Refining early Basque criteria

Stefan Georg Georg at home.ivm.de
Thu Jan 20 09:06:24 UTC 2000


>It is my understanding that oral consonants are articulated by children
>before nasal consonants. Is this incorrect in your view?

Yes, this is incorrect in my view. You will not repeat that keeping your
moputh shut and simply turn on your vocal cords ("/mmm/") is such a
difficult thing that it invariable comes after, say, /p/, /ts/ or /r/, will
you ? A different question is, however, whether this very simple, let's
call it an "articulatory gesture", is really a "consonant". It might be
sensible to reserve this technical term for non-vocalic articulatory
gestures *when they form part of linguistically meaningful items*, which is
certainly not the case for infants who first try out their articulatory
tracts.

So, again, the nasal /m/ is articulated *very* early, and it's association
with whatever semantic content - done by the parents, of course - is an
artefact of adults fancying that the child is actually speaking to them. At
this stage, the child is certainly *not* "using" "language"; it's babbling.
Babbling-research has established that, in phonetic terms, the sounds
produced by infants vary greatly in frequency: at the top of the list are
/m/ and /b/ alike, with the same frequency, followed by, in this order,
/p/, /d/, /h/, /n/, /t/, /g/, /k/, /j/, /w/, /s/, aso. (Locke, J.L.:
Phonological Acquisition and Change, NY: AP 1983).

>All you have written is, of course, very plausible.

Yes, it very obviously is ;-)

>But --- and there is always a but, is there not? --- this does not really
>address my argument, I do not think.

>I am *not* maintaining that *mama is the early term for 'mother' but rather
>*ama.

Some societies conventionalize the fully reduplicated form /mama/, some
don't. Some languages do know initial labials, some don't (and among these,
some may allow for such a thing in a marked nursery term, some won't). Some
cultures chose to conventionalize /deda/ for "mother", reserving /mama/ for
"father". Some will conventionalize a one-syllable /ma/, some won't. When
it comes to the act of conventionalizing, which is to be kept sharply apart
from the "pre-speech-act" of babbling, the phonotactic rules of the
language the poor child is about to learn from now on, do play a role
(after all, the happy parents of a Tlinkit child do of course think that
their napper is talking Tlinkit to them, and not Abkhaz, so even when the
little one constantly babbles /mama/ they'll assume for convenience' sake
that it actually wanted to say /ama/ [for argument's sake I assume that
Tlinkit is one of the labialless languages of the American NW,. and that
/ama/ is "mother" there; if this is not the case, take any language, where
it is, and replace this example]). There may be other reasons, apart from
phonotactics, and in some individual cases they may be hard to tell. So
what ?

>Larry brushes by this difference but I am hoping you will not.

This is not brushing by. The word is "linguistics" (;-).

>While *mama may be a perfectly plausible child's word for 'mother', the
>frequently found form *?ama can, with an implausible level of difficulty, be
>derived from it. Do you not agree?

Of course, such words may change, once they have become conventionalized
items of a language, they may be subject to any systematic change of the
sound structure of this language. They may, however, and this is the point,
why such terms shouldn't be allowed to play a decisive role in any
classificatory work, escape this, either remaining stable in spite of
sound-changes in other parts of the lexicon, or be innovated over and over
again to maintain their emotional connotations (or, for that matter, in the
case of imitative words, their iconicity).

St.G.

PS: Shoot, "mother" is /tla/ in Tlinkit, so think up a different example ...

Dr. Stefan Georg
Heerstraße 7
D-53111 Bonn
FRG
Tel./Fax +49-228-691332



More information about the Indo-european mailing list