Ergativity, or traces of it, in N.Am. pidgins? (fwd)

[David Robertson]

---------- Forwarded message ----------
[by the way, Henry, do you want me to subscribe you at the PSU address?
-- Dave]

Date: Tue, 14 Mar 2000 21:25:04 -0800 (PST)
From: Henry Zenk <psu18009 at pdx.edu>

Hi David,

[...]

I do have some thoughts about, if not ergativity in CJ exactly, at least,
of CHINOOKAN ergativity reflected in CJ.  These reflections are clearest
in the Grand Ronde var of the language (I know some people object to
calling CJ a "language"), with its clitic as well as independent forms of
pronouns.  Jacobs was aware of two pronoun sets in Victoria Howard's
CJ:  one which he felt was derived from Chinookan pronominal prefixes
(they consist of the Chinookan prefixes plus -a), the other from Chinookan
independent pronouns (yielding the regionally familiar CJ pronouns).  I am
aware of three:  the Chinookan prefix + -a set (na, ma, ya, nca, mca,
Las--that last has more than just an added -a), the Chinookan independent
set (nayka, mayka, ya(X)ka, ncayka, mcayka, Laska), and a set that appears
to consist of truncated versions of the latter (nay, may, ya, ncay, mcay,
Las) (I disregard variant forms such as nEsayka for ncayka, etc.).  Note
that the third person pronouns are identical for sets 1 and 3.  I should
also note that forms from all three sets are documented for most
speakers.  Most speakers tended to interchange pronouns from sets
two and three with corresponding pronouns in set one, with one
important restriction:  the forms na, ma, nca, mca NEVER appear
postposed (that is, as object forms, OR as subjects in Pred-S
constructions).  (I should also note that Victoria Howard was a
Chinookan speaker besides a CJ speaker, but that other,
non-Chinookan speaking Grand Ronde Indians also used clitic
pronouns.)

Anyway, about Chinookan ergativity.  I was struck when I read Boas's
Chinookan grammar by the "ergative" identity of form between the 3 sg (I
believe masculine) intransitive subject and the 3 sg transitive object
prefixes; the 3 sg masculine n/g prefixes match too (don't remember how
all the prefix sets compare on this point; the ergativity of Chinookan is
however widely noted).  Well, isn't there a hint at least of Chinookan
ergativity in the patterning of the CJ clitic vs. independent (including
truncated:  they're interchangeable with independent) sets?  The forms
na, ma etc. simply never occur as object forms, nor as subject forms
postposed to nouns and adjectives.  That is, they occur only as subject
forms going before the verb.  Independent (including truncated) forms, by
contrast, may indeed come post-verb:  either as objects of transitive
verbs, or as subjects in clauses with noun or adjective intransitive
verbs.

Dell, if you're out there, any thoughts?  Henry

On Sat, 13 Mar 1999, David Robertson wrote:

> LhaXayEm.
>
> This is a thought I've not had before; consequently, it's not well
> considered.  But perhaps you linguists will be interested in considering
> it.
>
> ChInuk Wawa largely derives from Chinookan languages.  (In fact it has
> been suggested that a label "Pidgin Chinook" would be more accurate.)
> Chinookan languages, as well as Salishan and many other languages of the
> Northwest --Wakashan as well?-- are ergative(-absolutive) in nature,
> rather than accusative(-nominative).
>
> Do you feel that ChInuk Wawa shows much trace of ergativity in its
> structure?  If so, does that reflect a preponderance of influence from
> indigenous languages over European language influence?
>
> To extend my question, does Mobilian Jargon, Delaware/Unami Jargon, or
> Eskimo Jargon show ergative features?  At least one of the named source
> languages for these ("Eskimo") certainly is ergative in its grammar.
>
> To offer a contrasting question, do pidgins tend toward accusativity at
> the expense of ergativity?  Do "word-order" languages, those which have
> little or no morphological case-marking, also tend toward accusativity?
> Would this tend to support an argument that languages of the world are
> underlyingly accusative?
>
> (I do not often think in terms of transformational grammar, let it be
> admitted.  Reading Silverstein's [1976] "Hierarchy of Features and
> Ergativity" is the immediate stimulus to these thoughts.  There he
> discusses the hypothesis that all languages are underlyingly accusative,
> and "that apparently ergative languages are really accusative languages
> with obligatory passive expression of transitive sentences."  [114]
> Another idea that I have in mind here is the notion that pidgins and
> creoles can show us something of the nature of human language and the
> acquisition thereof.  Also, some of Sally Thomason's work has used the
> idea that certain structural features of CJ may show us something about
> the degree of indigenous input into the formation of the language.  My
> question is inspired by each of these suggestions.)
>
> Ikta ma tEmtEm?
> Dave
>
>
>
>
>  *VISIT the archives of the CHINOOK jargon and the SALISHAN & neighboring*
> 		    <=== languages lists, on the Web! ===>
> 	   http://listserv.linguistlist.org/archives/salishan.html
> 	   http://listserv.linguistlist.org/archives/chinook.html
>