[HPSG-L] Highly ambiguous sentences with lots of quantifier scoipings and psycholinguistic work on underspeification
Stefan Müller
St.Mueller at hu-berlin.de
Tue Mar 11 11:19:53 UTC 2025
Thanks Emily,
Wow! 10^12 readings! This is indeed impressive. But the sentence is not:
Myrdal is the mountain terminus of the Flåm rail line (or Flåmsbana)
which makes its way down the lovely Flåm Valley (Flåmsdalen) to its
sea-level terminus at Flåm.
I guess a lot of the ambiguity is caused by using generalized
quantifiers for proper names and they should not scope with respect to
each other. So it is an engineering problem, although an interesting
one. Utool has some trick to cope with such unwanted ambiguities. There
is some way to specify equivalences. I forgot how it works but want to
find out.
So what I am looking for is something like:
Every daughter of an employee with at least four children knows somebody
with a daughter who owns a pony.
But naturally occurring and with lots of scopings. It would be fine if
it were only a million readings. =:-)
Something that convinces semanticsist or at least students that
underspecification is a good thing to have. =:-)
Best
Stefan
Am 10.03.25 um 15:35 schrieb Emily M. Bender:
> Hi Stefan, Hi all,
>
> Regarding computational processing of scope ambiguities I think you're
> thinking of the work of Koller & Thater on Utool:
>
> https://aclanthology.org/P05-3003.pdf
>
> Those impressive numbers involved the ERG, not the Grammar Matrix nor
> small grammars derived from it. (Though in principle, Utool should be
> compatible with Matrix-derived grammars.)
>
> Emily
>
> On Mon, Mar 10, 2025 at 6:46 AM Stefan Müller
> <St.Mueller at hu-berlin.de> wrote:
>
> Yes, I expected that this is true and that people will not sit
> there and
> imagine all the 20.000 possible worlds that correspond to the
> readings.
> But what do I know? I am not a psycholinguist.
>
> And after all our world turned into an impossible world due to
> unpredictable behavior of my fellow humans. By now I expect them
> to do
> strange things.
>
> Thanks for all the answers! This is enourmously helpful.
>
> I will post a summary.
>
> Best
>
> Stefan
>
> Am 10.03.25 um 14:35 schrieb Rui Chaves:
> >
> > There’s also the anecdotal fact that people who read a 80-word
> > sentence in a newspaper don’t spend 30 minutes parsing it.
> >
> > 😊
> >
> > --
> > Rui P. Chaves (he/him)
> > Chair and Professor of Linguistics, University at Buffalo, SUNY
> > Director of Graduate and Undergraduate Studies for the
> Computational
> > Linguistics Program
> >
> https://urldefense.com/v3/__https://www.acsu.buffalo.edu/*rchaves/__;fg!!K-Hz7m0Vt54!mPAi5C1HHo3E9FbgFknLV1atAM-iwf0T-92yDt4r8e8_3IBw6V8u0JA2FXttB3auQ2jHnpfjZ6sxcWf8CzIEY6M$
>
> >
> > *From: *Rui Chaves <rchaves at buffalo.edu>
> > *Date: *Monday, March 10, 2025 at 8:33 AM
> > *To: *Roussanka Loukanova <rloukanova at gmail.com>, Stefan Müller
> > <St.Mueller at hu-berlin.de>
> > *Cc: *hpsg-l at listserv.linguistlist.org
> <HPSG-L at listserv.linguistlist.org>
> > *Subject: *Re: [HPSG-L] Highly ambiguous sentences with lots of
> > quantifier scoipings and psycholinguistic work on underspeification
> >
> > Dear Stefan,
> >
> > There is evidence that comprehenders don’t compute all scopings:
> >
> >
> https://urldefense.com/v3/__https://journals.plos.org/plosone/article?id=10.1371*journal.pone.0081461__;Lw!!K-Hz7m0Vt54!mPAi5C1HHo3E9FbgFknLV1atAM-iwf0T-92yDt4r8e8_3IBw6V8u0JA2FXttB3auQ2jHnpfjZ6sxcWf88ly5M3E$
>
> >
> > --
> >
> > Rui P. Chaves (he/him)
> > Chair and Professor of Linguistics
> > Director of Undergraduate and Graduate Studies for the
> Computational
> > Linguistics BS and MS Programs
> >
> > University at Buffalo, SUNY
> >
> >
> https://urldefense.com/v3/__https://www.acsu.buffalo.edu/*rchaves/__;fg!!K-Hz7m0Vt54!mPAi5C1HHo3E9FbgFknLV1atAM-iwf0T-92yDt4r8e8_3IBw6V8u0JA2FXttB3auQ2jHnpfjZ6sxcWf8CzIEY6M$
>
> >
> <https://urldefense.com/v3/__https://www.acsu.buffalo.edu/*rchaves/__;fg!!K-Hz7m0Vt54!mPAi5C1HHo3E9FbgFknLV1atAM-iwf0T-92yDt4r8e8_3IBw6V8u0JA2FXttB3auQ2jHnpfjZ6sxcWf8CzIEY6M$
> >
> >
> > *From: *HPSG-L <hpsg-l-bounces at listserv.linguistlist.org> on
> behalf of
> > Roussanka Loukanova <rloukanova at gmail.com>
> > *Date: *Monday, March 10, 2025 at 8:25 AM
> > *To: *Stefan Müller <St.Mueller at hu-berlin.de>
> > *Cc: *hpsg-l at listserv.linguistlist.org
> <HPSG-L at listserv.linguistlist.org>
> > *Subject: *Re: [HPSG-L] Highly ambiguous sentences with lots of
> > quantifier scoipings and psycholinguistic work on underspeification
> >
> > Hi Stefan,
> > Hi Everyone,
> >
> > Your questions are extra interesting. I'm very much interested
> in the
> > answers.
> >
> > And, of course, I'm looking forward to your revised HPSG book!
> >
> > I work on Type Theory of Recursion, including its varieties of
> Type Theory
> > of Acyclic Recursion / Algorithms (TTAR / TTAA). TTAR provides
> > mathematical and algorithmic foundations of Syntax-Semantics
> (SynSem) in
> > Computational Grammars (CompGrs).
> >
> > The relational extensions of TTAA to Dependent-Type Theory of
> Situated
> > Info
> > (DTTSI) directly provides math, i.e., algorithmic foundations,
> of MRS and
> > its SynSem in HPSG.
> >
> > In all the versions of TTAR / TTAA and DTTSI, I include
> > underspecification.
> >
> > Best Regards,
> > Roussanka
> >
> >
> > On Mon, 10 Mar 2025 at 12:22, Stefan Müller
> <St.Mueller at hu-berlin.de>
> > wrote:
> >
> > > Hi,
> > >
> > > I remember that people working with the grammar matrix and
> large scale
> > > implementations reported natural occurring sentence with an
> enormous
> > > amount of readings. Something in the range of 100.000
> scopings. Is this
> > > documented somewhere in print?
> > >
> > > Is there psycholinguistic research showing that humans do not
> work with
> > > specific readings but leave scopings underspecified (maybe to
> a certain
> > > extend).
> > >
> > > I am revising my HPSG textbook and switching to MRS in this
> book now, so
> > > I am interested in this kind of information.
> > >
> > > Thanks a lot.
> > >
> > > Best wishes
> > >
> > > Stefan
> > >
> > > _______________________________________________
> > > HPSG-L mailing list
> > > HPSG-L at listserv.linguistlist.org
> > >
> >
> https://urldefense.com/v3/__https://nam12.safelinks.protection.outlook.com/?url=https*3A*2F*2Flistserv.linguistlist.org*2Fcgi-bin*2Fmailman*2Flistinfo*2Fhpsg-l&data=05*7C02*7Crchaves*40buffalo.edu*7Ca8b9b7346347478f06f008dd5fceb14c*7C96464a8af8ed40b199e25f6b50a20250*7C0*7C0*7C638772063545337622*7CUnknown*7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ*3D*3D*7C0*7C*7C*7C&sdata=4CDFd3jV49zrGkDtdZH0kdv5JcYC3egTLAtU5GxsPW4*3D&reserved=0__;JSUlJSUlJSUlJSUlJSUlJSUlJSUlJSUl!!K-Hz7m0Vt54!mPAi5C1HHo3E9FbgFknLV1atAM-iwf0T-92yDt4r8e8_3IBw6V8u0JA2FXttB3auQ2jHnpfjZ6sxcWf8mz-8umc$
> <https://urldefense.com/v3/__https://nam12.safelinks.protection.outlook.com/?url=https*3A*2F*2Flistserv.linguistlist.org*2Fcgi-bin*2Fmailman*2Flistinfo*2Fhpsg-l&data=05*7C02*7Crchaves*40buffalo.edu*7Ca8b9b7346347478f06f008dd5fceb14c*7C96464a8af8ed40b199e25f6b50a20250*7C0*7C0*7C638772063545337622*7CUnknown*7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ*3D*3D*7C0*7C*7C*7C&sdata=4CDFd3jV49zrGkDtdZH0kdv5JcYC3egTLAtU5GxsPW4*3D&reserved=0__;JSUlJSUlJSUlJSUlJSUlJSUlJSUlJSUl!!K-Hz7m0Vt54!mPAi5C1HHo3E9FbgFknLV1atAM-iwf0T-92yDt4r8e8_3IBw6V8u0JA2FXttB3auQ2jHnpfjZ6sxcWf8mz-8umc$>
>
> >
> <https://urldefense.com/v3/__https://listserv.linguistlist.org/cgi-bin/mailman/listinfo/hpsg-l__;!!K-Hz7m0Vt54!mPAi5C1HHo3E9FbgFknLV1atAM-iwf0T-92yDt4r8e8_3IBw6V8u0JA2FXttB3auQ2jHnpfjZ6sxcWf8IFxrgOM$
> >
> > >
> > _______________________________________________
> > HPSG-L mailing list
> > HPSG-L at listserv.linguistlist.org
> >
> https://urldefense.com/v3/__https://nam12.safelinks.protection.outlook.com/?url=https*3A*2F*2Flistserv.linguistlist.org*2Fcgi-bin*2Fmailman*2Flistinfo*2Fhpsg-l&data=05*7C02*7Crchaves*40buffalo.edu*7Ca8b9b7346347478f06f008dd5fceb14c*7C96464a8af8ed40b199e25f6b50a20250*7C0*7C0*7C638772063545362731*7CUnknown*7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ*3D*3D*7C0*7C*7C*7C&sdata=3EP*2BK2Ejp07qvqO*2FvOF6rSwOW1HQKXvD3aWeqSnFIaI*3D&reserved=0__;JSUlJSUlJSUlJSUlJSUlJSUlJSUlJSUlJSU!!K-Hz7m0Vt54!mPAi5C1HHo3E9FbgFknLV1atAM-iwf0T-92yDt4r8e8_3IBw6V8u0JA2FXttB3auQ2jHnpfjZ6sxcWf8JveBGbQ$
> <https://urldefense.com/v3/__https://nam12.safelinks.protection.outlook.com/?url=https*3A*2F*2Flistserv.linguistlist.org*2Fcgi-bin*2Fmailman*2Flistinfo*2Fhpsg-l&data=05*7C02*7Crchaves*40buffalo.edu*7Ca8b9b7346347478f06f008dd5fceb14c*7C96464a8af8ed40b199e25f6b50a20250*7C0*7C0*7C638772063545362731*7CUnknown*7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ*3D*3D*7C0*7C*7C*7C&sdata=3EP*2BK2Ejp07qvqO*2FvOF6rSwOW1HQKXvD3aWeqSnFIaI*3D&reserved=0__;JSUlJSUlJSUlJSUlJSUlJSUlJSUlJSUlJSU!!K-Hz7m0Vt54!mPAi5C1HHo3E9FbgFknLV1atAM-iwf0T-92yDt4r8e8_3IBw6V8u0JA2FXttB3auQ2jHnpfjZ6sxcWf8JveBGbQ$>
>
> >
> <https://urldefense.com/v3/__https://listserv.linguistlist.org/cgi-bin/mailman/listinfo/hpsg-l__;!!K-Hz7m0Vt54!mPAi5C1HHo3E9FbgFknLV1atAM-iwf0T-92yDt4r8e8_3IBw6V8u0JA2FXttB3auQ2jHnpfjZ6sxcWf8IFxrgOM$
> >
> >
> _______________________________________________
> HPSG-L mailing list
> HPSG-L at listserv.linguistlist.org
> https://urldefense.com/v3/__https://listserv.linguistlist.org/cgi-bin/mailman/listinfo/hpsg-l__;!!K-Hz7m0Vt54!mPAi5C1HHo3E9FbgFknLV1atAM-iwf0T-92yDt4r8e8_3IBw6V8u0JA2FXttB3auQ2jHnpfjZ6sxcWf8IFxrgOM$
>
>
More information about the HPSG-L
mailing list