Date: Mon, 19 Dec 2005 14:46:14 -0800 (PST) From: Olya Gurevich Subject: Morphology and the Web of Grammar: Essays in Memory of Steven G. Lapointe
EDITORS: Orgun, C. Orhan; Sells, Peter TITLE: Morphology and the Web of Grammar SUBTITLE: Essays in Memory of Steven G. Lapointe SERIES: Stanford Studies in Morphology and the Lexicon PUBLISHER: CSLI Publications YEAR: 2005
Olya Gurevich, Department of Linguistics, University of California, Berkeley
This volume is a collection of papers resulting from a workshop on morphology in memory of Steven Lapointe, held at UC Davis in April 2000. Lapointe, who passed away in 1999, worked on a variety of topics in theoretical linguistics, with a major focus on morphology and its relation to syntax and phonology. The papers in this volume relate to the wide variety of Lapointe's research interests. Some of the papers were presented at the 2000 workshop; some describe further developments on presented work; finally, some papers by Lapointe's close colleagues were added later.
The volume contains a preface by Lenora Trimm, Chair of the Linguistics department at UC Davis; a short introduction by the editors, C. Orhan Orgun and Peter Sells; eleven papers or varying lengths; and a full list of Steven Lapointe's publications. The papers are ordered roughly by linguistic subfield, from morphophonology to morphological expression and morphosyntax, to syntax and semantics.
Larry M. Hyman and C. Orhan Orgun, in ''Endocyclicity and Paradigm Non-Uniformity'', argue that cyclic phonology does not necessarily result from constraints on paradigm uniformity, i.e. the tendency of morphologically related word forms to be phonologically similar. They examine morphologically conditioned consonant mutation in Bantu, triggered by the addition of a causative suffix but not by the applicative morpheme. When both morphemes are present, different groups of Bantu languages demonstrate different phonological effects; in particular, one group ends up with a consonant that is different from the mutated consonant, but not the same as the underlying consonant. The authors argue for a cyclic analysis of all three types of languages within Paradigmatic Sign-Based Morphology (Dolbey and Orgun to appear, Orgun 1996), and suggest that cyclic phonology may disrupt paradigm uniformity.
''Remarks on Gerunds,'' by James P. Blevins, concerns so-called hybrid categories, like gerunds in English. He suggests that gerunds should be treated as lexical entries underspecified for word class properties. As a result, no special category is necessary to represent mixed cases; rather, structures that incorporate them license the appropriate interpretations. Blevins takes an intuition expressed in Chomsky (1970), expands it and updates earlier analyses, especially one by Pullum (1991). The advantage of the underspecification treatment is that it allows a unified representation of gerundive and derived nominals. The same approach is then expanded to analyze Welsh deverbal nouns.
In ''Rules about Paradigms,'' Gregory T. Stump shows that paradigms need to be an explicit part of a language's grammar, not just an epiphenomenon resulting from the application of morphosyntactic rules. He presents two types of rules that seem to refer explicitly to inflectional paradigms. The first is ''rules of abstraction'', where the realization of some paradigmatic cells is determined by other paradigmatic cells, regardless of the specific morphology of the forms in them. Several kinds of abstraction rules are presented, with examples from Sanskrit. The second type concerns interactions between ordered rule blocks: the particular interaction of rule blocks in one part of a complex inflectional system may explicitly refer to the rule-block interaction in a different part of the system. Since rule blocks make explicit reference to paradigms, this type of rule-block interaction also supports the explicit need for paradigms. The discussion is framed in terms of Paradigm-Function Morphology (Stump 2001). Finally, the author makes a plea for a theory of morphology that has separate principles from theories of syntax.
Jerry Sadock's ''Optimal Morphology'' makes an argument for autonomous morphology in which inflectional forms compete to satisfy constraints imposed by various syntactic, semantic, and morphological factors rather than being deterministically constructed by the syntax. The argument is based on an Internet exploration of case marking on conjoined pronouns ''you and I'' vs. ''you and me''. The suggested implementation is a sort of stochastically-based optimality theory, although the author does not present a specific formalism.
Andrew Spencer, in ''Towards a Typology of 'Mixed Categories,''' examines possible relationships between lexemes: morphological, syntactic and semantic. He suggests that most often, the relations between the three layers in a lexicon are one-to-one, so that a single morphological class (e.g. adjectival inflection) corresponds to a single syntactic class (modifier) and a single semantic class (denoting a property), but that there can exist mismatches in all possible combinations of properties. He presents examples of most types of mismatches, relying on various types of transpositions and unusual derivations. The major theoretical point is that languages can 'subvert the canonical mappings' between morphology, syntax, and semantics, in many possible ways, and these possibilities must be key in describing the lexicon of a given language.
''Dual Lexical Categories and Inflectional Morphology'' by James Yoon argues that inflected forms and mixed lexical categories are fundamentally alike in that the features determining phrase-internal syntax differ from the features determining external syntax. This fact is fairly well established for mixed categories. For inflectional categories, Yoon presents arguments from Korean and Salish to demonstrate that features of the root are responsible for internal syntax, whereas inflectional features are responsible for external syntax, and the two may be independent. Yoon examines these two types of categories within Lexicalist approaches to morphology (such as Pollard and Sag 1994), where they are treated as fundamentally different, and suggests an extension to Lapointe's Dual Lexical Category Theory (DLC; Lapointe 1993; 1999) so as to accommodate inflectional morphology.
Patrick Farrell's ''Prepositional Small Clauses in English: A Dual- Category Analysis'' looks at English phrases like 'the pope in a bikini' and argues that such phrases are subject-containing small clauses headed by a preposition. The author presents evidence that the small clauses behave like DPs with respect to the rest of the sentence, and like subject-containing PPs internally. He suggests an analysis along the lines of Lapointe's DLC, arguing that the heads of such small clauses are prepositions (P) with respect to internal syntax, and determiners (D) with respect to external syntax.
''Morphological and Constructional Expression and Recoverability of Verbal Features'' by Peter Sells deals with the auxiliary 'ha' ('have') in Swedish, which in some contexts is optional. A careful examination of such contexts reveals that 'ha' can be omitted when some other verb in the clause overtly expresses finiteness, or when finiteness can be recovered from context, such as from the presence of certain auxiliaries or nominative subject pronouns. An analysis is presented within Optimality-Theoretic Lexical-Functional Theory (Bresnan 2000, Kuhn 2001).
Almerindo Ojeda and Tamara Grivicic, in ''The Semantics of Serbo- Croatian Collectives,'' discuss a regular class of nouns in Serbo- Croatian that have separate forms for singular and plural individuals, and singular and plural collections of individuals. The semantics of such nouns are analyzed within a model-theoretic approach developed in (Ojeda 1993). The basic claim is that the sets of possible individuals and their combinations into groups form a mereological lattice, and the word forms pick out different sub-parts of this lattice.
Greg Carlson's ''When Morphology . . . Disappears'' examines several cases in which otherwise expected morphological markers are absent, e.g. lack of articles, case-marking, and doubled clitics. The contexts with missing morphology involve weak indefinite or generic interpretations of nouns. The author proposes a multi-level semantic analysis in which the interpretation of weak indefinites can happen at the initial, lexical levels of meaning assignment, whereas the interpretation of specific and definite noun phrases requires more contextual information and thus must happen at the later, sentence- related levels. Thus, there is no reason to assume that the absence of morphology carries meaning; rather, weak indefinites correspond to a ''zero'' meaning, and definites receive additional meaning carried by overt morphology.
In ''The Puzzle of Ambiguity,'' Thomas Wasow, Amy Perfors, and David Beaver examine the relation between linguistic theories and ambiguity. They suggest that despite being extremely widespread, ambiguity in natural language is not accounted for in any obvious way by existing theories of linguistic analysis, language use, evolution, or processing. They evaluate several speculative suggestions for the communicative function of possible sources of ambiguity, and come to the conclusion that it remains a very difficult problem, and much more research is needed to place it properly within the study of natural language.
The papers in this volume represent a wide spread of linguistic topics, reflecting the diversity of Steven Lapointe's interests. The quality of the papers is almost uniformly high, and many of the contributors are researchers well-known in their fields. A common thread in many of the papers is that morphology has separate principles of organization from syntax or phonology.
Thematically, the largest group of papers concerns ''dual'' (or ''mixed'') lexical categories, reflecting one of Lapointe's most important contributions to the discipline (Lapointe 1993, 1999). It is perhaps useful here to summarize the main point of Lapointe's theory. He looks at cases, such as gerunds, that represent mismatches between phrase-internal and phrase-external syntax, e.g. ''Their reading the paper was unexpected.'' The analysis involves dual categories , where X and Y are major lexical categories, such that X determines the external syntactic distribution of the phrase it heads, and Y determines internal syntactic properties of that phrase. Thus, English gerunds are represented as , behaving as nouns externally and as verbs phrase-internally.
Blevins deals with some of the same data as Lapointe and expresses a similar intuition that categorial hybrids fulfill multiple functions within a single phrase. However, his well-argued analysis involves under- specification rather than over-specification of the properties of such hybrids. Spencer, on the other hand, views the idea of dual, or mixed, categories as somewhat misguided, because it reifies the notion of 'category' and makes it sound more stable than it really is. He argues that what we think of as part-of-speech categories is simply one of the ways in which different levels of grammar (morphology, syntax, and semantics) can be mapped onto each other. However, he does not specify how such possible mappings may be implemented in a grammar, so it is conceivable that Blevins' under-specification analysis of gerunds could fit into Spencer's expanded typology.
The papers by Yoon and Farrell represent extensions of Lapointe's DLC theory. Farrell argues convincingly that prepositional small clauses have a DP-like external distribution and suggests that they are headed by prepositions. From here, it is easy to see how an analysis of prepositions as categories would work, although the arguments that the clauses are, indeed, headed by prepositions seem to depend crucially on the assumptions of X-bar theory, and more theory-independent evidence would be welcome. Yoon's suggestion that inflectional morphology is a case of mixed categories is interesting; however, his arguments against existing Lexicalist views of inflection seem somewhat problematic. He assumes a very simple version of the Head-Feature Principle in which all properties of the head must be passed up to the higher node. However, it is easy enough to imagine that only some of the features of a word (i.e. just the inflectional features) are passed up to determine external syntax, while only some of the features (i.e. just the root features) are used to determine internal syntax. Lapointe's DLC theory was designed primarily for lexical categories; it seems that modifying it to work for inflection is a radical step that may not be necessary given a sufficiently sophisticated view of the Lexicalist proposals.
Two of the papers touch on paradigms. Stump provides very convincing arguments for the inclusion of paradigm structure in grammar. Unfortunately, some of his arguments are a bit difficult to follow without knowing the details of Paradigm-Function Morphology (Stump 2001). Hyman and Orgun suggest that not only does grammar include paradigms, but it also explicitly specifies their structure. Although this was not a major part of their paper, some explanation as to why this should be, and what kinds of constraints grammar may put on paradigms, would have been a good addition.
Several papers concern Optimality Theory and, more generally, constraint satisfaction and ranking. Sadock, dealing with morphological choice, suggests that linguistic forms must comply with a set of probabilistically ranked constraints on form, syntax, semantics, etc. He makes a convincing argument that there are different kinds of pressures acting on a speaker when making this choice. This approach, however, needs more elaboration: it is unclear exactly how a constraint-ranking analysis would work and what role is given to individual speaker parameters such as dialect, speech register, etc. Sells analyzes optionality as two possible constraint rankings: one in which the auxiliary is omitted, and one in which it is present. However, it is unclear when speakers prefer one ranking over another, i.e. whether there are additional discourse factors affecting the choice, or whether some speakers omit the auxiliary more often than others.
The papers on semantics round out the volume nicely, although they seem somewhat disconnected from the rest of the papers and do not display the same sophistication as others, especially in morphological or syntactic matters. In Ojeda and Grivicic's analysis, it is unclear, given the denotational similarity between individualized plurals and collective forms, how the different word forms are semantically different. The assumption that morphological derivation must correspond exactly to semantic derivation also leads to a strange conclusion that the meaning of the root underlying all the count nouns is a mass noun. Carlson's analysis of missing morphology would benefit from a reference to other articles arguing in favor of paradigms, where the contrast between the presence and absence of morphology could automatically signal a contrast in meaning. While his suggestion that missing morphology corresponds to missing semantics is intriguing, there is too little data to support the overarching conclusions about how indefinites are marked morphologically or semantically.
Finally, some more general critical remarks are in order. It is stated in the introduction that some of the papers were presented at the 2000 workshop on morphology at UC Davis. However, it is often unclear from the papers themselves whether the same material was presented at the workshop, or whether they were included in the volume later. Such clarifications would have given a more accurate picture of the workshop. Also, there seems to be no interaction between the papers. Given that some of the authors attended the same workshop, and that many of the papers deal with similar topics, some amount of cross-reference would make the volume more coherent as a unit. In a similar vein, some of the authors present very technical analyses without explaining the necessary background for their respective theories. Given that this is a fairly broad-coverage collection of papers, it would have been nice to see more effort to make them accessible to the general reader.
Bresnan, Joan (2000) Optimal Syntax. In J. Dekkers, F. van der Leeuw, and J. van de Weijer, (eds.), Optimality Theory: Phonology, Syntax, and Acquisition. Oxford: Oxford University Press, 334-385.
Chomsky, Noam (1970) Remarks on Nominalization. In R.A. Jacoms and P.S. Rosenbaum, (eds.), Readings in English Transformational Grammar. Waltham: Ginn and Company, 232-286. [Reprinted in Studies on Semantics in Generative Grammar. The Hague: Mouton, 1- 61].
Dolbey, Andrew, and C. Orhan Orgun (to appear) Phonology- Morphology Interaction in a Constraint-Based Framework. In C. Reiss and G. Ramchand, (eds.) Handbook on Interface Research in Linguistics. Oxford: Oxford University Press.
Kuhn, Jonas (2001). Generation and Parsing in Optimality Theoretic syntax: Issues in the formalization of OT-LFG. In P. Sells, (ed.), Formal and Empirical Issues in Optimality Theoretic Syntax. Stanford: CSLI Publications, 313-366.
Lapointe, Steven G. (1980) A Theory of Grammatical Agreement, Ph.D. Dissertation. University of Massachusetts, Amherst.
Lapointe, Steven G. (1993) Dual Lexical Categories and the Syntax of Mixed Category Phrases. I In A. Kathol and M. Bernstein, (eds.), Proceedings of the Eastern States Conference on Linguistics. Dept. of Linguistics, Cornell University, 199-210.
Lapointe, Steven G. (1999) Dual Lexical Categories vs. Phrasal Conversion in the Analysis of Gerund Phrases. In P. DeLacy and A. Nowak, (eds.), University of Massachusetts Occasional Papers in Linguistics 24: Papers from the 25th Aniiversary. University of Massachusetts, Amherst: GLSA, 157-189.
Orgun, C.Orhan (1996) Sign-Based Morphology and Phonology with Special Attention to Optimality. Ph.D. Dissertation, University of California, Berkeley.
Pollard, Carl and Ivan Sag (1994) Head-Driven Phrase Structure Grammar. Chicago: University of Chicago Press.
Pullum, Geoffrey K. (1991) English nominal gerund phrases as noun phrases with verb-phrase heads. Linguistics 29: 763-799.
Stump, Gregory T. (2001) Inflectional Morphology: A Theory of Paradigm Structure. Cambridge: Cambridge University Press.
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Olya Gurevich is a PhD Candidate in Linguistics at UC Berkeley. Her research interests include morphology, syntax, and semantics; and cognitive and computational linguistics. She is currently completing her dissertation, "Constructional Morphology: The Georgian Version."