<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta http-equiv="Content-Type" content="text/html;charset=ISO-8859-1">
<title></title>
</head>
<body text="#000000" bgcolor="#ffffff">
Chickasaw (and most other Muskogean) also has two 'eat' verbs,
intransitive impa and transitive apa. (These look similar, but there is
no regular relationship between them.) The first has one arguments, the
second two. It's interesting how common this is!<br>
<br>
(I'm not completely sure what you mean by a "covert object", David. Is
this an object that can appear if the speaker wishes to specify it? A
non-agreeing object, like the patient of 'give'? Or a semantically
implied object, if I can use that term, like what is eaten with English
'dine'?)<br>
<br>
Pam<br>
<br>
david costa wrote:<br>
<blockquote type="cite"
cite="mid24638163.1112496627554.JavaMail.root@scooter.psp.pas.earthlink.net">
<pre wrap="">In Algonquian, the transitivity of 'eat' is very simple -- there is a stem that
takes inanimate objects ('eat it'), another stem that takes animate objects ('eat
him'), and yet another stem that's intransitive (plain 'eat', 'dine'). Moreover, all
three stems are suppletive.
Dave C
-----Original Message-----
From: ROOD DAVID S <a class="moz-txt-link-rfc2396E" href="mailto:rood@spot.Colorado.EDU"><rood@spot.Colorado.EDU></a>
Sent: Apr 2, 2005 6:30 PM
To: <a class="moz-txt-link-abbreviated" href="mailto:siouan@lists.colorado.edu">siouan@lists.colorado.edu</a>
Subject: transitivity of eat??
Alfred,
In Wichita there are two verbs to eat, one used when there is an
object (ka'ac), and one when there is none (wa:wa'a). When you call
people to dinner, you say "We're going to eat now" (ke'ecira:kwa:wa'a)
with the intransitive verb.
An English verb with a similar argument structure is 'to dine'.
Do you think that, too, has a covert object???
I do not think that 'eat' in English always implies an object.
David
David S. Rood
Dept. of Linguistics
Univ. of Colorado
295 UCB
Boulder, CO 80309-0295
USA
<a class="moz-txt-link-abbreviated" href="mailto:rood@colorado.edu">rood@colorado.edu</a>
On Sat, 2 Apr 2005, [ISO-8859-1] "Alfred W. Tüting" wrote:
</pre>
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">I run into similar problems when people claim that "eat" in "we
</pre>
</blockquote>
<pre wrap="">eat every afternoon at 4:00" has an "implied object" because you have to
eat something. I think it is purely intransitive in that kind of context,
and has only one argument, logic or no logic. <<
Yes, "logic or no logic" ;-) I assume that you refer to English (or most
Indo-European) grammar, otherwise this would be a somewhat biased view
on the matter. (I'm unsure whether or not this distinction of "to eat"
being two verbs, one transitive and one intransitive, is nothing but
kind of psycho-linguistic speculation.)
In Lojban (le logji bangu = the logic language) e.g. the "selbri" (say,
predicate) "to eat" is defined as:
eats citka (cti): x1 |/ingests/consumes (transitive verb) x2
which expands to zo'e (cu) citka zo'e (zo'e=the obvious value of an
indefinite unspecified sumti, say, argument).
So one is free to express:
citka - smb (obvious/unexpressed) eats smth/smb(!) obvious/unexpressed
(the second zo'e-slot actually can also refer to a human e.g. in some
cannibalistic invironment ;-) )
mi pu citka [zo'e] - I ate (smth. unspecified)
le mlatu cu citka loi ratcu - the cat(s) eat(s) mice
[zo'e] na mu'o citka le sanmi - smb. unspecified doesn't/didn't/will not
eat up the meal
I.e. in Lojban - logically/grammatically - there are always all
arguments present (although maybe unexpressed/unspecified) that are
defined as pertaining to the selbri (predicate). I tend to assume that
this actually reflects "nature" (space and time human utterances perform
in).
In Dakota language (grammar)also, parts of speech (verbs=one-word
sentences), if transitive (I'd say by their "nature"), always point to
(a) definite participant(s) (albeit unspecified and understood by
context). So sometimes there's need to make them generic (by affixation
of _wa-_, which maybe might derive from _wan_?).
In Hungarian, it is kind of the other way around: any verb able to be
transitive by "nature" has a basic form that is generic and special
endings to make it specific.
E.g.
Szeretek könyveket - I love books (generic direct object)
Szeretem a könyveket - I love the books (specific direct object)
Olvasok könyvet - I read a book (generic direct object)
Olvasom (a) könyveimet - I read my books (specific direct object)
Only 3rd person pronouns are regarded as specific, but not 1st p.p.:
Szereted õt/õket - You love him, her, it/them
Szeretsz engem(et) - You love me
So, the well-known question is usually expressed simply by "Szeretsz?"
with the direct object implied/understood: Do you love me? (And the
expected answer has a special form for 2nd p s/pl: "Szeretlek
(téged/titeket)!"
(These implied objects also work with "datives": e.g. Nekem hiányoz -
to-me s/he, it-lacks -> I miss him/her/it, but also simply: "Hiányzol!"
- you-lack (to-me) -> I miss you!)
Chinese "to eat" (chi) actually is transitive and needs to have a direct
object: not unlike in Dakota, kind of generic object has to go with the
werb, i.e. "chi fàn" (lit.: to eat "rice", rice=generic word for food).
So, my conclusion might be that at least it depends on each language's
grammar - and, nontheless, I'm quite hesitant with regard to "eat" in
"we eat every afternoon at 4:00" not having an "implied direct object". :((
Alfred
</pre>
</blockquote>
<pre wrap=""><!---->
</pre>
</blockquote>
<br>
<pre class="moz-signature" cols="72">--
Pamela Munro,
Professor, Linguistics, UCLA
UCLA Box 951543
Los Angeles, CA 90095-1543
<a class="moz-txt-link-freetext" href="http://www.linguistics.ucla.edu/people/munro/munro.htm">http://www.linguistics.ucla.edu/people/munro/munro.htm</a>
</pre>
</body>
</html>