/departments/linguistics/VowelsandConsonants/course/chapter1/ipaSOUNDS/Con-51a.AIFFl
R (this does sound close to the alveolar frictionless continuant in
"nerd")
http://hctv.humnet.ucla.edu/departments/linguistics/VowelsandConsonants/course/chapter1/ipaSOUNDS/Con-40a,AIFF
Y. (devoice this somewhat to get the 'zh' in "aazham")
http://www.unil.ch/ling/phonetique/SonsAPI/Consonnes/long_l.au
Y (devoice this somewhat to get the 'zh' in "mazha")
http://www.unil.ch/ling/phonetique/SonsAPI/Consonnes/inverted_r.au
References:
http://hctv.humnet.ucla.edu/departments/linguistics/VowelsandConsonants/course/chapter1/chapter1.html
If the above webpage is correct, then one or both of the ones below
might be wrong but that suits me well on this occasion since the
upside down r and long l are pronounced (or mispronounced as the case
may be) close enough to the Malayalam 'zh' to serve me well for this
illustration.
http://www.unil.ch/ling/phonetique/api33-eng.html
http://www.unil.ch/ling/phonetique/api322-eng.html
P.S. The Mandarin r seems like a fricative; it doesn't seem either
frictionless or trilled. In my opinion, it sounds close to the
r<hacek> in a native Czech's pronunciation of Dvorak.
M. Ranjit Mathews
links right this time.
R. (seems too trilled in this clip to be the retroflex frictionless
continuant in "nurse")
http://hctv.humnet.ucla.edu/departments/linguistics/VowelsandConsonants/course/chapter1/ipaSOUNDS/Con-51a.AIFFl
R (this does sound close to the alveolar frictionless continuant in
"nerd")