LINGUIST List 14.3374
Mon Dec 8 2003
Review: Syntax/Semantics: Lang, et al.(2003)
Editor for this issue: Naomi Ogasawara <naomilinguistlist.org>
What follows is a review or discussion note contributed to our Book
Discussion Forum. We expect discussions to be informal and
interactive; and the author of the book discussed is cordially invited
to join in.
If you are interested in leading a book discussion, look for books
announced on LINGUIST as "available for review." Then contact
Sheila Dooley Collberg at collberg
linguistlist.org.
Directory
Martin Schaefer, Modifying Adjuncts
Message 1: Modifying Adjuncts
Date: Mon, 8 Dec 2003 10:57:48 -0500 (EST)
From: Martin Schaefer <schaefemserver1.rz.uni-leipzig.de>
Subject: Modifying Adjuncts
Lang, Ewald, Claudia Maienborn and Cathrine Fabricius-Hansen,
ed. (2003) Modifying Adjuncts, Mouton de Gruyter, Interface
Explorations 4.
Announced at http://linguistlist.org/issues/14/14-820.html
Martin Schaefer, Institute of Linguistics, University of Leipzig
SUMMARY
The book presents a collection of papers originating from the Oslo
Conference ''Approaching the Grammar of Adjuncts'' in 1999. After an
introduction by the editors, the contributions are grouped into 4
sections: the argument-adjunct distinction, adjunct placement, case
studies on wieder/again and flexibility of event-related modification.
The section on the argument-adjunct distinction begins with David
Dowty's paper ''The dual analysis of adjuncts/complements in
Categorial Grammar'', pp. 33-66. Dowty argues that any complete
grammar should provide a dual analysis for complements as
adjuncts. Only this, he claims, can account for the fact that most
complements have same- category adjunct doubles. For him, the parallel
analysis as adjunct and complement allows one ''to formalize [] the
two endpoints of a complex psycholinguistic ''continuum'', a
psychological phenomenon where we are not ready to try to formalize
the intermediate points.''(p. 60)
In ''Genitives, relational nouns, and argument-modifier ambiguity'',
pp. 67-112, Barbara H. Partee and Vladimir Borschev discuss in detail
the possibility of providing a unified analysis of genitives for on
the one hand those genitives which are often argued to behave like
arguments, e.g. ''John's father'', on the other hand those genitives
that have modifier-like qualities, e.g. ''John's movie''. After a very
careful discussion of the data, which includes mainly English and
Russian examples, the authors finally reject a uniform analysis on the
basis of ellipsis data.
Manfred Bierwisch's ''Heads, complements, adjuncts: Projection and
saturation'', pp. 113-159, argues that the distinction between
complements and adjuncts is an essential property of
grammar. According to him, adjuncts discharge while complements
receive a Theta-Role provided by an argument position.
The second section on adjunct placement begins with ''Syntactic
conditions on adjunct classes'', pp. 163-209 by Werner Frey. Frey
argues that there is a distinction between five classes of adjuncts in
German and English, all of which are assigned base positions with
distinct structural requirements. The author also argues that adjuncts
can undergo scrambling in the middle field of German clauses. Frey
devotes a one page footnote to a defense of his base position for
manner adjuncts against the analysis proposed by Eckardt in this
volume.
Benjamin Shaer's '''Manner' adverbs and the association theory: Some
problems and solutions'', pp. 211-259, discusses whether or not the
reading of an adverb correlates with its syntactic position. According
to him, this correlation does indeed exist, though it cannot be
extended rightaway to fronted, parenthetical and afterthought
appearances of manner adverbs.
''Manner adverbs and information structure: Evidence from the
adverbial modification of verbs of creation'', pp. 261-305 by Regine
Eckardt investigates phenomena concerning transitive verbs in
combination with indefinite direct objects and manner adverbs. In
particular, she argues that the two available orders, manner adverb-
direct object vs direct object- manner adverb, are consequences of
different information structures assigned to the respective
sentences. A reflex of this are a limited number of readings for the
indefinite object in its pre-adverb position. Eckardt argues directly
against some points in Frey's paper, especially concerning syntactic
base positions.
Thomas Ernst's ''Semantic features and the distribution of adverbs'',
pp.307-334 argues that an adverb that carries the lexicosemantic
property of being subjective, defined here as being a) gradable and b)
whose mapping to a scale is influenced by the speaker's contextually
dependent judgement, cannot adjoin to the right in head-initial
languages, at least not in non-manner readings.
''Clause-final left-adjunction'', pp.335-362 by Inger Rosengren
investigates the syntactic behaviour of circumstantials, e.g. cause,
time and place modifiers, in German, English and Swedish. According to
her, the verbal head parameter is responsible for the reverse ordering
of these modifiers in German as opposed to Swedish and English.
The articles in the third section are all case studies of the
ambiguities surrounding wieder/again. ''Process, eventuality, and
wieder/again'', pp.365-391 by Karin Pittner argues for a syntactic
explanation of the ambiguity of wieder/again. In particular, Pittner
argues that ''wieder'' in its restitutive reading minimally c-commands
the verb in clause-final base position; in its repetitive reading it
minimally c-commands the base positions of all arguments. Pittner
believes that the meaning of ''wieder'' can in essence be reduced to
its repetitive meaning, the two readings being epiphenomena of which
entity ''wieder'' is applied to, either the whole eventuality
(=repetitive reading) or part of it (=restitutive
reading). Pittner incorporates into her discussion the data and
arguments given in the two next contributions.
''Competition and interpretation: The German adverb wieder
('again')'', pp. 393-416 by Gerhard Jaeger and Reinhard Blutner can be
divided into two parts. On the one hand, the authors hold that the
ambiguity of wieder/again cannot be handled by decompositional
accounts relying on scope to resolve the ambiguities, arguing directly
against von Stechov (1996). On the other hand, they present a
bidirectional optimality theoretic framework to account for the
ambiguity.
''How are results represented and modified? Remarks on Jaeger &
Blutner's anti-decomposition'', pp. 417-451, by Arnim von Stechow is a
direct reply to the previous contribution. The author discusses in
detail the criticism brought forward against the decompositional
approach, acknowledging that some data remain problematic. In the
second part, he incorporates the presented optimality theoretic
account into his decompositional framework.
The last section on flexibility of eventuality-related modification
begins with ''Event arguments, adverb selection, and the Stative
Adverb Gap'', pp. 455-474 by Graham Katz. Katz begins with the
observation that many adverbs, in particular manner adverbs, cannot be
combined with stative verbs, while one the other hand no adverb can
exclusively be combined with stative verbs (=the Stative Adverb
Gap). Katz then argues that stative verbs do not have an event
argument and for this reason do not allow for manner modification.
Claudia Maienborn's ''Event-internal modifiers: Semantic
underspecification and conceptual interpretation'', pp. 475-509 argues
for a two-way distinction in which adverbial modifiers are linked to a
sentence's eventuality argument. Event-external modifiers, located at
the VP periphery, relate to the event argument as a whole, while
Event- internal modifiers relate only indirectly to the event
argument. Technically, both types of modifiers bring with them a free
variable, which is either syntactically specified as the event
argument (=event- external modification), or is specified with the
help of contextually salient world knowledge, using abductive
inference.
''Flexibility in adverbal modification: Reinterpretation as contextual
enrichment'', pp. 511-552 by Johannes Doelling proposes a multi-level
model of meaning representation. This model assumes a level of
semantic form, where operations that lead to a strictly grammatically
determined, context-independent underspecified meaning operate. A
second level is the Parameter-fixed Structure, which results from
operations specifying the underspecified meaning with the help of
contextual enrichment. This model is then shown to account for
reinterpretation phenomena in temporal modification, as well as for
secondary predication.
In ''Secondary predication and aspectual structure'', pp. 553-590,
Susan Rothstein proposes to analyse secondary predicates, that is
resultatives and depictives, as aspectual modifiers. Secondary
predications introduce an additional event (besides the event
introduced by the matrix verb, that is). For depictives, the two
events in the sentence are subject to the constraint that the two
events must have the same running time and share one argument. For
resultatives, the same constraint must be met, but this time by the
culmination of the event introduced by the matrix verb and the event
introduced by the secondary predication.
''Real adjuncts in the Instrumental in Russian'', pp. 591-625 by
Assinja Demijanow and Anatoli Strigin is concerned with the analysis
of adjunct-DPs in the free instrumental case in Russian. After
presenting the many different interpretations such an adjunct-DPs can
receive, they concentrate on giving a uniform semantics for three of
these readings, using an abductive inference system fed by the
situations at hand and world knowledge.
The volume ends with Ilse Zimmermann's paper ''German participle II
constructions as adjuncts'', pp. 627-649. Zimmermann takes these
participle construction to be reduced sentences. She proposes two
different templates for the composition of modifiers with their
modificanda, distinguishing modification unifying two predicates
relating to participants/situations and modification in which the
modifier has operator status.
DISCUSSION
The papers in this volume give a comprehensive overview about
modifying adjuncts, with a strong emphasis on the semantic side of the
analysis. Two issues repeatedly occur in the contributions. First, a
major difficulty consists in finding a mechanism that is able to
derive all of the many different readings of modifying adjuncts
needed, especially when their derivation is context- and
world-knowledge-dependent. This problem is addressed in detail by the
papers of Doelling, Maienborn, and Demjjanow and Strigin, all of which
use inference via abduction (all three cite Hobbs et al. 1993 as a
major reference on this) in their derivation mechanisms.
The second issue is complementary to the first one, and concerns the
search for those constraints for such a mechanism that are given by
the language system itself, i.e. that are grammaticalized or
lexicalized. While two papers give more or less lexicalist
constraints on interpretations, or rather, on possible verb-adjunct
combinations (Katz, Ernst), many others concern syntactic
constraints. Thus, Frey, following up on his joint papers with Karin
Pittner (1998,1999), argues that the semantics of an adjunct place it
in on of the five different syntactic classes of adjuncts he
postulates. If this is correct, the syntactic position of an adjunct
can also be used in the derivation of its meaning. This is exactly
what two contributors (Maienborn and Pittner) argue for in the case of
locative modifiers and for the correct interpretation of wieder/again,
respectively. Both make use of the different syntactic classes
discussed by Frey. This picture is complicated by the data discussed
in other approaches. Thus, Eckardt's paper shows that the position of
a manner adverb relative to the direct object can reflect the
information-structural status of the direct object rather than a
different adverb reading, while Jaeger and Blutner's paper points to
the role of accentuation in the disambiguation of readings.
In addition, the data adduced by Shaer in his discussion of causal
readings of ''since'' sentences containing manner adjuncts are
support, in my view, for an information-structural account for some
instances of the syntactic positioning of pure manner adverbs. An
intriguing solution for the incorporation of these multiple factors
that play a role in the positioning of adjuncts into a semantic
analysis would be to extend the treatment for ''wieder'' offered by
von Stechow, which incorporates Jaeger and Blutner's optimality
theoretic approach, to the other cases. In this way, the basic insight
about the influence of syntactic position on adjunct interpretation
can be maintained while the other factors (information structure,
reading of the direct object, accentuation) could also be accounted
for.
Overall, the papers in this volume connect very well with each other,
be it explicitly or through their topics, and make for very
stimulating reading. The only downside is that the book uses endnotes
and not footnotes, which makes reading the text unnecessarily
complicated.
REFERENCES
Frey, Werner, and Karin Pittner (1998) Zur Positionierung der Adjunkte
im deutschen Mittelfeld. Linguistische Berichte 176:489-534
Frey, Werner, and Karin Pittner (1999) Adverbialpositionen im deutsch-
englischen Vergleich. In Sprachspezifische Aspekte der
Informationsverteilung. M. Doherty (ed.), 14-40. Berlin: Akademie
Verlag.
Hobbs, Jerry R., Mark E. Stickel, Douglas E. Appelt and Paul
Martin.(1993) Interpretation as Abduction. Artificial Intelligence
63:69-142
Stechow, Arnim von (1996) The different readings of wieder 'again': A
structural account. Journal of Semantics 13:87-138
ABOUT THE REVIEWER
I work in a DFG (www.dfg.de)-sponsored project on event structures and
am currently writing my PhD Thesis on the formal semantics of manner
adverbs. Feel free to visit my homepage at:
http://www.uni-leipzig.de/~semantik/people/martin.html