a little more on glottalization and sanskrit loans

[Waruno Mahdi]

> questionable. We know that the actual pronounciation of Tagalog and Ilokano
> final vowels does not differ in any considerable way in respect to
> aspiration, so why insist on positing a silent _h_? In none of the
> Philippine languages known to me which contain this "linking _h_" feature
> does there exist an audible final _h_. Furthermore we see that the real

I think we are simply confusing phonemics with morphophonemics.
Alternation of consonants (e.g. r/d), and alternation between a consonant
and nought (e.g. h/-), at rootmorph-final position, depending upon whether
followed by a suffix or not, is rather widespread in Austronesian languages.
Tagalog is thus not unusual in having this morphophonological feature.
In the orthography, of course, one is free to choose between a spelling
along phonemic principles (e.g. Malagasy). or one along morphophonemic
ones (e.g. French). I see it as a matter of convention, which spelling
one chooses, and a matter of the concrete topic of discussion, which
mode of transcription one prefers to implement.

> (audible) final _h_ which occurs in Indonesian languages is usually
> reflected in Tagalog and other Phil. languages by the final glottal stop
> e.g. Ind. suluh, Tag. sulo' "torch"; Ind. penuh, Tag. puno' "full"; Ind.
> basah, Tag. basa' "wet"; etc.

This are not, however, an instances of reflection of Malay -h as Tag. -',
but the reflection of PAN *q as Mal. _h_ and Tag. _'_ (which is the
regular sound correspondence for inherited items in the two languages).

> distribution of other phonemes. In many cases it seems to be a (late)
> Philippine innovation considering such examples as,
> Sebwano- bag'o, Bikol- ba'go, Tag.- bago  Ilokano.-baru  Ind.-baru   from
> PAN- *baRu  "new".

The protoform is actually *baqeRu(H) (I don't remember whether there's an
*H or not, don't have notes at hand), Cebuano has *q/*R metathesis, Malay
has doublets baru / baharu, whereas _bahru_ is attested either historically
or dialectally (my memory is a bit fuzzy here). Note Kalamian-Tagbanwa
_baklu'_ with regular *q > k, *R > l. I think there is a Lampung dialect
which has _bahyu_ (regular *q > h, *R > y) but I don't have my notes with
me.

> the only point of contention is that these Sanskrit loan words were borrowed
> through a post-glottalizing dialect such as Brunei Malay as previously
> suggested. Is there enough evidence that the Brunei kingdom held suzeranity
> over Luzon for such a protracted period of time to have such a deep impact
> on the language? We find highly-widespread Sanskrit loan words with
> post-glottalization such as _diwata'_ which must predate this period.

I don't think one should get too specific about concretely Brunei Malay
having been the precursor (at least, I never made that contention in such
a concrete form), only that several Malayic isolects / Malay dialects
along the sailing route, Brunei included, had this postglottalizing
feature.

The historiographic testimony of Pigafetta on Brunei suzerainty over Luzon
- who knows, whether long or short, real or purely formal - was just extra.
What we do know for sure, seems to be that there had been relatively
intensive lexical intake as a result of contacts with Malay speakers over a
long period of time. Part of that time, borrowings apparently did not
feature postglottalization of final vowels. And I merely wish to say, that
a certain contingent of the loans (or perhaps two or more such contingents)
apparently reflect contacts with speakers of Malay dialects with automatic
postglottalization of  final vowels (e.g. Brunei Malay, Banjarese). That
seems to be the simplest explanation. It would have been more difficult to
explain how they managed to ignore the final glottal in words they perhaps
borrowed from Malay dialects with automatic post-glottalization, seeing that
Malay dialects along the route tend to postglottalize......, or how they
might have managed to make so many Malay loans while carefully avoiding
borrowing from those dialects which automatically postglottalize.

I mean, once one allows for Tagalog having borrowed some of the stuff from
Malay dialects with postglottalization of final vowels, and these just
happen to be the most prominent dialects along the route, one must permit
at least part of loanwards with postglottalized final vowels to be of this
origin (unless we can account for loss of that final glottal at borrowing,
beside parallel appearance of glottal after final vowel in loans by some
other, yet unidentified mechanism).

There are other circumstances testifying to prominence of Brunei Malay.
Ambon Malay of the 17th century had suffix -akan for the -kan of
Standard Malay. Brunei Malay seems to be the only other Malay dialect with
-akan instead of -kan, suggesting that the Brunei area may have been the
place of origin of Ambon Malay (spoken in Central Maluku) or at least
played an important role in its emergence.

> Furthermore, the historic Brunei sultunate being a Muslim state, we should
> expect to find more Arabic loan words as we do in Tausug and other languages

If I remember correctly, there are relatively early mentions of Brunei
in Chinese sources as Boni or Buni, the earliest of which may pertain to
the pre-islamic period. But whatever the case, one shouldn't assume
Islamization of various polities to have happened as an abrupt cut in
the cultural history. The transition was usually very very gradual. The
word _dewata_ "(the) gods" (where _e_ is not schwa but mid front) exists
up to this day in Javanese (muslim since some 5 centuries) as well as in
Malay (muslim since even longer). The Malayan version of the Ramayana,
the Hindu epic of Valmiki, survived in several copies as Arabic-based
Jawi-script manuscripts with the title "Hikayat Seri Rama". The Minya
Tujuh inscription in Aceh, being a muslimic tombstone for the wife of a
sultan, is in the (Indic) Pallava-based Old-Sumatran script. Some
oldest Indonesian mosques (e.g. that of Demak in central Java) exhibit
unmistakable features of architecture of Hindu temples, whereas the old
mosque of Tidore, of which I once saw an engraving, even reflected older,
pre-Hinduistic sacral architecture (bipyramidal roof).

But we do find some Arabic/Persian loans, like that _salawAl_ (A = stressed
a) Jean-Paul refered to. It was first assigned by Bob Blust (1970, Proto-
Austronesian Addenda, Oceanic Linguistics 9:104-162) to a *saluqar (Ar.),
subsequently corrected to *saruqal (1974, The Proto-Austronesian Word for
'Two': a Second Look, Oceanic Lingusitics 13:123-161, see there p.142). Bob
Blust also cites Bikol _sarwAl_ from Malcolm Mintz's dictionary, besides
indicating that Wilkinson's Malay dictionary lists variants _seluar_,
_sarewal_, and _sarawal_ (_e_ is schwa in both instances). But Luzon had
never been directly muslim, the way the Sultanate of Sulu had been, which
would explain the greater number of ultimately Arabic loans in Tausug.

Regards to all,   Waruno

-----------------------------------------------------------------------
Waruno Mahdi                  tel:   +49 30 8413-5411
Faradayweg 4-6                fax:   +49 30 8413-3155
14195 Berlin                  email: mahdi at fhi-berlin.mpg.de
Germany                       WWW:   http://w3.rz-berlin.mpg.de/~wm/
-----------------------------------------------------------------------