just a quick question

[R. Joe Campbell]

Fran,
   I *thought* that "case of the missing n's" should probably be resolved
in the direction of some phonetic deletion, but then I fussed and fidgeted
about the question of whether it might possibly have something to do with
whether plurality was being marked at the morphological level or not.
Leonel's comment on plural marking in the rest of the sentence pointed
away from that, though.
   Syllable-final devoicing was one of my first neat surprises in Nahuatl
-- when I made my first stumbling steps in Tepoztlan and Santa Catarina
(Morelos), the devoicing of /l/ left the /l/'s with a great deal of
friction and speakers of Spanish in Tepoztlan who knew little or no
Nahuatl had it in their speech too.  I recorded some Spanish narrative and
lent a tape of it to one of my graduate professors in Spanish linguistics
for his commentary.  He told me that my speaker "had a very curious /s/"
(in obvious reference to the voiceless /s/'s) and his credibility dropped
like a rock.
   Of course, /y/ and /w/ also devoice too, in "classical" and modern
dialects.  /y/ ---> [x] is the basis for an early "gee whiz" experience
(quipia, she takes care of it; oquipix, he took care of it [cl.], where
the /y/ fails to show up in the present tense;;; tlaoya, he shells [corn];
otlaox, she shelled [corn] (Hueyapan, Morelos).  Many modern dialects have
converted syllable-final /w/ (spelled 'uh') into [h], so for years I
doubted that I would ever hear a "real" voiceless /w/.  Then I went to
Ameyaltepec and heard it.

                       Ameyaltepec          Canoa

she leaves it            quicahua            quicahua
he left it              oquicauh            oquicah

   On the nasals, I think what happens in syllable-final position may be a
different mechanism.  Again, Tepoztlan and Santa Catarina have some
relevant data:

                         Tepoztlan           Santa Catarina

house                    calli               calli
houses                   caltin              calti~
                                             (where i~ represents
                                             nasalized 'i')

People (particularly a friend of mine) from Tepoztlan would laugh
scornfully about the ignorant n-dropping of the folks from Santa Catarina
-- nomas no la saben pronunciar!!

I think that the voicing in the  i~  continues throughout the vowel, so
the mechanism of "dropping" looks like what happened in French and
Portuguese.  --And even in Spanish.  I've heard [e~fermo] for "enfermo"
and [nara~ja] (excuse the 'j') for "naranja".  Since the consonant that
follows the nasal consonant is not a "stop" (complete closure) the
articulators anticipate that small opening and the velum drops early,
nasalizing the vowel.  The historical process is categorically clear in
Portuguese, where nasal vowels contrast with oral ones only before
fricatives (e.g. f s z, etc.) and in final position.  The nasal consonants
remained stable before stops (e.g. p t k b d g ch, etc.).

   It seems to me that beginning in the 16th century, we can see nasal
consonant dropping happening in three environments:
  1. before /w y/
  2. word final
  3. before fricative consonants (i.e. /s x/)

I have assumed that 1 was a strong catalyst of /n/ deletion and that 3 was
somewhat weaker -- and I have no strong intuition about 2.

   nantli        mother
   nanyotl       motherliness
   nayotl        motherliness

    I know that we're not supposed to be subjective about the object of
our analysis, but I really *like* n-dropping.  San Agustin Oapan has it,
resulting in:

   patlani       it flies
  opatla         it flew
  opatlanqueh    they flew

-- resulting in a little more depth in the phonology of the language --
and it can use it!

..and I didn't refer once to any PG13 topics....

Best regards,

Joe


On Wed, 27 Oct 1999, Frances Karttunen wrote:

> One possible explanation is the invisibility of syllable-final nasal
> consonants in written Nahuatl.  Orthographically they are often represented
> by a tilde or bar over the preceding vowel.  Or they are simply left out.
> Phonetically there seems to be a reason for this.  Syllable-final resonants
> [n,m,l] are "devoiced" in Nahuatl.  That is, they are whispered.  So they
> are hard to hear.  Hence, they get left out of writing.  Speakers in some
> sence "know they are there."  It's only for nonspeakers that the writing
> system is confusing.
>
> >>(...) oncan quicenquixtiaya in tlatlaloque:
> >>auh in ihcuac oquicenquixtique, niman oncan quinmictiaya:
> >>auh in ihcuac in oquimonmictique, niman (...)
> >>( there they gathered them together [the sacrificial victims
> >>called] tlalocs: and when they had gathered them together,
> >>then they slew them there: and when they had slain them,
> >>then etc.)
> >>
> >>Would someone be kind enough to explain to me why is it
> >>found twice qui- instead of the expected quin-/quim- ,
> >>namely 'quicenquixtiaya' (they gathered them together) and
> >>'oquicenquixtique' (they had gathered them together)? After all
> >>"tlatlaloque" is plural isn't it? Are those mere 'typos'
> >>or is there an explanation for them that I can't see? And if so,
> >>why then the "correct" quinmictiaya and oquimonmictique?