Uto-Aztecan [was: What is an IE language]

At 12:10 AM -0500 12/20/02, Amanda Babcock wrote:
>On Thu, Dec 19, 2002 at 11:21:47AM +0100, Christophe Grandsire wrote:
>
>> As for IE features, as I said, they are very varied. But IE languages seem to
>> have always some things in common (note that the list doesn't apply to *all* IE
>> languages. Rather, those are features that are found in a majority of IE
>> languages each, but not all IE languages have each of them):
>
>Wow!  Such an impressive list!  I wish I could see something like that
>for some of the other language families in the world, especially the more
>obscure ones...
Here are some Uto-Aztecan (UA) traits, in no particular order. Many of these are
not unique to UA, but they are all characteristic of the family. Probably the
most unique are the absolutive suffixes and the number-sensitive verb stem
suppletion. (I gleaned most of this information from Ronald Langacker's _An
Overview of Uto-Aztecan Grammar_ and my own lecture notes for a course in the
structure of Uto-Aztecan I taught a few years ago at the University of Utah.)

* basic word order: UA shows all of the symptoms of being verb-final (GN, AN,
postpositions, etc), even in languages where canonical word order has shifted
the verb to initial position (Tepiman, Nahuatl). There is also a tendency for a
second-position clitic string containing elements of person and number of the
subject and tense and apsect for the sentence as a whole. (Of course, second
position phenomena are not confined to UA, but they are characteristic of it.)

* pronoun "copies": By this is meant the copying of nominal constituents in
pronominal form. There are four grammatical constructions involving copying: i)
possessives, ii) postpositions, iii) object agreement on the verb, and iv)
subject agreement on the verb or in the clitic group.

* non-distinct argument phenomena: PUA had three prefixes whose function was to
indicate unspecificity of the subject or object: *ta- indicated unspecified
subject, *t1- indicated unspecified object, and *n1- indicated unspecified
subject where this subject was animate and coreferential with the subject
(reflexive). Most UA languages have lost one or more of these prefixes, but
they are preserved intact in Shoshoni.

* syncretism of interrogative and indefinite pronouns: In UA, the indefinite
pronouns are not distinct from the interrogative pronouns. These pronouns
typically begin with /h/, and are thus "h-words" (cf English wh-words).

* negative imperatives: Negative imperatives in UA are not simply the regular
negation of positive imperatives but show a higher degree of nominalization or
subordination than do positive imperatives. While individual languages have
adopted distinct strategies, the most distinct tendency is for negative
imperatives to be marked through modal elements such as dubitative or irrealis.

* absolutive suffixes: The Uto-Aztecan absolutive is not a case suffix; it is a
suffix (or more typically a group of suffixes) which appears on a noun in
citation form but may drop when a noun is subjected to various morphological
processes, such as affixation (possession, postpositions), compounding, or
reduplication. (The characteristic -tl of Nahuatl is an absolutive with
allomorphs -tli and -lli.)

* case: When case marking is present, is nominative-accusative. There are no UA
languages that I know of which extend the case inventory beyond nominative,
accusative, genitive.

* number: A distinguishes between singular and plural. Some of the northern UA
languages (Numic, Tübatulabal, Hopi) have developed a dual, but this is not a
family trait. Plural marking tends to be restricted to animate nouns, and may
be marked by suffixation (*-m1) or reduplication or both.

* participles in *-t1 and *-p1: *-t1 marked the active participle in PUA, and
*-p1 marked the passive participle. Their reflexes still retain these functions
in the daughter languages; e.g. Shoshoni -t1n and -pp1h.

* incorporating verbs meaning "have", "make/do": UA languages often express
predicative possession ("I have a rock.") by means of a suffix which converts a
noun into a verb with the meaning "have NOUN". E.g. Shoshoni -pa'i:

Ne tempimpa'i
ne tempin -pa'i
1s rock   -have
'I have a rock.'

* stem suppletion: In almost every UA language, there are a few verbs which are
suppletive for number. Suppletion is sensitive to the number of the subject for
transitive verbs, and to the number of the direct object for transitive verbs.
Here are some examples from Shoshoni:

sing    dual        plural
kat1    y1k_wi      y1k_wikka  'sit'
w1n1    tacakkihka  topoihka   'lie'
paikka  was1        was1       'kill'
uttu    himi        himi       'give'
yaa     hima        hima       'take'

What makes this suppletion so unusual is that it follows an ergative/absolutive
pattern, although UA has a definite nominative/accusative case alignment.

* perfective and imperfective stems: There is a basic distinction in the verbal
morphology of every Uto-Aztecan language between stems which are "perfective"
and stems which are "imperfective". This distinction has been reinterpreted in
different ways in the daughter languages, but this basic distinction is
available everywhere.

Dirk
--
Dirk Elzinga                                               Dirk_Elzinga@byu.edu

"It is important not to let one's aesthetics interfere with the appreciation of
fact." - Stephen Anderson

Replies