The following is selected from the discussion on le on Chinese Teachers' Net. It consists of exchanges of opinions among Benjamin Ao, Harold C. Hill, and Patrick Edwin Moran
I've run into an interesting (and seemingly anomalous) situation in usage of
the "double le" pattern with a quantified object. My TA (she is a native
Chinese) and I agree that the sentence Ta1 chi1 le san1 wan3 fan4 le (He
has eaten three bowls of rice so far [and is still eating].) is correct.
On the other hand, this pattern is unacceptable to my TA when the word
zhi3 (only) [or jiu4] is inserted. She says that the "so far" idea here
would require that the sentence start out with something like Dao4 xian1zai:
dao4 xian4zai ta1 zhi3 chi1 le san1 wan3 fan4.
I'd be interested to know how others construe this, and whether a
statement can be made as to why. Thanks in advance.
---------------------------
Harold C. Hill
Professor of Chinese
Head, Department of East Asian Languages and Literatures
Washington and Lee University
Lexington, VA 24450-0303
U.S.A.
TEL.: (540) 463-8829
FAX: 540-463-8478
e-mail: hhill@liberty.uc.wlu.edu
_________________________________
moran@wfu.edu
http://www.wfu.edu/~moran
Patrick Edwin Moran
East Asian Languages and Literatures
Wake Forest University
Winston-Salem NC 27109
office phone (910) 759-4959
BAO@firstbyte.davd.com Fri Apr 12 15:09:26 1996
Received: from artemis.davd.com (artemis.davd.com [198.147.119.22]) by mail0.uts.ohio-state.edu (8.7.5/8.7.3) with SMTP id PAA28450 for ; Fri, 12 Apr 1996 15:09:20 -0400 (EDT)
From: BAO@firstbyte.davd.com
Received: from hermes.davd.com (hermes.davd.com [198.147.119.24]) by artemis.davd.com (8.6.5/8.6.5) with SMTP id LAA17911 for ; Fri, 12 Apr 1996 11:08:28 GMT
Received: from ccMail by hermes.davd.com
(IMA Internet Exchange v1.04) id 16e9c820; Fri, 12 Apr 96 11:10:10 -0700
Mime-Version: 1.0
Date: Fri, 12 Apr 1996 11:57:34 -0700
Message-ID: <16e9c820@hermes.davd.com>
Subject: Rule: Re: Double le
To: wenzhu@magnus.acs.ohio-state.edu
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Description: cc:Mail note part
X-status: N
Your TA's intuition is correct. As I see it, this is not an anomaly at
all. It has to do with what we call change of state. A sentence like
Ta1 chi1 le san1 wan3 fan4 is ambiguous: it may describe either a
static fact, or the arrival of a new state. With the addition of the
sentence-final le, the second interpretation is made explicit and
exclusive. On the other hand, the sentence Ta1 zhi3 chi1 le san1 wan3
fan4 is a statement about a static fact of the past. It is a state all
by itself, i.e. it is not developed from another state. To use the
jargon of possible world semantics, it's a statement that is true only
in the world the speaker refers to. Adding a sentence-final le would
suggest otherwise and therefore result in a semantic conflict, and
hence it's unacceptable. The same reason explains why we cannot say
He's hit the jackpot yesterday, because He hit the jackpot is true
only in the world of yesterday. However, notice that Ta1 zhi3 chi1
san1 wan3 fan4 le is perfectly okay. The conversational implicature of
this is he used to eat more than three bowls. Here, there IS a change
of state involved. Also correct is Ta1 chi1 san1 wan3 fan4 le, which
implies either he used to eat more or he used to eat less.
The following examples are a partial illustration of what we can and
cannot do with the sentence-final le:
Ta1 chi1 san1 wan3 fan4. (+stative, -stative)
Ta1 chi1 san1 wan3 fan4 le. (-stative)
Ta1 zhi3 chi1 san1 wan3 fan4. (+stative, -stative)
Ta1 zhi3 chi1 san1 wan3 fan4 le. (-stative)
Ta1 chi1 le san1 wan3 fan4. (+stative, -stative)
Ta1 chi1 le san1 wan3 fan4 le. (-stative)
Ta1 zhi3 chi1 le san1 wan3 fan4. (+stative)
*Ta1 zhi3 chi1 le san1 wan3 fan4 le. (+stative? -stative?)
Benjamin Ao
______________________________ Reply Separator
Subject: Re: Double le
Author: moran@wfu.edu (Patrick Moran) at Internet
Date: 28/2/96 17:33
Prof. Harold Hill asked about the sentence:
dao4 xian4zai ta1 zhi3 chi1 le san1 wan3 fan4.
Or maybe his native speaker/informant did not really lop off the final le,
so it would be:
dao4 xian4zai ta1 zhi3 chi1 le san1 wan3 fan4 le.
Let's look at what the original sentence _should_ mean (possible
contradictions or whatever included):
Ta1 zhi3 chi1le san1 wan3 fan4 le.
As of now, he has only eaten three bowls of rice.
My intuition says that "He has only eaten three bowls of rice" makes it
sound like there is an expectation that he should eat more. To me, the
verb-le answers not the question "Have you done it?" (which would get
answered in positive fashion with guo), but "Did you do (that which I
expect you to do) yet?" (Did you eat your spinach?)
If the above paragraph is correct, then there would be what Aquinas calls a
"conflict of notes" in the sentence in Chinese, because on the one hand you
are admiting that "he's still at it" and on the other hand you're _blaming_
him for not having done enough -- it sounds like it would be more fair and
reasonable to make that judgment of "only" after he actually finishes
however much he is going to eat. I think that is why the native speaker
feels the sentence is wrong.
If you say:
Dao4 xian4 zai4, ta1 zhi3 chi1le san1 wan3 fan4 le.
You are saying something like,
Up to now, he has only eaten three bowls of rice (and I know he is probably
going to continue to eat, but I'm concerned enough about the situation to
take note of it.)
Of if you insist on:
Dao4 xian4 zai4, ta1 zhi3 chi1le san1 wan3 fan4.
Then it simply becomes:
Up to now, he has only eaten one bowl of rice (and I guess he is finished,
but it does not seem to me that he has eaten enough).
But then I'm not a native speaker....
Pat
______________________________________________________________
>From BAO@firstbyte.davd.com Fri Apr 12 15:09:28 1996
Received: from artemis.davd.com (artemis.davd.com [198.147.119.22]) by mail0.uts.ohio-state.edu (8.7.5/8.7.3) with SMTP id PAA28449 for ; Fri, 12 Apr 1996 15:09:20 -0400 (EDT)
From: BAO@firstbyte.davd.com
Received: from hermes.davd.com (hermes.davd.com [198.147.119.24]) by artemis.davd.com (8.6.5/8.6.5) with SMTP id LAA17908 for ; Fri, 12 Apr 1996 11:08:27 GMT
Received: from ccMail by hermes.davd.com
(IMA Internet Exchange v1.04) id 16e9c800; Fri, 12 Apr 96 11:10:08 -0700
Mime-Version: 1.0
Date: Fri, 12 Apr 1996 11:57:34 -0700
Message-ID: <16e9c800@hermes.davd.com>
Subject: Rule: Re[2]: Double le
To: wenzhu@magnus.acs.ohio-state.edu
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Description: cc:Mail note part
X-status: N
The sentence-initial adverbial phrase is irrelevant, i.e.
dao4 xian4zai ta1 zhi3 chi1 le san1 wan3 fan4 le
is still ungrammatical.
Benjamin Ao
Separator _________________________________
Subject: Re[2]: Double le
Author: moran@wfu.edu (Patrick Moran) at Internet
Date: 29/2/96 16:37
Benjamin Ao writes:
> *Ta1 mei2 chi1 san1 wan3 fan4 le.
>
> The semantic scope of mei2, like that of the post-verbial le, is
> generally considered to be the verb. Why, then, is "Ta1 chi1 le san1
> wan3 fan4 le" okay but not "Ta1 mei2 chi1 san1 wan3 fan4 le"?
Consider:
At time 1: Ta1 you3 yi4 dian3 pang4.
and then a change,
At time 2: Ta1 shou4 le.
To be able to say "Ta1 mei2 chi1 san1 wan3 fan4 le,"
there would have to be weird world in which:
At time 1: Ta1 chi1le san1 wan3 fan4 (le).
and then a change,
At time 2: Ta1 mei2 chi1 san1 wan3 fan4 *le*.
But that's not like:
At time 1: Ta1 you3 wu3 shi2 kuai4 qian2.
and then a change,
At time 2: Ta1 mei2 you3 qian/ le.
Because you can spend the money but you cannot "go back in time" and make
it such that it is no longer true that yesterday (or whenever, as long as
we are talking about the same dinner) he did not eat those three bowls of
rice.
In practice, it's always worked well for me to give students a rule:
If nothing has happened, don't use le.
If he did not eat the three bowls of rice, then that event did not
transpire, so you can't add le. (That works for "Ta1 mei2 you3 chi1 *le*
fan4," too.)
Pat
moran@wfu.edu
http://www.wfu.edu/~moran
Patrick Edwin Moran
East Asian Languages and Literatures
Wake Forest University
Winston-Salem NC 27109
office phone (910) 759-4959
From BAO@firstbyte.davd.com Fri Apr 12 15:09:09 1996
Received: from artemis.davd.com by postbox.acs.ohio-state.edu (8.6.9/5.901231)
id PAA21461; Fri, 12 Apr 1996 15:09:07 -0400
From:
Received: from hermes.davd.com (hermes.davd.com [198.147.119.24]) by artemis.davd.com (8.6.5/8.6.5) with SMTP id LAA17895 for ; Fri, 12 Apr 1996 11:08:15 GMT
Received: from ccMail by hermes.davd.com
(IMA Internet Exchange v1.04) id 16e9c790; Fri, 12 Apr 96 11:10:01 -0700
Mime-Version: 1.0
Date: Fri, 12 Apr 1996 11:57:32 -0700
Message-ID: <16e9c790@hermes.davd.com>
Subject: Rule: Re[3]: Double le
To: wenzhu@magnus.acs.ohio-state.edu
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Description: cc:Mail note part
X-status: N
You got it, Pat! I think the explanation you offered is exactly right.
To follow up on this line of thinking, I suppose we may stretch our
imagination a little bit and consider a situation where you play a
videotape backward (like what they do on AFHV) to your class and then
you point to a frame prior to the one in which ta1 chi1 le san1 wan3
fan4 le and say (zhei4 huir) ta1 mei2 chi1 san1 wan3 fan4 le? This is
such a far-fetched situation I can't trust my intuition any more. I
wonder what my fellow native speakers think about this.
Benjamin Ao
______________________________ Reply
The problem Harold Hill pointed out, that one can't
use "zhi3, only" for double "le" construction,
is more general than double "le" construction.
Hill's pair of sentences are
OK: Ta1 chi1 le san1 wan3 fan4 le
* Ta1 zhi3 chi1 le san1 wan3 fan4 le
But note that "zhi3" is often imcompatible with
final "le" even without the other "le" after the verb:
OK: Ta1 chi1 le
* Ta1 zhi3 chi1 le
OK: Ta1 zhi3 chi1 le san1 wan3 fan4.
Also note the parallel situation in English:
OK: He has finished eating. (He has eaten)
* He only has finished eating. (He only has eaten)
OK: He only ate three bowls of rice.
I think the semantics of final "le" and "zhi3" are
not quite compatible. Final "le" covers the whole verb
phrase in it's scope, indicating that is what's accomplished,
while "zhi3" also takes the verb phrase in its scope,
in particular, focusing on the numerals if present,
indicating what falls short of the expectation.
The two purposes clashed so they are not used together.
______________________________ Reply Separator
>From BAO@firstbyte.davd.com Fri Apr 12 15:09:26 1996
Received: from artemis.davd.com (artemis.davd.com [198.147.119.22]) by mail0.uts.ohio-state.edu (8.7.5/8.7.3) with SMTP id PAA28437 for ; Fri, 12 Apr 1996 15:09:16 -0400 (EDT)
From: BAO@firstbyte.davd.com
Received: from hermes.davd.com (hermes.davd.com [198.147.119.24]) by artemis.davd.com (8.6.5/8.6.5) with SMTP id LAA17902 for ; Fri, 12 Apr 1996 11:08:23 GMT
Received: from ccMail by hermes.davd.com
(IMA Internet Exchange v1.04) id 16e9c7b0; Fri, 12 Apr 96 11:10:03 -0700
Mime-Version: 1.0
Date: Fri, 12 Apr 1996 11:57:33 -0700
Message-ID: <16e9c7b0@hermes.davd.com>
Subject: Rule: Re[2]: Double le
To: wenzhu@magnus.acs.ohio-state.edu
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Description: cc:Mail note part
X-status: N
Chilin brought up the interest issue of semantic scope, and I'd like
to follow up on that. I agree that the scope of the quantifier zhi3
seems to be the entire VP, but I'm not sure if the scope of le is the
VP or the S'. Whatever decision we make, we should be able to explain
why "Ta1 zhi3 chi1 san1 wan3 fan4 le" is okay but "Ta1 zhi3 chi3 le
san1 wan3 fan4 le" is not, since both contain zhi3 and le. Please also
notice the compatibility between sentence-final le and the negative
quantifiers bu and mei:
Ta1 bu4 chi1 san1 wan3 fan4.
Ta1 bu4 chi1 san1 wan3 fan4 le.
Ta1 mei2 chi1 san1 wan3 fan4.
*Ta1 mei2 chi1 san1 wan3 fan4 le.
The semantic scope of mei2, like that of the post-verbial le, is
generally considered to be the verb. Why, then, is "Ta1 chi1 le san1
wan3 fan4 le" okay but not "Ta1 mei2 chi1 san1 wan3 fan4 le"? It's
this kind of paradox that makes me think the real reason may have more
to do with time reference and change of state.
By the way, I think "He only has finished eating" is okay, though it's
probably better to say "He has only finished eating". The implication
is that there are some other things that he's expected to have
finished by now but still hasn't.
Benjamin
______________________________ Reply Separator
Back to