Re: Niggahiita in IPA?

My suggestion on how to represent niggahiita in IPA is as follows.

a.m  =  a~a
i.m  =  i~i
u.m  =  u~u
e.m  =  e~e
o.m  =  o~o

where '~' preceding a vowel means that vowel with a tilde above it, the
standard IPA representation of a nasalized vowel.  That is, niggahiita
is phonetically a nasalized extension of the vowel which precedes it.

This interpretation is not incompatible with passages like those quoted
from Buddhaghosa or Buddhapiya, but neither is it based on them.  I
doubt that they are intended primarily as phonetic descriptions; rather
they seem to be attempts to explain or motivate the name 'niggahiita',
as suggested by Ole Pind.  The absence of concern with phonetic detail
is noteworthy in the Pali tradition.  Every grammar treats sandhi, but
there are no phonetic treatises (to my knowledge).  Perhaps it was felt
that the Sanskrit phoneticians had exhausted the subject.

This interpretation is based in part on Sanskrit phoneticians'
descriptions of anusvaara, and in part on general phonological
considerations.  Pali has five distinct nasal consonants at the
beginning of a syllable (the vagga nasals; I second the proposals to
represent these in IPA except I don't think any subscript bridge is
necessary for the dental n).  But it only has one nasal consonant at
the end of a syllable.  This is assimilated to a following vagga
consonant; if it precedes a non-vagga consonant (such as s) or is not
followed by any consonant it will appear as niggahiita (a bit
oversimplified, but its pronunciation is either fixed or freely
variable).  Pali also has another, oral syllable final consonant: the
first part of geminates such as kk, ggh, ss, etc.  There is no distinct
orthographic symbol for this one (and no special name) because unlike
the nasal syllable final consonant it does not occur unassimilated to a
following consonant.

This kind of syllable structure represents a relatively distinct type
found in oherwise unrelated languages.  Its resemblance to modern
Japanese, for example, is uncanny.  Japanese also has a single syllable
final nasal consonant (often romanized as N) and a single oral nasal
consonant (sometimes romanized as Q, but more often simply as a
geminate).  In Japanese kana orthography (a syllabary with symbols
largely derived from Chinese characters, but whose organization is
Indian) there is a unique symbol for each of these even though they are
often (in the case of Q, always) phonetically the same as one or the
other syllable initial consonant.  The pronunciation of N when
unassimilated is arguably like that suggested above for niggahiita.
Japanese differs from Pali is having a more restricted number of
consonants and in leaving the application of assimilation (sandhi)
unexpressed in its orthography. 

Given that IPA exists to represent phonetic detail, that we do not have
access to Pali speakers or recordings from early days, and that
grammars provide only vague evidence for the details of pronunciation,
we have to recognize that any attempt to render Pali in IPA is
speculative.  However I submit that the interpretation of niggahiita
given above is not implausible and goes some way to account for the
rather confusing account given of it in the grammars.  There it is
rather grudgingly included among consonants, but given a unique
position at the end of the inventory of sounds.  This is because it is
really a vowel, one without parallel to other vowels and no fixed
pronunciation at that.

George Bedell



*   *   *   *   *
George Bedell
120/2 Palm Springs Place
Mahidol Road, Chiang Mai 50000
THAILAND
+66 (0)53-241342
correspondence in Japanese may be addressed to
      gbedelljp@...

Send instant messages to your online friends http://uk.messenger.yahoo.com