Non‐restrictive modification
A modifier M is non‐restrictive if it is irrelevant for the denotation or reference of the modified phrase,
Non‐restrictive modification by evaluative predicates
e.g. [[ M H ]] = [[ H ]]
cf. workshop "Semantic and Pragmatic Properties of (Non)Restrictivity"
(www.ims.uni‐stuttgart.de/~arndt/restrictivity.html)
Carla Umbach
Non‐restrictive modification
Zentrum für Allgemeine Sprachwissenschaft (ZAS), Berlin
[email protected]‐berlin.de
• relative clauses, appositions, attributive adjectives, …
• definite NPs (cf. Fabricius‐Hansen 2009), indefinite NPs, ….
Information, discourse structure and levels of meaning “IDL12″
workshop, Barcelona, October 25‐26th 2012
(1) In der Müllverbrennung entstehen gesundheitsschädliche Dioxine.
' Waste incineration produces noxious dioxines.'
1
Empirical vs. evaluative predicates (1)
Empirical vs. evaluative predicates (2)
Non‐restrictive modification by evaluative predicates:
Non‐restrictive modification by empirical predicates:
(2)
a. weiße Schimmel tautology
(3)
'white whitehorses'
b. gesundheitsschädliche Dioxine a.
schöne Blumen speaker assessment
'beautiful flowers'
fact
b.
empirical
'noxious dioxines'
c. heißer Kaffee
2
unschuldige Zivilisten speaker assessment
evaluative
'innocent civilians'
context‐dependent
c.
'hot coffee'
blöde Beamte speaker assessment
'stupid officers'
d.
dreckige Gauner ambiguous
'dirty crooks'
3
Observation (German): Most non‐restrictively modified indefinites contain
evaluative adjectives
(cf. Castroviejo & Schwager 2008 for Catalan, Demonte 2008 for Spanish)
4
Potts (2005): Conventional implicatures
Core question / plan
Non‐restrictive modification is clearly not part of the assertion

Two problems with Potts' solution:
non‐restrictive modification is a conventional implicature
cf. Potts (2005), Morzycki (2008)
Potts (2005): two meaning dimensions i.
impossible to distinguish empirical adjectives from evaluative ones; ii.
no link to evaluative adjectives in predicative position – no link to the
faultless disagreement debate.
<at‐issue, CI>
?? Why are evaluative predicates more susceptible for non‐restrictive interpretation than regular ones? conventional implicatures (CI):  comments by the speaker on the assertion
 scopeless, no accomodation, no truth value gaps, informativity required
 include non‐restrictive relative clauses, appositions, expressives etc.
In this talk:
 semantics/pragmatics of evaluative predicates in predicative position
 adapt these findings to attributive position
confirm: explain: Non‐restrictive evaluative predicates are CIs Why are evaluative predicates considered non‐restrictive, and how do they differ from regular ones?
5
Non‐restrictive modification and focus
(4)(a) The idea of faultless disagreement
In dem Film geht es um DRECKIGEF Gauner. 'The movie is about dirty crooks (not about clean ones).'
(b) In dem Film geht es um dreckige GAUNERF. 'The movie is about dirty crooks (not about dirty citizens).'
(c) In dem Film geht es um [dreckige GAUNER]F. 'The movie is about dirty crooks (not about citizens).'
6
(5)
Ann:
Ben:
Licorice is tasty.
No, it isn't. It tastes terrible.
Intuition: Ann and Ben may both be right
 "faultless disagreement"
Contextualist accounts (Stephenson 2007): Evaluative predicates include a hidden experiencer variable,
e.g., Licorice is tasty [to me].
 Non‐restrictively used adjectives must not be narrowly focussed , (1a), and not be backgrounded, (1b)
No correspondence between non‐restrictive adjectives and background:
Focus ==> restrictiveness (Riester 2012)
Background =/=> non‐restrictiveness
Relativist accounts (Lasersohn 2005, 2009, Stephenson 2007): The truth of evaluative propositions depends on a judge,
e.g., Licorice is tasty judged by Anne
 evidence for CI interpretation
7
8
Faultless disagreement is a misconception
Empirical / subjective propositions – universal / private judgments
Faultless disagreement presupposes a bird's eye point of view.
From the point of view of the discourse particpants, disagreement is never faultless (Stojanovic 2007, Umbach in prep.)
empirical propositions
subjective propositions (empirically testable)
(ascribed by the speaker)
Licorice contains sugar.
Licorice is tasty .
Osnabrück is in Denmark.
Flowers are beautiful.
 either there is genuine disagreement
 or the dissent is relativized (experiencer, finden, ….)
(6)
Ann:
Ben:
universal judgment
(intended to enter
the CG)
Licorice is tasty.
No, it isn't. It tastes terrible.
a. Ann: It is tasty. And it’s not just that I find it tasty; it’s tasty tout court.
b. Ann: OK. To my taste, Licorice is tasty; that’s all I’m saying.
(cf. Stojanovic 2007)
c. Ann: OK. Ich finde Lakritze lecker. (cf. Saebo 2009)
'I think Licorice is tasty.'
private judgment
(intended to be an individual discourse
commitment only,
cf. Farkas & Bruce 2010)
Licorice is tasty to me.
Flowers are beautiful to me.
Ich finde Lakritze lecker.
Ich finde Blumen schön.
'I think licorice is tasty / …'
9
10
Which propositions are subjective? and why?
Subjective cut‐off points
Use German finden ('think/consider') as a testbed:
Subjective propositions include
 positive / (comparative) form of evaluative adjectives (lecker, schön)
 positive form of dimensional adjectives (groß)
A proposition  is subjective iff Ich finde  is grammatical.
Consider gradable predicates:
(7) a. Ich finde Lakritze lecker / das Bild schön. Subjective propositions are about cut‐off points:
 for dimensional adjectives: meta‐linguistic interpretation (Barker 2002)
'I find licorice tasty / the painting beautiful.'
b. Ich finde die Wohnung groß.
'I find the appartment big.'
 for evaluative adjectives:
descriptive and metalinguistic interpretation cannot be distinguished, because the scale is not proportional (cf. also Sassoon 2011)
d. Ich finde die Skulptur schöner als das Bild.
'I find the sculpture more beautiful than the painting.'
e. *Ich finde Sue schwanger.
'I find Sue pregnant.'
f.
*Ich finde Rauchen gesundheitsschädlich.
'I find smoking harmful to health.'
c.
.
*Ich finde Sue größer als Sven.
'I find Sue taller than Sven.'
11
(8) [[pos]] = f<e,d>. x. f(x)  standard(f)(norm(c))(C) empirical propositions
subjective propositions
[[pos]] = f<e,d>. x. f(x)  standard(f)(ag(c))(C) where f is an adjective denotation, c is a context, C is a comparison class
12
Empirical / subjective propositions – universal / private judgments
Metalinguistic interpretation of degree assertions (Barker 2002)
Feynman is tall.
(i) descriptive interpretation: inform the hearer about Feynman's height.
"What is Feynman's height?"
 Feynman's height exceeds the tall-standard
(ii) metalinguistic interpretation: inform the hearer about the standard of
tallness in the context.
"Who counts as tall in this context?"
 the tall-standard is below Feynman's height
descriptive interpretation
empirical propositions
(empirically testable)
subjective propositions (ascribed by the speaker)
 commonly agreed on cut‐off  speaker‐dependent
cut‐off Licorice contains sugar.
Licorice is tasty .
Osnabrück is in Denmark.
Flowers are beautiful.
universal judgment
(intended to enter
the CG)
metalinguistic interpretation
height
height
tall standard
private judgment
tall standard
13
Private judgments need not be shared
Farkas & Bruce (2010):
An individual discourse commitment of a discourse participant X is a set of propositions DCX – propositions that are public commitments of X but are not in the common ground.
(intended to be an individual discourse
commitment only,
cf. Farkas & Bruce 2010)
Licorice is tasty to me.
Flowers are beautiful to me.
Ich finde Lakritze lecker.
Ich finde Blumen schön.
'I think licorice is tasty / …'
14
Evaluative predicates in attributive position
(9)
In Anna Garten stehen schöne Blumen 'There are beautiful flowers in Ann's garden.'
?? Why are evaluative predicates more susceptible for non‐restrictive interpretation than regular ones?  Restrictive modification
must be able to carry a focus, and trigger alternatives;
 must come with a well‐defined complement
 requires commonly accepted cut‐off point
universal judgments (empirical as well as subjective ones) are intended to enter the common ground
 universal judgments containing subjective propositions come with a normative intention
 If there is no commonly accepted cut‐off point, the modifying predicate
is considered as a comment by the speaker,  "non‐restrictive"
private judgments are intended to be individual discourse commitments only
15
 private judgments allow to express dissent without disagreement
 Evaluative predicates are more susceptible for non‐restrictive interpretation because they often cannot be used restrictively. 16
Conclusion
References
Adapting the semantics/pragmatics of evaluative predicates in predicative position to attributive position
confirms: Non‐restrictive evaluative predicates are CIs moreover: CIs are individual discourse commitments (Farkas & Bruce 2010)
explains: Evaluative predicates are susceptible to non‐restrictive
interpretation because they frequently lack a commonly agreed upon denotation
Why can beautiful flowers not be used restrictively to mean something like "the flowers that I think are beautiful" (as opposed to the ones I don't think are beautiful)?
 because you would need to know what I think in order to pick the intended ones.
17
Castroviejo, E. & M. Schwager (2008) Amazing DPs. In T. Friedman and S. Ito (eds.) Proceedings of SALT XVIII. Ithaca, NY: CLC Publications, 176‐193.
Demonte, Violeta (2008) Meaning‐form Correlations and Adjective Position in Spanish. In. Ch. Kennedy & L. Mc Nally (eds.): The semantics of adjectives and adverbs. Oxford University Press.
Farkas, D. & Kim Bruce (2010) On reactions to Assertions and Polar Questions. Journal of Semantics 27(1), 1‐37.
Fabricius‐Hansen, Cathrine (2009) Überlegungen zur pränominalen Nicht‐Restriktivität. Linguistische berichte 16. 89‐111
Kant, Immanuel (1790) Kritik der Urteilskraft.
Kölbel, Max (2004) 'Faultless disagreement', Proceedings of the Aristotelian Society 104.53‐73 Lasersohn, P. (2005) Context Dependence, Disagreement, and predicates of Personal Taste. Linguistics and Philosophy 28:643–686.
Lasersohn, Peter (2009) Relative Truth, Speaker Commitment, and Control of Implicit Synthese. 166 (2), 359‐374. Martin, Fabienne (2011) Decomposing (non‐)restrictivity. Evaluative modifiers in post‐
head positions. Handout, workshop on Modification with/without modifiers, Madrid.
McNally, Louise & Gemma Boleda Torrent (2003) Relational adjectives as properties of kinds’. In O. Bonami & P. Cabredo Hofherr (eds.), Empirical Issues in Syntax and 18
Semantics, vol. 5, pp. 179–196.
References
Morzycki, M. (2008) Nonrestrictive Modifiers in Nonparenthetical Positions. In McNally & Kennedy (eds.) Adverbs and Adjectives. Oxford University Press. Potts, Christopher (2005) The Logic of Conventional Implicatures. Oxford University Press. Riester, Arndt (2012) To Restrict is to Focus. Talk at Workshop "Semantic and Pragmatic Properties of (Non)Restrictivity", www.ims.uni‐stuttgart.de/~arndt/restrictivity.html
Saebo, K.J. (2009) Judgment Ascriptions. Linguistics and Philosophy 32, 327–352.
Sassoon, Galit (2011) Be positive! Norm‐related implications and beyond, Proceedings of
Sinn und Bedeutung 15, Universität des Saarlandes, Saarbrücken, Germany.
Stephenson, T. (2007). Judge dependence, epistemic modals, and predicates of personal taste. Linguistics and Philosophy, 30, 487–525.
Stojanovic I. (2007) Talking about taste: disagreement, implicit arguments, and relative truth. Linguist and Philosophy 30:691–706
Umbach, C. (2006) Non‐restrictive Modification and Backgrounding. Proceedings of the Ninth Symposium on Logic and Language, Hungarian Academy of Sciences, 152 ‐159.
Umbach, C. (in prep.) Evaluative predicates, subjective propositions and private judgments.
Zangwill, Nick (2007) Aesthetic Judgment. Stanford Encyclopedia of Philosophy.
Workshop "Semantic and Pragmatic Properties of (Non)Restrictivity", Stuttgart, March 2012, organized by Fabienne Martin & Arndt Riester, cf. www.ims.uni‐stuttgart.de/~arndt/restrictivity.html
19
20