Veena Dixit, Center For Indian Language Technology, Department of Computer Science and Engineering, Indian Institute of Technology, Bombay.
This is the tenth volume in the series 'Interface Explorations' edited by Artemis Alexiadou and T. Allen Hall. The volume contains a collection of eleven articles with a scholarly introduction. The articles are based on the talks given at the workshop 'Feature Distribution in the Noun Phrase', part of the 24th annual DGfS conference in Mannheim (February 27 - March 1, 2002) and at the workshop 'Inflectional Paradigms: Primitives and Structures' at IDS, Mannheim (May 23-24, 2003). The articles draw examples from over 45 typologically diverse languages.
As the nominal inflection is considered in this volume to be an interface between morphology and syntax, the articles remain focused either on the syntactic component or on the morphological component. The major morphosyntactic features, namely, case, gender, inflection class and number are discussed. The emphasis changes from article to article. A number of theoretical approaches are discussed. They are, Distributed Morphology, Optimality Theory, Minimalist Program, Lexical Approach, Inferential Approach, and mixed Lexical-Inferential Approaches.
EVALUATION BY PAPER
Introduction, Gereon Müller, Lutz Gunkel, & Gisela Zifonun: The introduction opens with examples of determiner inflection in German and noun inflection in Russian & German. This leads to a discussion on nominal inflection as an interface phenomenon focusing on inflection classes along with the syncretism and underspecification in German language. The issues regarding morphosyntactic features such as, linkage between morphology and syntax in nominal inflection and how morphology has direct impact on syntax are discussed. Theories of Inflection are put in four major classes based on the views towards the inflectional markers and the information rendered by them. There are Incremental Theories, Realizational Theories, Lexical Theories and Inferential Theories. Different approaches like Minimalist, Optimality are discussed with reference to the features based on above referred four classes of inflectional theories, null suffix, position of inflectional markers in syntax and status of such markers in lexicon. The notion of paradigm as seen in various approaches is delineated to supplement this discussion. The introduction ends with a brief note on the contributions made to this book.
Comment: The authors have not discussed theory of Whole Word Morphology in the context of the classification of Inflectional Theories. Whole Word Morphology has its application in the field of unsupervised Natural Language Processing Tasks.
Inflectional Class, Gender and DP Internal Structure, Artemis Alexiadou: This paper discusses the relations between nominal features like gender and inflection class. The question whether these give rise to a functional projection in the extended context of the noun or not is scrutinized. The writer poses that the gender and the inflection class features are not syntactically active. He argues that phenomenon such as noun movement and noun ellipsis in his view appears to be related to the status of gender and class markers. Noun ellipsis seems to be related to the presence of morphological gender agreement between nouns and the adjectives modifying them. In the course of the argument he considers different parameters like ellipsis, class markers, gender, number, case, adjective-noun agreement and theme vowel. The argument is supported with the examples from Spanish, Greek, Italian, Hebrew and French.
Inflection Classes and Economy, James P. Blevins: The basic issues regarding inflection class raised in this paper are as follows. * Which elements of a morphological system are assigned to inflection classes? * Which principles govern class assignment? * How are classes distinguished? * Is there any bound on the number of possible classes within a given system? * Why do these classes play no role in agreement or other grammatical process?
In the word-and-paradigm model, full word forms are assigned to inflectional paradigm, which are in turn grouped into inflection classes. The attempt is to locate the answers in terms of 'economy', word and paradigm economy, affix and paradigm economy, and lexical economy. The economy of an inflectional system rests on patterns of interpredictability. The 'Paradigm Economy Principle' and 'No Blur Principle' are targeted in the discussion. These principles account for the relation between the 'leading form' and other members of the paradigm. The paper concludes that there is no need for dedicated stem-based or affix-based economy principles; recognition of words and paradigms as the basic components of a morphological system can resolve the problems arising from shared sources.
Left of Number, Animacy and Plurality in German Nouns, Peter Eisenberge & Ulricke Sayatz: The paper is descriptive in nature. It talks about the order of suffixes in the maximally inflected form. It establishes the hierarchy of suffixes in German Language as follows. (verb / noun stem-masculine/feminine/neuter-± animacy) > (-er/-ler- masculine- +agent) (-ling/-in- masculine/neuter- +sex-marked) (- schaft/-tum- feminine/neuter- +collective) > (-chen/-lein- neuter- +diminutive) (plural markers- no gender- +plural).
Verb and noun stems are referred as 'gender-inherent' while plural markers are marked as 'gender-determined' and others are called as 'gender-determiner'. Similarly the hierarchy of abstract nouns derived from verbs as well as adjectives is established. Further, the quest on whether these hierarchies are structured only by plurality is examined. The interdependence of countability and plurality is demonstrated in terms of semantic features [± bounded] and [± internal structure]. In short, plurality value is judged with respect to the Animacy hierarchy.
Comment: A more detailed discussion about the nature of abstract nouns and the way they are related to plurality would have been apposite.
Feature Sharing in DPs, Peter Gallmann: The paper attempts to prove the validity of the following hypotheses. Firstly, 'Syncretism in affix paradigms is the result of two independent regularities'. The first regularity consists in that the constraints determine the features which may be combined in phrase heads. The second regularity would be that the formal expression of the available feature bundles obeys the Principle of Maximal Paradigmatic Contrast. The paper is chiefly concerned with markedness and faithfulness constraints on morphosyntactic features. The manner in which the affix paradigms are stored in the mental lexicon is considered as the base for the Principle of Maximal Paradigmatic Contrast. Secondly, the distribution of the strong and weak declension of German adjectives depends on the percolation of case features in the NP, and so does the distribution of noun forms with and without case suffixes. The paper demonstrates the interaction of co-occurrence constraints and percolation constraints. The writer foresees the 'visibility constraints' determining morphosyntax of DPs which are not discussed in the paper due to 'space constraint'.
A Typological Schema for Noun Phrases in German, Pawel Karnowski & Jürgen Pafel: It is correctly stated that typology based discussion provides an opportunity to bring to the fore elementary questions without committing to any particular syntactic framework. The issue 'What is typology' and the related terms like schema, slot, field and position are explained before the discussion commences. It introduces the typological schema, (Z Def. X Nom Y), for noun phrase in German. The number and the nature of the slots in a typological schema are determined by the restrictions the elements filling the slots have to comply with. The restrictions concerning Z (field), Def. (position) and Noun (position) are formulated as three generalizations. The schema and the restrictions are further discussed in detail with exhaustive illustrations. The paper acknowledges that the typological theory of German noun phrases is inspired by the typological theory of German sentences. The proposition advanced is that a noun phrase in German is well formed with respect to word order if and only if it is constructed in accordance with above referred schema and its restrictions. The writers propose to supplement this schema in future with the suitable restrictions for the X-field and the Y-field and make the theory descriptively more adequate.
On Decomposing Inflection Class Features: Syncretism in Russian Noun Inflection, Gereon Müller: Inflection class features are more abstract and differ from other morphological features. These features, unlike gender, number and case, are not grounded in any way, they are not independently motivated and they do not seem to play any role in syntax. There is a general tendency to either avoid inflection class features in analysis of noun inflection or to accept them as an imperfection in grammar design. The writer refuses to agree with either of the views. He argues that inflection class features can be decomposed into more primitive features. Such decomposition offers explanation for instances of trans-paradigmatic syncretism. The decomposed inflection class features play a role in morphology that is analogous to the role played by uninterpretable features in syntax. Being uninterpretable in the syntactic component, inflection class features drive morphological operations that delete the class features before syntax is reached. Hence these features are not instances of imperfection. He claims that this reasoning provides an argument against late insertion in a minimalist grammar. The writer presents the main paradigms of noun inflection in Russian and shows that inflection class features are necessary to account for them. He further illustrates the decomposing of both case and inflection class features into more primitive features. The primitive features like ([± subj(ect)], [± gov (erned)], [± obl(ique)]) for case features and ([± α] and [± β]) for inflection class features are proposed. This is claimed to be useful to account for most of the instances of intra-paradigmatic and trans- paradigmatic syncretism in interaction with the Specificity condition. He argues that inflection class features can be seen as the triggers of inflection with portmanteau markers.
Comment: The primitive features used to decompose inflection class features are more abstract than those used for case features.
A Factorial Typology of Number Marking in Noun Phrases: The Tension between Economy and Faithfulness, Albert Ortmann: This paper is the result of work done on capturing the various empirical and theoretical aspect of number marking restricted by language economy. It aims to establish a factorial typology of plural morphology in the framework of Optimality theory. This paper highlights the typological aspects of plural marking in the DP. The major parameters that the factorial typology accounts for are as follows. * DP-internal number agreement vs. non-redundant plural marking as the result of the relative ranking of MAX constraints and an Avoid constraint, * Lexical vs. phrasal plural markers, * Sensitivity to such concepts as specificity and animacy in DP plural marking.
The parameters that determine cross-linguistic variation with respect to the distribution of number markers over the constituents of the noun phrase are described. After introduction of constraints on the economic number agreement approach, analysis of the above mentioned main parameters of variation is developed. Discussion is supported with examples from Hungarian, Georgian, German, Tagalog, Persian and Turkish. An overview of the subtypes that result from the various possible rankings of the involved constraints is provided. The paper ends with a summary of partial rankings established for the languages analyzed in the paper.
Feature Checking, Case, and Agreement in German DPs, Wolfgang Sternefeld: The paper aims to reanalyze the phenomenon that certain morphological case markings of German nouns can or must be dropped under specific syntactic conditions. The phenomenon is studied in terms of Optimality Theory as well as (Minimalist) Checking Theory. It concludes that the observed phenomena are more adequately described in terms of the conditions regarding the shape of certain suffixes than in terms of their grammatical contents. This paper talks about the other side of the issue.
Feminine vs. Non-Feminine Noun Phrases in German, Rolf Thieroff: This paper examines and confirms the Gallmann's Suffix Corollary (German),Nominal words are underspecified with respect to case unless they are preceded by an adjectivally inflected word-form with case suffix within its DP. This is supported by extensive examples from the language. Further, the predominance of Feminine morphology in German language is shown while considering 'Functional Verb Constructions', 'Measure Constructions', and Prepositions Governing the Genitive and the Dative'. The dominance of Feminine morphology extends to syntax of noun phrase in general. This is examined and confirmed considering Accusative and Dative Singular, Genitive Singular and Dative Plural. The reason for the dominance of Feminine morphology is located in its weakness, i.e., the feminine noun phrase cannot mark case on the noun. It is accepted that there is a possibility that the opposite process exists, i.e., the adoption of non-feminine morphology by feminine morphology as in the instance of genitive marker-s overtaken by feminine proper nouns from masculine nouns. Categories and Paradigms.
On Underspecification in Russian Declension, Bernd Wiese: An uneconomic complexity of homonymous endings is claimed as an artifact of uneconomic descriptions. It is argued that the distribution of markers over forms or cells in paradigms is structured. A tangled web of many-to-many relations between form and function can be reduced to a rather well organized common structure that underlies declensional paradigms. The categorization can be assigned to the endings to account for their functional unity and for their diversity of application. Categorization refers to bundles of categories. The approach is classificatory than being feature-based. A brief outline of an analysis of Russian pronominal declension based on a conception of underspecified paradigms is provided. The conception developed is extended to nouns. A brief on types of syncretism focuses on the Russian genitive-accusative. Instead of dealing with syncretism in terms of a combinatorial system of syntactic or semantic features, the present investigation is based on a detailed inspection of formal marker, i.e., endings, as they are used to distinguish word forms of paradigms. A limited inventory of pairs of inflectional endings and categorizations is established.
Is There Any Need for the Concept of Directional Syncretism?, Dieter Wunderlich: The directional syncretism is favored as it can be formalized by a grammatical rule (rule of referral) that operates on paradigms. Wunderlich doubts whether directional syncretism is an adequate concept. He claims that rules of referral are undesirable for theoretical reasons. It is shown that A/N syncretism is most naturally captured by underspecification and the syncretism, which is subject to animacy (accusative-nominative of inanimate nouns), can be captured by the interaction of lexical entries with well motivated constraints. G.sg/N.pl syncretism of Russian nouns is captured as symmetrical. The paper concludes that all syncretism should be regarded as a matter of lexical information supplemented by a set of constraints that determine the choice between inflected forms. The concept of lexical economy makes it plausible that independent grammatical features may share their exponents.
Comment: By making available the dimension of acquisition to understand syncretism makes the study multifaceted.
The book is primarily meant for research scholars, working in the area of inflectional morphology. A good grounding in the subject is a prerequisite for reading. The technical issues are handled in a scholarly manner. The focus is on inflectional classes and their properties. It is studied from various angles offering various answers to the question. Many of the issues raised in the book merit further discussion. Each paper contains an exhaustive list of references which can serve as a resource for other researchers. The reader may well wish that he should have been a participant of the base workshops and the discussions.
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Veena Dixit is engaged in research on less-studied and resource-poor language, Marathi, the state language of Maharashtra State of India. She is a significant contributor to the development of Morphology Rule- based spellchecker for Marathi. At present she is working on Rule- based Part-of Speech Tagger for Marathi. She has presented her work in national and international conferences.