Lombardi, Linda, ed. (2001) Segmental Phonology in
Optimality Theory: Constraints and Representations.
Cambridge University Press, vii+300pp, hardback ISBN
Adam Ussishkin, University of Arizona.
Book announcement on Linguist:
As made obvious by the title, this book is mainly
concerned with segmentally- and featurally-based
phonological phenomena, and the chapters in this volume
provide accounts of these phenomena within the framework of
Optimality Theory (OT; Prince & Smolensky 1993). As
Lombardi notes in the introduction, OT's early success
centered mainly on issues connected not to featural and
segmental issues, but rather on issues relating to the
interaction between phonology and morphology, as well as on
prosodic phonology. A goal of this volume is to showcase
some of the recent body of work that concentrates on
segmental issues within OT. The book is divided into three
sections. The first, "The content of representations,"
contains four chapters that make use of OT-internal
arguments for adopting particular representations within
work on segmental phonology in OT. The second section,
titled "The content of constraints," contains three
chapters which involve examining particular types of OT
constraints. The broad goal of this section is to
demonstrate what elements constraints may refer to, and
deals with both markedness and faithfulness constraints.
Finally, the third section is titled "The structure of the
grammar." This section contains two chapters which deal
with the issue of phonological opacity. This issue has been
especially problematic for OT, and different mechanisms
have been proposed to account for various types of opacity.
The two chapters in this section each propose different
ways of accounting for opacity.
Introduction: Linda Lombardi.
1: Why Place and Voice are Different: Constraint-Specific
Alternations in Optimality Theory, Linda Lombardi.
Synopsis: The main issue in this chapter is the difference
between the behavior of laryngeal and place features.
Although these feature types have much in common (such as
their behavior in coda position), Lombardi argues that only
place features can be subject to position-dependent
constraints. Additionally, Lombardi argues for the
existence of constraints of the form Max-Feature, which are
essentially a mechanism to ensure preservation of features
present in input representations, analogous to the Max-
Segment constraints introduced in correspondence theory
(McCarthy & Prince 1995). The data brought to light in the
second section of the chapter involve coda position effects
on both laryngeal features (voicing) and place features.
The analysis that follows captures the different behavior
of these two types of features, resulting in an account
that recognizes laryngeal features as privative, as well as
Max-Feature constraints. Additionally, Lombardi argues that
laryngeal segments are not placeless segments but must be
considered as pharyngeal.
2: Constraints and Representations in Subsegmental Phonology,
Synopsis: In this chapter, Zoll argues that no underlying
representational difference is necessary to distinguish
floating features from latent segments, and that the
distinction between these two types of elements is easily
captured within OT. Crucially, all subsegments are
uniformly represented as features with no affiliation to a
root node underlyingly. The realization of the subsegment
depends on the ranking between faithfulness constraints
demanding preservation of the subsegment, and markedness
constraints regulating both the output position of the
subsegment as well as feature combinations. Languages
analyzed include Amharic, Inor, and Yawelmani.
3: Phonological Contrast and Articulatory Effort, Robert
Synopsis: Kirchner argues in this chapter for a view of OT
in which phonological representations are enriched, such
that they include the full range of phonetic detail. This
move, Kirchner demonstrates, is necessary in order to
account for the phenomenon of lenition, which requires
reference to articulatory effort, a non-contrastive
phonetic property. A major goal here is to unify the array
of processes that typically fall under the heading of
"lenition" through an effort-based approach. A detailed
analysis of Tmpisa Shoshone lenition, which includes
spirantization, voicing, nasal weakening, and elision,
shows that constraints on articulatory effort are necessary
in order to account in a unified manner for these different
4: Markedness, Segment Realization, and Locality in Spreading,
Mire N Chiosin and Jaye Padgett.
Synopsis: In this chapter, Padgett and N Chiosin
concentrate on the issue of locality in spreading, with an
empirical focus on Turkish vowel harmony. The theoretical
issue at stake in this chapter is strict locality, which
prevents feature spreading from skipping segments. For
vowel harmony, a consequence of strict locality is the
participation in vowel harmony of intervening consonants.
However, a problem that immediately arises from this
consequence concerns these participating consonants: if
these consonants are coarticulated with the vowel-harmonic
features, why don't the consonants with secondary
articulations occur contrastively in the language? Padgett
and N Chiosin show that under Dispersion Theory (DT;
Flemming 1995), this problem disappears, because of
constraints on both articulatory and perceptual markedness.
5: Austronesian Nasal Substitution Revisited: What's Wrong
with *NC (and What's Not), Joe Pater.
Synopsis: This chapter, comprising the first of three in
the second part of the book, concerns the content of
constraints. Specifically, Pater addresses the issue of
nasal substitution in Austronesian languages. The chapter
takes as a starting point earlier analyses (e.g., Pater
1999) of nasal substitution phenomena based on a constraint
penalizing nasal+voiceless obstruent sequences. Here, Pater
focuses on the Muna language, which shows evidence for a
richer set of constraints that account for the complex
array of processes associated with nasal+obstruent
sequences in the languages. The chapter provides convincing
evidence that rather than a simple ban on nasal+voiceless
obstruent sequences these processes are best accounted for
using constraints regulating the prosody-morphology
interface, and in addition posits a feature for voiced
obstruents that excludes both nasals and voiceless
6: A Critical View of Licensing by Cue: Codas and
Obstruents in Andalusian Spanish, Chip Gerfen.
Synopsis: Gerfen's chapter addresses what has become a very
heated topic in recent years: the issue of whether
contrasts are licensed by prosodic position or by phonetic
cues. This issue clearly relates to the content of
constraints; arguing against recent proposals by Steriade
(e.g., Steriade 1997), Gerfen's analysis of the licensing
of [s] in Eastern Andalusian Spanish requires reference to
syllable position. In a cue-based approach, generalizations
regarding the patterning of different clusters is lost,
claims Gerfen, thus rendering a cue-based approach
7: Segmental Unmarkedness versus Input Preservation in
Reduplication, Moira Yip.
Synopsis: Yip's chapter focuses on segmental alternations
exhibited by various patterns of reduplication in Chinese
languages, and puts forth the claim that the main drive
behind these alternations is markedness. Based on
contextual markedness constraints, the analysis recognizes
no special base-reduplicant faithfulness relation, so the
only correspondence relations are between input and output.
This chapter provides an interesting alternative to the
more standard approach to reduplication within OT (cf.
McCarthy & Prince 1995), and additionally supports other
recent proposals regarding reduplicative faithfulness, such
as that of Struijke (2001). Rather than being driven by a
base-reduplicant correspondence relation that demands
copying of base material (i.e., Max-BR), Yip's analysis
makes use of the two constraints Alliterate and Rhyme to
drive reduplicative identity.
8: Local Conjunction and Extending Sympathy Theory: OCP
Effects in Yucatec Maya, Haruka Fukazawa.
Synopsis: Fukazawa's chapter extends sympathy theory (cf.
McCarthy 1999) to OCP-based patterns in Yucatec Maya.
Originally designed to handle cases of derivational
opacity, sympathy theory is used here to account for a
phenomenon that would not be classified as a case of
derivational opacity in a serial approach. Additionally,
Fukazawa shows that local constraint conjunction is
required in order to capture certain cooccurrence
restrictions. Although the analysis is based on only six
pieces of data, the discussions regarding the need for
conjunction and for sympathy exemplify the use of these
important developments in OT.
9: Structure Preservation and Stratal Opacity in German,
Junko Ito and Armin Mester.
Synopsis: Based on cases of opacity in German, Ito and
Mester's chapter argues for a particular (and novel)
implementation of OT. Specifically, Ito and Mester
demonstrate the need for Weak Parallelism, a model which
recognizes a lexical and postlexical level and thus
dispenses with the strict parallelist single-level version
of OT. Rather than handle opacity with an opacity-specific
mechanism (such as sympathy theory), the analysis here
proposes a separation between the lexical and postlexical
modules which interact to produce stratal opacity.
This book is a nicely composed collection of papers on
current and relevant topics within Optimality Theory. The
organization into three distinct domains (representations,
constraints, and the specific issue of opacity) is well-
motivated and provides a solid foundation for the papers it
contains. Linguists at the graduate level should find it
accessible, and will probably be able to discern from the
papers important unresolved questions, so its value for
furthering research in OT is clear. The papers in the book
present an appreciable variety of approaches, from the
phonetics-based paper by Kirchner as well as that by
Padgett and N Chiosin, to the Weak Parallelism proposed
in Ito and Mester's paper. These papers represent
approaches that are less conventional within what one might
call classical OT, giving this collection the advantage of
containing much food for further thought.
Flemming, Edward. 1995. Auditory Representations in
Phonology. Ph. D. Dissertation, UCLA.
McCarthy, John. 1999. Sympathy and phonological opacity.
McCarthy, John, and Alan Prince. 1995. Faithfulness and
reduplicative identity. In Jill N. Beckman, Laura Walsh
Dickey, and Suzanne Urbanczyk, eds., University of
Massachusetts Occasional Papers 18: Papers in Optimality
Pater, Joe. 1999. Austronesian nasal substitution and other
NC effects. In R. Kager, et al., eds., The Prosody
Morphology Interface. Cambridge: Cambridge University
Prince, Alan, and Paul Smolensky. 1993. Optimality Theory:
Constraint interaction in generative grammar. Ms., Rutgers
University and University of Colorado, Boulder.
Steriade, Donca. 1997. Phonetics in phonology: The case of
laryngeal neutralization. Ms., UCLA.
Struijke, Caro. 2001. Existential Faithfulness: A study of
reduplicative TETU, feature movement, and dissimilation.
Ph. D. Dissertation, University of Maryland.
ABOUT THE REVIEWER
Adam Ussishkin is a Visiting Assistant Professor in the
Department of Linguistics at the University of Arizona.
His research centers on the interaction between phonology
and morphology and OT-based implementations of this