17.2099, Review: Syntax, Semantics: Pafel (2006)

[linguist at LINGUISTLIST.ORG]

LINGUIST List: Vol-17-2099. Thu Jul 20 2006. ISSN: 1068 - 4875.

Subject: 17.2099, Review: Syntax, Semantics: Pafel (2006)

Moderators: Anthony Aristar, Wayne State U <aristar at linguistlist.org>
            Helen Aristar-Dry, Eastern Michigan U <hdry at linguistlist.org>
 
Reviews (reviews at linguistlist.org) 
        Laura Welcher, Rosetta Project / Long Now Foundation  

Homepage: http://linguistlist.org/

The LINGUIST List is funded by Eastern Michigan University, Wayne
State University, and donations from subscribers and publishers.

Editor for this issue: Laura Welcher <laura at linguistlist.org>
================================================================  

This LINGUIST List issue is a review of a book published by one of our
supporting publishers, commissioned by our book review editorial staff. We
welcome discussion of this book review on the list, and particularly invite
the author(s) or editor(s) of this book to join in. To start a discussion of
this book, you can use the Discussion form on the LINGUIST List website. For
the subject of the discussion, specify "Book Review" and the issue number of
this review. If you are interested in reviewing a book for LINGUIST, look for
the most recent posting with the subject "Reviews: AVAILABLE FOR REVIEW", and
follow the instructions at the top of the message. You can also contact the
book review staff directly.


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

1)
Date: 20-Jul-2006
From: Michael Putnam < mtputnam at umich.edu >
Subject: Quantifier Scope in German (Syntax) 

	
-------------------------Message 1 ---------------------------------- 
Date: Thu, 20 Jul 2006 12:34:28
From: Michael Putnam < mtputnam at umich.edu >
Subject: Quantifier Scope in German (Syntax) 
 

Announced at http://linguistlist.org/issues/17/17-357.html 

Author: Pafel, Jürgen
Title: Quantifier Scope in German
Publisher: John Benjamins
Year: 2005

Michael T. Putnam, Department of Languages and Literatures, Michigan State
University

*This review benefited greatly from comments and correspondence with Winfried
Lechner, Jie Zhang and Jürgen Pafel. Any remaining shortcomings are my own. 

INTRODUCTION

As clearly stated in the title, the central topic of this monograph focuses on
quantifier scope in German. In this revised version of his Habilitationsschrift
"Skopus und logische Struktur. Studien zum Quantorenskopus im Deutschen" (1997)
(Universität Tübingen) Jürgen Pafel delivers a comprehensive account of
quantifier scope in German. The scope behavior both of ordinary quantifiers and
negative, adverbial, interrogative, relative and particle quantifiers is
investigated. In the preface to this work Pafel accurately affirms the bold
challenge of analyzing quantifier scope within the parameters of linguistic theory:

"Quantifier scope is a challenge to linguistic theory as it is a phenomenon
which is determined by the interplay of different factors. Such interactions
have not been studied in depth and theoretical frameworks are not well prepared
to cope with true interactions ...Quantifier scope is basically a semantic
phenomenon, it is, however, an interface phenomenon par excellence too, as the
interaction of factors determining quantifier scope is part of the constraints
relating syntax and semantics, more precisely, part of the constraints
connecting syntactic and semantic structure" (xiii).

In this work Pafel develops an independent level of linguistic structure known
as "semantic structure", or quite simply "s-structure", at which semantic
phenomena such as quantifier scope are constructed and interpreted. This
predominantly functionalist approach proposes "linear models" with linear
equations "which predicts the observed scope readings and their degree of
preference quite accurately" (p. 51). In establishing the form and function of
s-structure and how it would properly working within a linguistic theory, Pafel
reaches the conclusion that transformational models of grammar are flawed in
their inability to accurately predict (i.e. a lack of both descriptive and
explanatory adequacy) quantifier scope constructions in German. Likewise, Pafel
also points out potential conceptual weaknesses quantifier scope constructions
in German encounter within Optimality-Theory. The purpose of this review is to
offer a critical analysis of Pafel's contribution to linguistic theory in this
book, namely, the concept of a separate, non-configurational linguistic level
labeled semantic structure. 

REVIEW OF GENERAL ARGUMENT

In Chapter 1 (Preliminaries), Pafel gives an introduction to the empirical and
theoretical aspects of quantifier scope and presents the syntactic and semantic
assumptions that enter into this investigation. In my opinion, Pafel does an
excellent job of setting up his general argument with a thorough discussion and
explanation of the basic terminology that he makes regular use of throughout the
course of the text. Pafel even goes into substantial detail in describing and
defending his data elicitation and evaluation procedures. In Section 1.1.2,
Pafel attacks the "syntactocentristic" view of linguistic theory as developed by
Noam Chomsky since 1975. The main premise behind this criticism of the Chomskian
model lies in the lack of independence awarded to semantic structure. Within
this transformational system there is a level of semantic interpretation called
logical form (LF), where scope takes the form of a syntactic theory of scope;
however, to a large extent the semantics of a given sentence is dependent upon
its syntactic structure. In contrast, Pafel favors a non-configurational view
the linguistic system which copes with diverse aspects of linguistic signs (cf.
their phonetic-phonological, prosodic, morphological, syntactic, and semantic
aspects) by generating relative autonomous, parallel structural descriptions of
these aspects. While campaigning for "semantic independence" one of the
fundamental issues ? if not the fundamental issue ? discussed in this book is
the true nature of semantic structure and its relationship to syntactic structure. 

In Chapter 2 (Relative scope), Pafel further develops his linear model of how
quantifier scope is determined in German and comes to the conclusion that
relative scope is configurationally represented on a level of representation
which is not a syntactic, but a semantic, level dubbed "semantic structure". In
his discussion of "some basic patterns of Q/Q-interaction" (2.1), Pafel
introduces an apparent weakness in attributing quantifier scope to syntactic
considerations (i.e. by assuming that relative scope can be determined by the
c-command relations of quantifiers and their traces). Pafel contends that
"configurational theories" such as minimalism and their treatment of quantifier
scope would lead to ungrammatical constructions such as Weak Cross Over (WCO)
effects. Interestingly here Pafel provides only English data (p. 59) to
illustrate WCO effects. The generalization that changes in the word order of
arguments do not lead to WCO effects in German is attested by many authors (cf.
Webelhuth 1992, Frank, Lee & Rambow 1996, Putnam 2006). Therefore the relevance
of mentioning WCO effects as a possible shortcoming of transformational theories
such as minimalism in relation to quantifier scope bears no effect on the
discussion of German data. 

In the middle of Chapter 2, Pafel establishes a healthy laundry list of multiple
linguistic properties responsible for determining relative scope. The author
identifies no less than 13 properties that determine relative scope, which can
be divided into eight groups.

1.  Syntactic constellation
2.  Grammatical function
3.  Thematic Property
4.  Distributivity
5.  D(iscourse)-binding
6.  Definiteness
7.  Focus
8.  Negation Attraction

In discussing the "syntactic constellation" in relation to quantifier scope,
Pafel collapses linear precedence and c-command into this aforementioned term.
This is somewhat of a misnomer in that constituents may indeed reside in a
position of linear precedence while simultaneously not c-commanding lower
constituents (for example, if the linear superior element participates in
adjunction). The lack of clear distinctions between linear precedence and
c-command sometimes blurs the exact nature and function of "syntactic
constellation" in Pafel's model. According to Pafel, the critical data are to be
found in the relations between attributive quantifiers inside a noun phrase,
i.e. the scope relations between quantificational PP or genitive attributes in
noun phrases (p. 63-4). Pafel introduces sub-domains of VP to support his
properties of internal-precedence (IN-PRE) and external-precedence (EX-PRE),
making a distinction between quantifiers that reside in the Vorfeld and
Mittelfeld domains of German clauses. The ontology of these 'sub-domains'
tangentially resemble Chomsky's notion of strong phases (vP and CP).
Surprisingly, Pafel does not make mention of Chomsky's seminal work on
minimalism (the latest work of Chomsky's that he sites in his bibliography is
Barriers (1986)). 

The remainder of Chapter 2 demonstrates how the interplay of these various
grammatical properties of quantifier scope are calculated and assessed through
weighted, numerical constraints (pp. 88-106). At first blush it would appear
that such constraints may be able to be mapped into a representational theory of
grammar such as Optimality Theory (OT), Pafel addresses this claim by noting a
stark contrast between linear models and OT. In the former model, structures are
evaluated as possible/impossible and preferred/not preferred such that there
might exist several possible structures with distinct or the same degree of
preference. The structures do not compete with one another, but ever structure
is evaluated independently of the other ones. Finally, in linear model the
constraints are numerically weighted rather than hierarchically arranged as in
OT (cf. p. 124). Although Pafel rejects traditional OT, it appears that his
approach is representational at its core. Arguing for a level where quantifier
scope is configurationally represented ? distinct from the syntactic level ?
also offers a blueprint for a potential new design of LF (p.115). 

Chapters 3 (Absolute scope), 4 (Indefinites and quantifiers) and 5
(Interrogative quantifiers) provide an in depth discussion and treatment of the
aforementioned aspects of quantifier scope and their existence in German. In
Chapter 3, the quantifier scope properties of complex sentences in German are
investigated with an integrated model of absolute and relative scope
determination being introduced as the appropriate mechanism to handle these
structures. In Chapter 4, different types of indefinite noun phrases are
distinguished: quantificational and non-quantificational ones, and, among the
non-quantificational ones, non-specific, (weakly and strongly) specific and
generic indefinite noun phrases. This chapter uncovers the quirky behavior of
non-quantificational indefinite noun phrases: these different indefinites are
"uniformly analyzed as names (of a kind), which, in semantic structure, may lead
to, or trigger, the introduction of a quantifier which ranges over the instances
of the kind (the quantifier can be an existential one or some variety of a
generic quantifier)" (pp. xiv-xv). This special type of quantifier is argued to
not have a syntactic counterpart and thus only appears in the level of semantic
structure. Chapter 5 completes Pafel's investigation of the major areas of
quantifier scope in German and confirms the hypothesis established in earlier
chapters that wh-phrases and are quantifiers subject to the same regularities as
non-interrogative quantifiers. 

Chapter 6 (Alternative scope accounts) and Chapter 7 (Towards a theory of
semantic structure) function as the locus of Pafel's central claims in this
book. In Chapter 6, a survey of alternative theoretical models (configurational,
semantically and pragmatically-based theories as well as multi-factor theories)
are introduced and critiqued. In Chapter 7, the basic construction and
interpretation of the independent level of semantic structure is delivered.
Turning first to Chapter 6, Pafel asserts that modern configurational (syntactic
theories) attempts to explain scope can be deduced purely from formal structural
configurations (i.e. c-command). In particular, Pafel takes issue with May's
(1985) reformulation of LF to account for wh/Q-interaction and its inability to
account for the facts of this relationship in both English and German (cf.
Section 5.2). To support this claim Pafel provides two pivotal data sets showing
that the reading with the direct object outscoping the subject depends on the
lexical properties of the quantifiers involved ((1) and (2) below) and sentences
with topicalization ((3) ? (6) below) (both data sets taken from Pafel pp. 239-40). 

(1)  Most of the students read every book.

     (every > most: *)

(2)  Some of the students read every book.

     (every > some: OK)

Topicalization examples:

(3)  All of us have read many of these books with great enthusiasm.

     (ambiguous)

(4)  Many of these books, all of us have read with great enthusiasm.

     (only: many > all)

(5)  Many people come to New York every summer.

     (ambiguous)

(6)  Every summer, many people come to New York.

     (only: every > many)

As noted by Pafel, "Changing the relative precedence of two quantifiers by
topicalization in English has the same effect as topicalization and scrambling
in German: the scope value of the preceding quantifier becomes distinctly great
such that the wide scope reading of the quantifiers becomes available or even
the only possible reading ? depending on how high the scope value is without
taking syntactic constellation into account" (p. 240). In conclusion to his
discussion of configurational theories of scope interpretation and their
subsequent weaknesses, Pafel notes that since the mid 1980s it has become common
place to interpret not only the position of the quantifier in syntactic
structure, but also the position of its traces. A blatant shortcoming of these
sorts of theories is that they are ununified with respect to exactly where scope
determination takes place: on logical form, on s-structure or on a reconstructed
structure (cf. p. 244).  

Concerning multi-factor theories (cf. Section 6.3), Pafel primarily focuses his
attention on the work of Beghelli, Stowell and Szabolcsi in Szabolcsi (ed.)
(1997). The basic idea behind these theories is that quantifiers can or must
occupy distinct position at LF depending on the quantifier type they belong to.
Under this approach, scope determination "becomes an epiphenomenon of feature
checking" (p. 255). Such a minimalist theory makes the following predictions
(data from Pafel 256-57).

(7)  Every (/each) student reads two books.  (ambiguous)

(8)  Two students read every (/each) book.  (ambiguous)

(9)  Jeder Pianist hat zwei Beethoven Sonaten in seinem Repertoire.  (unequivocal)
     every piano play has two Beethoven sonatas in his repertoire
    'Every piano player has two Beethoven sonatas in his repertoire.'

(10) Zwei Beethoven-Sonaten hat jeder Pianist in seinem Repertoire.  (ambiguous)
     two Beethoven sonatas has every piano play in his repertoire
     'Two Beethoven sonatas, every piano player has in his repertoire.'

In both (7) and (8) we have a group-denoting quantifier phrase (GQP) (two books,
two students) and a distributive-universal quantifer phrase (DQP) (every or
each). Accordingly, "a GQP should be scopally ambiguous with respect to a
clausemate DQP, depending on whether the GQP move to Spec of RefP or to Spec of
ShareP" (Beghelli/Stowell 1997:80). The data in (9) and (10) show that the
predictions of the theory are at variance with the facts in German (and several
other languages). According to Beghelli/Stowell, any sort of contrast in (9) and
(10) is unpredicted and unexpected; both sentences should produce an equally
ambiguous scope reading. As a result, Pafel concludes that fixed positions at LF
do not and can not explain the various grammatical properties involved in
determining relative scope. 

Chapter 7 provides a sketch of the nature of semantic structure, the level of
analysis where the meaning-compositional aspects of the readings of a sentence
and its parts are represented. In this chapter, Pafel focuses on three principal
aspects of his newly defined semantic structure, namely, the semantic features
and semantic structures that make up this level of analysis (7.1), the
interpretation of semantic structures (7.2) and the construction of semantic
structures from syntactic ones (7.3). Chapter 8 (Conclusion) ? which consists of
only two pages ? wraps up this work.  

CRITICAL ANALYSIS

First and foremost, this book is an excellent resource for anyone pursuing an in
depth study of quantifier scope in German. The ocean of data provided in the
first five chapters function as a great springboard for anyone researching this
topic or topics closely related to it. My main criticisms with this work lie
within the theoretical assumptions and conclusions put forth by Pafel.  First,
let us consider Pafel's statements about OT. Pafel's brief discussion of OT (p.
124) concludes that the main problem that besets this representational framework
is its lack of cumulativity. According to a cumulative view of OT, a candidate
is deemed a loser in OT if it violates a higher ranked constraint that other
competing candidates. With cumulativity through the introduction of numerical,
weighted constraints in a linear model this is not a problem: A candidate can
violate a higher ranked constraint and be grammatical if it fulfills enough
lower ranked constraints, i.e. the violation can be compensated. Thus "a
quantifier's tendency to wide scope gets stronger if it satisfies more
scope-relevant properties ... In other words, the fact that the values a
quantifier gets relative to each scope-relevant properties are added up to its
scope value is what makes the linear model cumulative, or, additive" (Jürgen
Pafel, personal communication). Such a conclusion rests solely upon a primitive
notion of OT and ignores the works of those such as Flemming (2001) on how to
appropriately include weighted constraints into the OT-framework. As shown by
Flemming, the application of weighted constraints in OT can be achieved without
introducing additional complex machinery to the language system. Speaking of
complex machinery and economy considerations, similar to OT, Pafel's linear
model is incredibly complicated in that multiple (perhaps an infinite number of
them?) candidates must be evaluated along with the calculation and assessment of
each and every numerically weighted constraint. Therefore although this model
addressed a multitude of grammatical properties involved in determining
quantifier scope although with their numerical weight, it is inherently complex
from a mental processing standpoint. 

Considering its potential relation to derivational, configurational theories of
syntax-semantics interaction (i.e. minimalism), Pafel states that his semantic
structure is "a new kind of argument for 'logical form'" (p. 115). Adherents to
the minimalist program should focus on Section 7.3 (the construction of semantic
structures from syntactic structures) in any attempt to better understand how
"semantic structure" could potentially serve to improve and revise an
understanding of LF. Many of the theoretical shortcomings pertaining to
configurational (syntactic) and multi-factor frameworks brought to light by
Pafel in Chapter 6 should, however, be seriously considered. For example,
Pafel's claims that since the mid 1980s most, if not all, derivational
treatments of scope determination rest not only on the quantifier's final
position but also the placement of its traces exposes a current weakness in the
minimalist program, namely, how exactly are lower traces, i.e. copies, of moved
constituents interpreted? If quantifier raising involves adjunction, is there
perhaps a different sort of trace/copy mechanism involved in this mechanism of
traversal? Regarding the argument for or against fixed positions at LF, Pafel
(personal communication) points out that many operations ? such as middle field
scrambling in German ? that also appear to lack fixed positions and can hardly
be attributed solely to operations in the narrow syntax. This, of course, raises
the timeless question behind the motivation of movement operations in the narrow
syntax. Not only that, but such data and inquiries also question the premise of
a catographic approach to the construction of phrase structure and natural
clauses. In this regard, Pafel's work also finds a home among scholars of
minimalism in their quest to better understand LF. 

In conclusion, this book is well-written and has a nice, logical progression
with its presentation of data and relative arguments. In Chapters 6 and 7, Pafel
raises many questions for proponents and opponents to dominant linguistics
schools of thought (e.g. minimalists, adherents of OT, those in support of
non-transformational frameworks, etc.). Although what Pafel provides his reader
regarding his notion of "semantic structure" in Chapter 7 is only a sketch at
this point in time, it will be interesting in time to see what developments and
adaptations stem from this work, not only by the hands of the author, but also
by those who use this book as a reference source.

REFERENCES

Beghelli, Filippo and Tim Stowell. (1997) Distributivity and Negation: The
Syntax of Each and Every. In Anna Szabolcsi (ed.), Ways of Scope Taking.
Dodrecht: Kluwer, 71-107.

Chomsky, Noam. (1986) Barriers. Cambridge: MIT Press. 

Flemming, Edward. (2001) Scalar and Categorical Phenomena in a Unified Model of
Phonetics and Phonology. Phonology. 18(1): 7-44.

Frank, Robert, Young-Suk and Lee, and Owen Rambow. (1996) Scrambling
Reconstruction and Subject Binding. Rivista di Grammatica Generativa. 21: 67-106.

Putnam, Michael T. (2006) Scrambling in West Germanic as XP-Adjunction: A
Critical Analysis of Prolific Domains. Ph.D. dissertation. University of
Kansas-Lawrence. 

Szabolcsi, Anna (ed.). (1997) Ways of Scope Taking. Dordrecht: Kluwer.

Webelhuth, Gert. (1992) Principles and Parameters and Syntactic Saturation.
Oxford: Oxford University Press. 


ABOUT THE REVIEWER

Michael T. Putnam is a Visiting Assistant Professor of German at Michigan
State University. His primary research areas lie in German linguistics and
syntactic theory, with a particular focus on the syntax-pragmatic
interface, Germanic languages (diachronic and synchronic) and
German-American dialects (with a focus on Pennsylvania German).





-----------------------------------------------------------
LINGUIST List: Vol-17-2099