9.1268, Sum: Almost

[LINGUIST Network]

LINGUIST List:  Vol-9-1268. Mon Sep 14 1998. ISSN: 1068-4875.

Subject: 9.1268, Sum: Almost

Moderators: Anthony Rodrigues Aristar: Wayne State U. <aristar at linguistlist.org>
            Helen Dry: Eastern Michigan U. <hdry at linguistlist.org>

Review Editor: Andrew Carnie: U. of Arizona <carnie at linguistlist.org>

Associate Editors:  Ljuba Veselinova <ljuba at linguistlist.org>
                    Martin Jacobsen <marty at linguistlist.org>
                    Brett Churchill <brett at linguistlist.org>

Assistant Editors:  Elaine Halleck <elaine at linguistlist.org>
		    Julie Wilson <julie at linguistlist.org>

Software development: John H. Remmers <remmers at emunix.emich.edu>
                      Zhiping Zheng <zzheng at online.emich.edu>

Home Page:  http://linguistlist.org/


Editor for this issue: Brett Churchill <brett at linguistlist.org>

=================================Directory=================================

1)
Date:  Mon, 14 Sep 1998 15:56:58 +0900
From:  "Hiroaki Tanaka" <hiro-t at ias.tokushima-u.ac.jp>
Subject:  Summary: almost

-------------------------------- Message 1 -------------------------------

Date:  Mon, 14 Sep 1998 15:56:58 +0900
From:  "Hiroaki Tanaka" <hiro-t at ias.tokushima-u.ac.jp>
Subject:  Summary: almost

Dear all,
   I posted a query about implicature/entailment status
of the word "almost" called "approximates" at the end
of  July about two months ago. To this, 6 people
responded, whose names and e-mail addresses are
listed below. I want to thank all of them. Their answers
are of great help to me.

Partrick Griffiths <griffiths_p at usp.ac.fj>
Willaim morris <morris at ling.ucsd.edu>
Debra Ziegeler <DZIEGELE at vaxc.cc.monash.edu.au>
Rick Mc Callister <rmccalli at sunmuw1.MUW.Edu>
Bernard Kripkee <Bernard.Kripkee at InternetMCI.com>
Larry Horn <laurence.horn at yale.edu>
Keith J. Miller <keith at mitre.org>

Among them, Larry Horn introduced his and J. Atlas's
papers written about this matter, which are the following.

Horn, L. (1991) "Given as new: When redundant
affirmation isn't." Journal of Pragmatics 15: 305-328,
(part of which is a critique of the diagnostics in sadock
1979)
- -. (1996) "Exclusive company: Only and the Dynamics
of vertical inference." Journal of Semantics 13: 1-40,
(some discussion of 'almost' and 'barely')
Atlas, J. (to appear) "Negative adverbilas, prototypical
negation, and the De Morgan taxonomy." Journal of
Semantics 14(4).

   And Debra Ziegeler told me that she presented a
paper at the LACUS Forum 25, Claremont California (31
July 1998) and it will be published in the LACUS Forum
Proceedings No. 25. Please refer to her for her paper in
person.

   For those who missed my query in July (and because
it has been quite a long time since I posted, and I'm
sorry for the delay of the summary), I repeat my
question,  summarizing the result of the judgement test
and  some commnets of the respondents.

- -------------------------------------
I'm working on conversational implicature theory,
about which I am particulary interested in the word
"almost" called "approximates."  I wonder if
"almost P" conversationally implicates or entails
"not P." See the following examples.

     Bill almost swam the English Channel.-->Bill did not
swam the English Channel.
    Bill almost killed her.-->Bill did not kill her.

By (generalized) "conversational implicature", I mean "almost P"
implcates "not P" by/via Grice's Quantity maxim which
requires the speaker S to say as much as requried.
Because S doesn't say P, I infer that S doesn't say more
than is required and conclude that saying "almost P" P
does not hold in this case, hence "not P". By "entailment", I
mean  "almost P" is always true/holds wherever "not
P" is true/holds, but not vice versa. Bill's almost
crossing the channel always means his not crossing the
channel, but not vice versa.

   Some contraversial claims made in deciding whether "not P"
is implicated or entailed by "almost P" are that "almost" affects
(i)the acceptability of contradictory sentences rather than
"some"/"not all" contrast, (ii)the acceptability of "but not..."
phrase and (iii)the acceptability of "if not..." phrase. Please
check *, ? or OK in each slot below, which are mostly adapted
from Hitzman (1992) and Sadock (1981)(see the references below),
and please explain what the real meaning of each acceptable
sentence and the reason of the unaccaptable sentences.

- --------------------------------------
The result of the inqury.
(i)Contradictory sentences:
   (1) (* ) Mary is almost a cooporal and she's a corporal.
           OK: 0 people / ?: 0 people / *: 4 people
      P. Griffiths says that ALMOST entails NOT P
because he cannot accept
     *Bill almost swam the Channel and in fact he did
By contrast, he can accpet
     I thought I heard a nightngale and in fact there was
one.
Which means THOUGHT that P implicates NOT P, to
which I totally agree.

  (2) (  ?   ) There are two dogs in the yard and there are
three dogs in the yard.
     OK: 1 / ?: 2 / *: 1
All of the people regarded this sentence as having two
situations, one being the case that there are two dogs in
one part of the yard and the other being that there are
three in another part of the yard. In such
situation/context, 3 people accepted/put a ? in (2).
However, Hitzeman (1992) (see the references below)
accepted this sentence meaning that there is only one
situation and the speaker denies the conversational
implicature (not 3 dogs in the yard) in the latter part of
the sentence. I wonder if we can accpet the sentence
like this. Rather, if the _in fact_ sentence is not odd at
all, the sentence quite successfully cancels the
conversational implicature. What do you think?
   (   )  There are two dogs in the yard and in fact there are three dogs in
the (same) yard.

  (3) a.  A: John washed some of the windows yesterday,
           B: ( OK     ) I heard that he did wash all of the
windows yesterday.
   OK: 3 / ?: 1 / *: 0
Again this result contrasts with Hitzeman's judgement, in
which (3a) is unnatural, saying that "by claiming that
john is taking _none P_, Person A also claims that John is
taking _not some P_. Person B then may deny that John
is taking _not some P_ using a stressed auxiliary, saying
that it IS true that _some P_. In cases of conversational
implicature such as (3a) the aux cannot be inserted."
One person who is questionals about the acceptability of
(3a) says that (3a) would be better without an emphatic
"do" in B's response, which means he agree with
Hitzeman.

   Hitzeman  aslo says that unlike conversational
implicature, the entailemt case like "almost" allows
insertion of the stressed aux. like (3b).

     (3)  b. A: John almost finished his homework
yesterday.
          B: (  OK    ) He did finish it. I checked it this
morning.
        OK: 4 / ?: 0 / *:0

(3a) and (3b) contradict Hitzeman's judgement. I
conclude from this that there seems to be no justifiable
distinction between two denials of implicaure and
entailent at least in the case of (3a) and (3b).

(ii)Acceptability of "but not..." phrase. "His voices"
should have been "His voice" as most people pointed
out. This is adapted from a TV series "Columbo's"
noverization. Without some contexts it is meaningless.

  (4) a. ( OK    )His voice was almost, but not quite,
angry.
        OK: 4 / ?:0$B!!(B/ *:0

       b. (  ?*  )His voice was almost, and not quite, angry.
      OK: 0 / ?:3 / *:1

       c. ( ?*   )His voice was not quite, but almost, angry.
       OK: 1 / ?: 2 / *:1

       d. ( *   )His voice was not quite, and almost, angry.
       OK: 0 / ?: 0 / *:4

          e''. ( OK   )Bill ate some, but not all, of the cake.
      OK: 4 / ?: 0$B!!(B/ *:0

          e'''. (  OK   )Bill ate some, and not all, of the cake.
     OK: 3 / ?: 1 / *:0

          e''''. ( OK   )Bill ate not all, but some, of the cake.
     OK: 3 / ?: 1 /*:0

          e'''''. ( *   )Bill ate not all, and some, of the cake.
    OK: 0 / ?: 0 / *:4

Atlas (1981) says that the status of "almost" and
"some" is implicature because "almost but not quite"
parallel "some but not all" in (4a) and e''. However, the
actual parallelism is like the following.

    OK: almost but not quite
    OK: some but not all

   ?*: almost and not quite
   OK: some and not all

   ?*: not quite but almost
   OK: not all but some

   *: not quite and almost
   *: not all and some


(iii)Accaptability of "if not..."
   (5) a. (  OK    )Moore has almost begun to grasp the
concept, if he hasn't already grasped it.
   OK: 4 / ?: 0 / *:0

        b. ( OK?    )Moore has almost begun to grasp the
concept, if he hasn't quite grasped it.
   OK: 2 / ?: 1 / *:1

   (6) ( OK    )He hit almost forty home runs, if not forty.
  OK: 4 / ?: 0 / *:0

   (7) a. ( OK    )Frankenstein's monster was almost
human, if not human.
  OK: 4 / ?:0 / *:0

        b. ( OK?    )Frankenstein's monster was almost
human, if not quite human.
  OK: 2 / ?: 1 / *: 1

   "If not" can suspend the implicature of the weaker
word like "some", in which case the same holds for
"almost".

   I cannot reach the conclusion only from these
evidences that "almost P" entails "not P", although I
am convinced that it is time to restudy the possibility of
using "and", "but", and "if not" as an implicature
cancelling/suspending device.

References
Atlas, J.D. (1984) "Comparative adjectives and adverbials of
degree: an introduction to Radically Radical Pragmatics."
Linguistics and Philosophy 7: 347-377.

Hitzeman, J. (1992) "the selectional properties and entailement
of 'almost'." CLS 28: 225-238.

Sadock, J.M. (1981) "Almost." In Radical Pragmatics,
Academic Press, pp. 257-271.

   Thank you very much for all the replies to my query,
which are very insightful and helpful to me. Please don't
hesitate to continue any discussion via e-mail address
below.

Best wishes,

Hiroaki Tanaka

Associate Professor
Faculty of Integrated Arts and Sciences
Tokushima University, Japan

1-1, Minamijousanjioma,
Tokushima, 770,
Japan

phone & fax: +81 886 56 7125
e-mail: hiro-t at ias.tokushima-u.ac.jp




---------------------------------------------------------------------------
LINGUIST List: Vol-9-1268