Starting to learn Nahuatl

Campbell, R Joe campbel at
Mon Aug 27 17:03:00 UTC 2007


   Another member of the Nahuat-l list and I have corresponded about
the Andrews grammar and he suggested that I contribute some of my
observations to the list.  This is an attempt to communicate some of
my disorganized notes and worries to the group and encourage
discussion of the issues.

   I would preface these notes with a few general statements:

1)   Anyone who is seriously interested in Nahuatl should read Fran's
    review article on twelve publications (including Andrews'
    Frances Karttunen, "Nahuatl for the Twenty-First Century",
    Ethnohistory, Vol. 52, No. 2, pp. 449-477 (2005).

     A PDF version is available for download:

2)   In spite of any worries and criticisms that I have about his
    grammars (the first edition and the considerably revised edition),
    I admire his work and feel fortunate to have gained from using it.

3.   Most of comments pertain to my own interest in Nahuatl, focused
    of the details of word formation.  As a "grammarian" I hope that
    our statements about morphology will enlighten people about
    the way the language "really" works; I do not consider our job to be
    making arbitrary hypotheses about how the language "might"
    work.  I think that our statements about grammatical mechanisms
    should be backed up by some kind of evidence.
      About forty years ago, one of my iconic experiences was hearing
    about one of my respected colleagues teaching a linguistics class;
    his reponse to a graduate student's solution to a problem was "That
    has     nothing to say for it except for the fact that it *works*."
    He was making the point that our solutions aim at some degree
    of "truth", not at being just *one* of the candidates of
    possible answers.

* * * * * * * * *

Some specific quibbles (all page numbers refer to the revised edition):
I will enclose quotes from Andrews' grammar in ((...)).

* * * * * * * * *

p. 574.

  ((ohhuia: to be like a road, i.e., to pose a danger, to be

It might be that danger and road are related concepts in Nahuatl,
but there is no reason given to believe this.  Morphological
relationships rest mainly on two kinds of evidence: 1) form (i.e.,
the phonological shape, and 2) function (i.e., semantic content,

* * * * * * * * *

p. 189.

((Among the destockal verbstems with  ni  as the stem formative are
two that are no longer extant but serve as the source for derived
stems (see #27.4.3):

*po-o:-ni > *(po:-ni) = to emit smoke

*to-o:-ni > *(to:-ni) = to run))

[[referring, respectively, to  popo:ca  and  toto:ca]]

This is based on the behavior of verbs like  pozo:ni  and patla:ni
that have intransitive reduplicated forms (in addition to their
normal ones) of:  popozo:ca  and  papatla:ca.
While one certainly wants to attend to possibly relevant data such
as the mechanisms involved in these latter verbs, it is *not* the
case that any two identical syllables in sequence are necessarily to
be interpreted as due to the process of reduplication.  It is
entirely possible that  popo:ca is either 1) lexically just what it
appears to be (i.e.  popo:ca) or 2) a reduplication of  po:ca.

* * * * * * * * *

p. 436

(under the topic: particle-look NNCs)

((cue:l ... it is in the manner of a crease; i.e., briefly suddenly,

cue:lihui = to become bent/folded))

I believe that semantic relationships may be hard to see because
there may be a succession of "bridges" or because one's culture and
the semantics of his language is so different from the language
being analyzed, but, in the absence of some explicit rationale as to
why "crease" and "suddenly" are might be related, this claim leaves
me dubious.  I don't send my bank account number to someone in
Canada or Germany who promises to reward me with millions of dollars
from unclaimed funds in a Nigerian bank account.

* * * * * * * * *

p. 441

((ichtaca (steal)  <  ich-tequi
    in the manner of one who cuts maguey leaves))

Aside from the unaccounted for difference between -taca and the
well-known verb  tequi, is there some cultural evidence to lead one
to believe that the language codifies a relationship between theft
of maguey leaves and stealing?

* * * * * * * * *

p. 283

((unique noun stems as fillers

*(te:l)-li + (po:ch)-tli-, "incense smoke" = (te:l-po:ch)-tli-,
  "a youth, young man"))

If one were looking for an example to teach morphology, this one
would be a highly ranked candidate for what to reject.  One of the
elements is unattested with no semantic connection; the other
element matches the *form* for "incense smoke", but has no semantic
connection to the notion of "youth".

* * * * * * * * *

I will have some more to add, but getting it ready will take time... |8-)



Nahuatl mailing list
Nahuatl at

More information about the Nahuat-l mailing list