NEG Raising

Christopher T Collins cc116 at NYU.EDU
Mon May 20 13:15:11 UTC 2013


Dear Typologists,

A view perhaps related to the one sketched by Paolo Ramat is that there is
no syntactic NEG Raising,
but rather via some pragmatic principles, one gets the downstairs reading
of negation
(this is the claim of Bartsch 1973). The reasoning goes as follows:

Step 1:
"think" is a verb that has an Excluded Middle presupposition.
I think p or I think  not p.

Step 2:
"I don't think p" is never derived by NEG Raising, and the negation
only has matrix scope.

Step 3:
However, "I don't think p" plus the Excluded Middle presupposition
entails "I think not p".

But from a typological perspective, I think we can step back from the
proposed
solutions (syntactic or pragmatic) and simply ask which languages display
the kinds of data that would be labelled (pre-theoretically) NEG Raising.

I think it would be very interesting to have a relatively small set of
criteria
and apply it to a large number of languages to see what the distribution
of such languages is.

Chris



On Mon, May 20, 2013 at 6:06 AM, Paolo Ramat <paoram at unipv.it> wrote:

>   Dear All,
> I sympathize with Bjoern views: how can we  be sure that “I don't think
> that Floyd hit Roscoe” is derived via NEG-raising from  “I think that Floyd
> didn't hit Roscoe”?  Both sentences refer to the same  state of affairs,
> but the speaker *makes use* of different  focussing  strategies. Unless
> we accept the so-called Universal Base Hypothesis and the Linear
> Correspondence  Axiom of Kayne 1994, the assumption that NEG has  been
> raised in the second sentence is not well-founded.  As Haberland rightly
> states, “I hope that he is not a werewolf” is certainly not equivalent to
> “I don't hope that he is a werewolf” –but ‘hope’ is not an epistemic verb
> and the NEG refers primarily to ‘hope’: no need to speak of  a raising of
> ‘not’. My (strong and, may be, not very popular) conclusion: NEG-raising
> is an a priori assumption, as the other ‘alpha-movements’. An usage-based
> approach to linguistic facts needs not this kind of assumptions.
>
> Best,
> Paolo Ramat
>
>
>
>
>  *From:* Bjoern Wiemer <wiemerb at UNI-MAINZ.DE>
> *Sent:* Friday, May 17, 2013 12:24 PM
> *To:* LINGTYP at LISTSERV.LINGUISTLIST.ORG
> *Subject:* Re: NEG Raising
>
>  Dear Christopher (and all),
> thanks for raising the issue of NEG-raising! In light of your mail and the
> response by David Gil I would like to make two comments and one request on
> what you two observed.
>     In his classical paper on complementation Noonan (2007 [1985]: 100f.)
> gave examples with NEG-raising in English with the CTP-verbs "think,
> believe, want". He gave the examples cited below, the first pair coincides
> with yours. Now, in a footnote (f. 21) Noonan remarked that sentence (214b)
> is ambiguous, since it allows for both a reading with and without
> NEG-raising (i.e. in the latter case there would only negation of the CTP).
> He added that this ambiguity might be conditioned by "a
> commitment/non-commitment interpretation of the speaker's evaluation of the
> complement proposition" (with further references).
>
> (214a)    I think that Floyd didn't hit Roscoe.
> (214b)    I don't think that Floyd hit Roscoe.
>
> (215a)    Zeke believes that Martians don't live in caves.
> (215b)    Zeke doesn't believe that Martians live in caves.
>
> (216a)    Hugh wants Mary Ann not to win.
> (216b)    Hugh doesn't want Mary Ann to win.
>
> David Gil wrote that "'I don't think John is a werewolf' cannot mean 'I
> think that John is not a werewolf'". Would other native speakers judge the
> same way?
>
> Thus, two questions arise (in my view). First, wouldn't this judgment
> depend on how much commitment you ascribe to your epistemic attitude toward
> the proposition in the complement? Second, do such ambiguities (and
> possible differences in judgments between native speakers of the same
> language) show up with other verbs of the same conceptual domain (epistemic
> attitude, report on speech acts, volition, etc.)? Noonan made his remark
> quoted above only with respect to "think", "believe" seems to behave
> differently. What about other verbs denoting epistemic attitudes in English?
>     From this my request arises: Has anybody  worked on such ambiguities
> and tried to make up a classification of CTP-verbs (of epistemic attitude,
> volitional, etc.) within ONE language, and be it English. That is to say:
> apart from _cross_linguistic variation with respect to the liability toward
> complementation in general (and the way complementation is marked
> syntactically or by lexical means), it would be interesting to understand
> whether predicates denoting epistemic attitudes show variation within even
> one language, and what are the conditions.
>     I would be ready to collect such information and make a small digest
> out of it, if anybody sends me pertinent references or reports. Anyway, I'd
> be grateful to know more about this issue.
>
> Best regards,
> Björn Wiemer.
>
>
> Dear Typologists,
>
> Could you tell me if there are languages that you know or know of that do
> not permit NEG Raising.
> On a NEG Raising reading of (a), it is felt to mean the same thing as (b):
>
> a. I don't think John is a werewolf
> b. I think that John is not a werewolf
>
> Also, strict NPIs are licensed:
>
> c. John won't be here until 6:00
> d. I don't think John will be here until 6:00
>
> In these sentences 6:00 is a strict NPI, and it needs a negation.
> (d) contrasts with (f):
>
> e. I regret that John won't be here until 6:00
> f. *I don't regret John will be here until 6:00
>
> Other NEG Raising predicates include: think, believe, imagine, intend,
> want.
>
> Chris Collins
>
>
>
> --
> Björn Wiemer
> Professor für Slavische Sprachwissenschaft
> Johannes-Gutenberg-Universität
> Institut für Slavistik
> Jakob-Welder-Weg 18
> D- 55099 Mainz
> tel. ++49/ 6131/ 39 -22186
> fax ++49/ 6131/ 39 -24709
> e-mail: wiemerb at uni-mainz.dehttp://www.staff.uni-mainz.de/wiemerb/
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20130520/f0307426/attachment.htm>


More information about the Lingtyp mailing list