Conference report by
UC Santa Cruz
The 21st West Coast Conference on Formal Linguistics was hosted by UC Santa Cruz, April 5-7, 2002. The conference took place mainly at UCSC's Cowell College; the stunning view of Monterey Bay was only occasionally obscured by the typically foggy springtime weather. The conference banquet, held on a crisp, clear Saturday night, served also as an impromptu birthday celebration for UCSC professor Bill Ladusaw (a contributor to the first WCCFL (Ladusaw 1982) and many others since). With 141 registered participants, and a strong turn-out from UCSC undergraduate linguists, the event was bustling and intellectually stimulating.
Although the conference did not have an advertised theme, it turned out to be in large part an extended exploration of the interfaces. The Optimality Theory (OT) perspective, still dominant on the "P" side (and, increasingly, in acquisition and learnability), affords a natural interleaving of phonetics, phonology, and morphology. Most current views of syntax and semantics also encourage principles that reference categories and types, movement and lifting, etc. Notable also were the recent sophisticated appeals to intonational contours in syntactic and semantic analysis. As a group, WCCFL-21 presenters displayed a proficiency in multiple subdomains, an exciting development suggesting a trend towards an increasingly unified theory. It was also encouraging to see a rise in experimental results and corpus-based research.
I turn now to brief discussions of each of the papers presented during the conference, beginning with the three invited talks and then moving through the others in the order of presentation at UCSC. The program and the accepted abstracts are available at the conference website, http://ling.ucsc.edu/~wccfl-21.
WCCFL-21 had three invited talks. Joe Pater (UMass) and Caroline Heycock (Edinburgh) were invited by the conference organizers in an effort to highlight and promote the work of young, increasingly influential linguists. Irene Heim was invited to deliver the closing paper. She was a terrific choice for the 2002 WCCFL. Twenty years on, Heim 1982 remains a standard reference on most issues in semantics, and her subsequent work figures centrally in research on topics ranging from VP-ellipsis to presupposition projection (see, e.g., Heim 1983 in WCCFL-2). Additionally, her textbook with Angelika Kratzer (Heim and Krazter 1998) articulated the approach to the syntax–semantics interface that engendered much of the work reported on during the conference.
Joe Pater spoke at the end of the Friday sessions. His paper "Form and substance in phonological development" concerned the question of how the OT perspective can inform issues in child language and language acquisition. In particular, Pater sought an explanation in terms of constraint ranking for restrictions on young children's productive competence. The work continues the admirable tradition in OT of taking seriously issues of learnability and acquisition and their impact on grammar design. (See also the discussions below of Bruce Tesar's talk and also Meghan Sumner's.)
Caroline Heycock's "Topic, focus, and syntactic representations" (based on collaborative work with Anthony Kroch (UPenn)) was an innovative approach to the analysis of specificational copular constructions, including canonical copular constructions and their more exotic kin, specificational pseudoclefts and inverted specificational pseudoclefts. The nomenclature is cumbersome, but the proposals were not. Perhaps most valuable are the tools Heycock provided for exposing distinctions between (what appear to be) truth-conditionally identical representations. She made use of composition schemes and the structural properties of interpreted representations to achieve these results. The ideas synthesize, negotiate, and elaborate on the work of previous researchers in this area, including (but not limited to) Büring 1998, Sharvit 1999, Graff 2001, Mikkelsen 2001, and Steedman 2000. A related talk was given by David Adger and Gillian Ramchand.
Heim's "On the negation theory of antonymy" was also in part a synthesis, one with an apparently simple starting point: "What is antonymy?" Heim began with a review of existing approaches to gradable adjectives of both downward and upward polarity. As we have come to expect, the review was a refreshing, original perspective on the phenomena. With this unfamiliar view of the familiar as background, Heim used cross polar anomaly (#The jockey is shorter than the basketball player is tall), studied most extensively in Kennedy 1999, to develop a lexical decomposition analysis of predicates like short as little tall. The talk was innovative and empirically rich; it is likely to lead to an outpouring of work on this topic and related ones. Like Heycock, Heim acknowledged a debt to the existing literature, citing work by Kennedy in particular, and also Seuren (1978), von Stechow (1984), and others. (I recommend Graff 2000 as relevant to those intending to pursue this topic.) Heim's talk was notable for, among other things, its enthusiasm for tentative, original investigation using formal methodology — just the right message to send everyone home (back to work) with.
Daniel Büring (UCLA) began the conference with "Attributes stranded by A-bar movement", proposing a complex 'double-headed' analysis of German relative clauses like die Argumente, welche gegen seine Theorie vorgebracht wurden ('the argument, which against his theory adduced were'), in which an attribute (gegen seine Theorie) is separated from its associated nominal (Argumente). He argued that a double-headed analysis has the virtues of head-raising without the morphological and interpretative vices. His derivations involved remnant movement, a link supported by parallels between these relatives and the remnant movement constructions of Müller 1996. Alternative structures were dispatched on both syntactic and semantic grounds.
Benjamin Bruening (Delaware) used complex and novel data from Passamaquoddy (an Algonquian language of the northeast United States), mostly involving operator scope interactions and semantic variable binding, to argue for the position that the syntactic A/A-bar distinction is based not on relational positions but rather on feature-driven movement. In turn, the conflicting evidence that raising-to-object in Passamaquoddy has both A and A-bar properties was argued to reduce to the fact that there are two different structures in play here.
Alexander Williams (UPenn) and Jeffrey Lidz (Northwestern) closed the first session with "Reflexivity and resultatives". Their approach appealed to the denotation domains of nodes in trees and the nature of the semantic combinatorics to regulate the syntax — a surprising interaction, but one that, they argued, simply makes maximal use of standard assumptions about semantic composition. Theirs is not the first such proposal (for example, Fox (2000) appeals to nodes denoting in <t> to regulate his syntactic operation of quantifier raising), and they see potential applications for it elsewhere. In the question period, Williams characterized their hypothesis as an attempt to exploit the primitives of the theory to their full extent. It seems to me that this appropriately describes much of the interface work reported on during the conference.
The second morning sessions continued themes from the first. Peter Svenonius (Tromsø) and Gillian Ramchand (Oxford) offered an account of verb-particle constructions in Germanic and Scandinavian (and elsewhere) based on a projection within the lexical syntax, in the spirit of the lexical decomposition analysis of Hale and Keyser 1993, and extending some arguments in Svenonius 1994. Ida Toivonen's (Rochester) "Verbal particles and results in English and Swedish" served as a counterpoint to Svenonius and Ramchand's. The two papers overlap in their empirical domains, but differ markedly in their conclusions. Toivonen's deserves special mention as the only purely syntactic work of the weekend whose principles are stated in a grammar formalism (LFG), rather than just in a natural language.
Lotus Goldberg (McGill), in "An elucidation of null direct object structures in Modern Hebrew", contributed to our understanding of null direct objects. Her paper offered a method for distinguishing null direct object structures from those created by VP-ellipsis following movement of V out of the to-be-elided phrase (as proposed first for Irish by McCloskey (1991)). This done, she showed that null direct objects cannot be animate, and that VP-ellipsis in Hebrew requires strict verbal identity. Goldberg also used island-related facts to show that null direct objects are formed by A-bar movement.
In "Functional projections in Walpiri", Julie Anne Legate's (MIT) focus was the basic clausal design of Walpiri sentences. She argued that Walpiri is not a nonconfigurational language, appealing to a rich functional layer atop the clause to explain word-order patterns, which she correlates with those found in uncontroversially configurational languages. She concluded that the parallels indicate a hidden rigidity to the clausal organization of Walpiri, one that is captured by the 'exploded' functional layer of Cinque (1999) and others working in the Minimalist Program.
The first phonetics/phonology session of the conference began with Travis Bradley's (UC Davis) paper "Gestural timing and derived environment effects in Norwegian clusters". Bradley, in developing an OT theory, used a mix of phonetic and phonological principles to explain derived environment effects involving rhotic-consonant clusters in Urban East Norwegian.
In "Contour tone licensing and moraicity", Jie Zhang (Harvard) deployed evidence from a variety of typologically distinct languages for the position that contour tones licensing is not in any way determined by bimoraic syllables, contra much recent work on the topic. Instead, Zhang linked contour tones with the syllable rime, analyzed as the tone bearing unit in tone systems. Like Travis Bradley's, Zhang's proposal mixed phonetic and phonological principles in a single OT grammar to achieve a unitary account.
Maria Gouskova's (UMass) "Syllable contact as a relational hierarchy" was a revision of the standard 'two constraint' view of syllable contact laws. Gouskova argued that these laws are properly decomposed into a hierarchy of OT constraints with a fixed ranking, supplemented by Relational Alignment, new to this work, which merges onset and coda markedness scales. Relational Alignment is inspired by the Harmonic Alignment of classical OT, but is, Gouskova argued, superior in the area of syllable markedness constraints.
Mark Gawron (San Diego State) led off the second afternoon session with a paper by him and Andrew Kehler (UCSD), "The semantics of the adjective respective". Gawron and Kehler distinguished the adjective from the adverb respectively using a variety of contrasts, and then developed a formal semantics that captures these contrasts while maintaining the obvious links between the two forms. The proposal is informed by the semantics of plurality and appealed to pragmatically supplied sequencing functions to model distributive readings.
Daphna Heller (Rutgers), in "Possession as a lexical relation: evidence from the Hebrew construct state", offered evidence from Hebrew that, she argues, favors the proposal of Vikner and Jensen (2001) and others that possession is a lexical relation, not just one of many contextually supplied relations (Partee 1997). Her proposal was that a noun in the morphologically distinguished construct state denotes a function in <e,e>, taking the genitive nominal as argument to yield an individual. In addition, she pursued the claim that lexical entries encode 'qualia structure', which expresses essential properties of objects (e.g., telicity, part–whole relations).
The final regular session talk on Friday was given by Brady Clark (Stanford), presenting a piece of joint work with David Beaver (Stanford). Entitled "The proper treatment of focus sensitivity", it was a part of Beaver and Clark's larger research program exploring the premise that focus sensitivity is regulated by both pragmatic and semantic factors. The empirical locus of the paper was mainly the contrasts between only, which they give a 'weak' (narrowly semantic) analysis, and always, which they give a 'strong' (pragmatic) account. The generalization was intended to include these morphemes' counterparts in other languages.
The comparatively small number of presenters at this year's WCCFL-21 meant that it was necessary to have just two parallel sessions. These were scheduled for Saturday; a high overall attendance meant that all the talks were well-attended.
Shinichiro Ishihara (MIT) began the Saturday semantics session with "Invisible but Audible Scope Marking: Wh-Constructions and deaccenting in Japanese". Shinichiro's is a well-supported account based on the interaction of syntax and prosody, and boasted experimental evidence for specific focus contours and their correlations with syntactic phenomena. His proposals are likely to inform research into islands effects and related phenomena in Japanese and perhaps more widely. In addition, Ishihara's evidence suggests that, if one accepts that intonation contours are properties of sentences, then utterances previously thought to correspond to multiple sentences are not in fact ambiguous in this way. (See also the synopsis of Shin-Sook Kim's "Focus matters: two types of intervention effect" given below.)
Jason Merchant's (Chicago) "PF output constraints and elliptical repair in SAI comparatives" continued the line of analysis featured in Kennedy and Merchant 2000 and Merchant 2001, which assumes that ellipsis is able to 'repair' otherwise ungrammatical structures. In this paper, Merchant concentrated on the obligatory VP-ellipsis that accompanies subject–auxiliary inversion in comparatives (Abby knows more languages than does her father (*know).). He also proposed to re-instate (a version of) the Empty Category Principle, now relativized to PF-head government (roughly, government by a pronounced head).
In their "Derivational parallelism and ellipsis parallelism", Ash Asudeh (PARC and Stanford) and Richard Crouch (PARC) used Glue semantics, which employs a fragment of linear logic, to analyze scopal interactions under ellipsis. Asudeh presented the paper, arguing that proof parallelism is the crucial notion for ellipsis parallelism. Their argument's main empirical interest is its discussion of the examples in Jacobson 1997 and Johnson and Lappin 1999 which suggest that surface scope is preferred, but not required, when one of two clauses related by ellipsis contains just a single scope-bearing element. Asudeh and Crouch's paper was also theoretically provocative: it is in part an argument for a distinguished level of semantic representation (Glue proofs). The paper was a refreshing take on an old, difficult problem, and it is terrific to see linguists exploring nonclassical logics, both for their inherent value and because the move could renew ties with theoretical computer science.
Saturday's opening phonology talk was given by Robert Kennedy (Arizona), who presented "A stress-based approach to Ponopean reduplication". Kennedy succeeds in modelling the challenging range of reduplicative morphemes in Ponopean with a single OT subgrammar. That grammar characterizes the shape of the reduplicant as an instance of the emergence of the unmarked, which brings this case in line with the OT expectation that this is the way base-reduplicant relationships work in general.
Following Kennedy, Nicole Nelson (UC Irvine) presented results supporting Positional Anchoring Theory via an appeal to the overwhelming tendency for reduplication to be prefixing ("Deriving the "prefixing" preference in reduplication"). The guiding intuition was that this should be assimilated to other positional effects.
In "A restricted model of UR discovery", Adam Albright (UCLA) used evidence from Lakhota to motivate the hypothesis that underlying representations (URs) are posited by language learners on the basis of unique surface alternants. Forms that deviate from this posited UR receive complex derivations (however those are formally conceived, as transformations or imperfect relations between URs and surface forms). Albright's arguments might interact significantly with those of Nathan Sanders, whose "Dispersion in OT: color contrasts in Middle Polish nasal vowel" is reviewed below. See also Meghan Sumner's experimental results, also discussed below.
Shin-Sook Kim (Universität Konstanz) investigated intervention effects (e.g., Nobody gave John/*most people a red cent in English, German, Korean. "Focus matters: two types of intervention effect" motivated the hypothesis that the class of interveners is characterized by focus-related properties of lexical items, rather than a quantificational semantics per se (as proposed by Linebarger (1987) and Beck (1996), among others). (For partially related ideas, see the discussion above of Shinichiro Ishihara's "Invisible but Audible Scope Marking: Wh-Constructions and Deaccenting in Japanese".)
Elsi Kaiser's (UPenn) "Case alternations, disjunction and questions", which began the morning syntax session, sought a unified account of the distribution of partitive and accusative case in Finnish and the way case-marking interacts with the two Finnish disjunctions, vai, which is restricted to alternative questions, and tai, which can appear in alternative and yes/no questions. Kaiser proposed that partitive case is anti-licensed by veridicality (Giannakidou 1999), thus properly correlating its distribution with negative polarity items. The limitation of vai to alternative questions is due to its presuppositional, and hence veridical, nature. If used in a yes/no question, it would invariably presuppose a "yes" answer, making it pragmatically anomalous (uninformative). Its incompatibility with partitive case follows from the antiveridicality of that case: their combination yields a veridicality clash.
Sei Rang Oh (UConn) developed a semantics for the English expression one at a time and its Korean counterpart han peney hanassik in "Decomposing one at a time". Oh analyzed the Korean expression as composed of a distributivity operator, hana-ssik ('one-DIST '), and the phrase han pen-ey ('one-time-at'). In addition, Oh used this construction to motivate the presence of event arguments in interpreted structures, and suggested a potential expansion of the analysis to include phrases like sea to shining sea. It would be worth seeing what modifications are required in order to obtain the semantic fact that this phrase means shining sea to shining sea, in which both occurrences of sea are modified by shining.
Running in parallel to the morning syntax talks was a session devoted to morphology, though, as expected, there was plenty overlap with adjacent domains, making this categorization slightly infelicitous. Loren Billings (Providence) and Abigail Konopasky (Duke) addressed the nature of the relationships between and among syntax, morphology and prosody in their paper "Morphology's role in ordering verb-adjacent clitics", providing evidence that clitic movement in Bulgarian and Tagalog is guided by different versions of the EPP property, located in TP. They treated Kayne's (1994) Linear Correspondence Axiom as a fixed premise, and noted that this combines with their other assumptions to entail a nonsyntactic linear ordering of clitics. See also Graham Horwood's OT based approach to morpheme order, described briefly below.
A welcome departure from the audiocentric nature of formal linguistics, the paper "Optimal reciprocals in German sign language", by Roland Pfau (Amsterdam) and Markus Steinbach (Mainz), described the multiple ways of expressing reciprocal relations in German Sign Language (DGS: Deutsche Gebärdensprache), a project they began in Pfau and Steinbach 2001. In some dialects, there are up to four modes of expression for reciprocals, each characterized by specific phonological or morphological properties of the signs. Pfau and Steinbach offered four ranked OT constraints to model these manifestations, thereby showing (perhaps unsurprisingly) that the morphophonology of sign languages is amenable to the same treatment as that given to spoken languages.
Graham Horwood (Rutgers) motivated the idea that morpheme order is specified in inputs, with surface ordering in turn determined by general well-formedness constraints. "Relational faithfulness and conflicting directionality" was an argument that this approach to linear order has desirable typological consequences and provides a unified account of phenomena like infixation and metathesis. For a different take on some similar problems involving clitics, see Loren Billings and Abigail Konopasky's paper, reported on just above.
In "A beauty of a construction", Ora Matushansky (MIT) offered a spirited and insightful analysis of the construction mentioned and used in her title (which, she reports, is crosslinguistically common). She explored the structure and derivation, with emphasis on the semantics of modification inside them (this oceanic barge of a woman, an attested example of hers). She ultimately concluded that the second nominal (a woman in the example) is the semantic head, and goes on to link this construction with degree nominals like How beautiful1 of a t1 construction, assigning them a parallel derivation with movement of NP rather than AP (a beauty1 of a t1 construction). Matushansky apologized that twenty-minutes would not permit her enough time to discuss the Romance analogue, but she managed to sneak in more than a few facts about the French version of this puzzle of a construction.
David Adger (York) and Gillian Ramchand's (Oxford) paper "Predication and equation" connected with Heycock's work on copular constructions, though the empirical domain was mainly Scots Gaelic rather than English. Ramchand delivered the paper, an analysis of predicational constructions that is based on a highly articulated functional layer both inside the predicate and atop the clause. Adger and Ramchand argue that there is just a single predication structure, and that it is both syntactically and semantically asymmetrical. The copula denotes a function from properties to functions from individuals to truth values, containing a primitive 'holds' relation.
The paper "Triggers and alternations in compensatory lengthening", by Darya Kavitskaya (Yale), argued from a broad typological investigation that compensatory lengthening of vowels and consonants are best handled by separate mechanisms in the synchronic grammar. Cases of opaque alternations are argued to be nonphonological. Instead, their sporadic nature suggests a lexicalization, a stance resembling that of Nathan Sanders, whose analysis of opaque alternations in the history of Polish is reviewed below.
Bruce Tesar (Rutgers) provided a way of measuring restrictiveness of OT grammars in terms of their r-measure, which is given as the sum of the number of markedness constraints dominating each faithfulness constraint. (Tesar calls this 'markoverfaithedness'.) This in turn leads to a method for reducing structural ambiguity, by appeal to a premise that says learners will favor the most restrictive grammar (highest r-measure) that is consistent with input data. Tesar's claims were supported by computational results; probing questions in the answer period revealed that he has a more developed theory than could be presented in a mere twenty-minutes. The work is a novel, impressively precise new entry in the distinguished OT tradition of fusing theoretical linguistics with learnability issues.
Nathan Sanders (UCSC) tackled opacity in Polish using the apparatus of Faithfulness, Dispersion, and Markedness in OT (FDM-OT), a variant of OT in which inputs and candidates are entire languages and dispersion constraints reference relative contrasts among segments, balancing articulatory and perceptual ease roughly in the way that OT balances markedness and faithfulness. Sanders proposed to maintain the consequence of OT that (most) opacity cannot be phonologically productive. Instead, he attains morphologically-conditioned opacity by appeal to historical change. His empirical focus was color (backness and roundness) contrasts in the nasal vowels of Middle Polish; the work provided a concise survey of the historical alterations these segments have seen, using this narrative to reconstruct the underlying representations that speakers are apt to posit. See also Adam Albright's "A restrictive theory of UR discovery: evidence from Lakhota" (for related ideas about underlying representations), and Darya Kavitskaya's proposals about opacity in compensatory lengthening (for a similarly lexical view of opacity), both briefly summarized above.
In "Ablaut as feature coalescence", Ji-yung Kim (UMass) addressed a challenging interface problem: in ablaut alternations, "the alternation itself is phonological, but [...] the conditioning of that alternation is morphological". Kim OT used phonological and morphological constraints to capture ablaut with a single constraint set while maintaining a distinguished morphological aspect to the alternations.
Daylight savings time began at 2 am Sunday morning. The program warned of the hour's loss, but the organizers still feared Sunday's sessions would suffer from sparse attendance, due to the hour's loss and the nature of Saturday night at these events. However, the fears were unfounded. Sunday's talks were well-attended; strong and timely coffee made for a surprisingly lively series.
Elliott Moreton (Johns Hopkins) and Paul Smolensky (Johns Hopkins) argued, in "Typological consequences of local constraint conjunction", that local conjunction should have a primary place in OT grammars. Moreton presented their arguments that this constraint type is useful for handling certain classes of opacity effect. The work offered a typology of chain-shift effects, and used it to show that, with this kind of opacity, OT actually excels in terms of specific analyses as well as typological predictions.
Evan Mellander (McGill) derived generalizations about the relative markedness of various rhythm patterns ("On rhythmic asymmetries in metrical groupings") from a set of ranked OT constraints that exploits the notion of domain size. Meghan Sumner's (SUNY) "The psycholinguistic reality of abstract representations" was a report on priming experiments done with native Hebrew speakers, the aim of which was to get at the nature of the underlying representations speakers posit. Her results yield insights into, among other things, the influence of orthography and the cognitive differences between transparent and opaque alternations in paradigms.
Jeffrey T. Runner, Rachel S. Sussman, and Michael K. Tanenhaus (all at Rochester), presented their experimental results concerning speakers' treatment of reflexives inside possessed nominals. Their results are based on eye-tracking experiments done at Rochester. In constructions like John saw Bill's picture of himself, their subjects allowed the matrix subject to antecede the reflexive in thirty percent of the cases, a significant departure from most statement of the binding theory. (However, as they note, there has always been skepticism about extending the binding theory to include genitives as subjects; see, e.g., Reinhart 1993:683.) Similar results were obtained by Asudeh and Keller (2001) using magnitude estimation methodology. Runner et al.'s theoretical conclusion is that some reflexives need not be bound, but rather can receive coreferential analyses; Asudeh and Keller (2001) reached closely related theoretical conclusions from their magnitude estimation data. (I offer here a bit of attested evidence for Runner et al.'s and Asudeh and Keller's conclusions: "C.B.'s father [...] still resented his wife for her low opinion of himself, [...]" (Richard Russo. Empire Falls, p. 4).)
A quite different take on reflexives was found in "Deriving reflexives", by Rose-Marie Dechaine (UBC) and Martina Wiltschko (UBC). They argued for a tripartite typology of reflexives: pro-DP, pro-phiP, and pro-NP, with these syntactic categorizations in turn determining what sort of binding relation the form can participate in. Evidence from English, French, Greek, Haitian Creole, Norwegian, Plains Cree, and Spanish bolstered this view — the typological survey was impressive. The central role of functional projections in this work recalls "Predication and Equation", Adger and Ramchand's talk, (partially) summarized above.
WCCFL officially came of age this year, its 21st, and with unqualified success. The overall quality of the presentations was at least as high as it has ever been, and the presenters' keen awareness for the diverse range of possible analyses bodes extremely well for the continued vitality of formal linguistics.
My thanks to Line Mikkelsen for reviewing an early draft of this report. And thanks also the WCCFL-21 contributors who provided input and advice: Ash Asudeh, Loren Billings, Brady Clark, Rose-Marie Dechaine, Maria Gouskova, Daphna Heller, Gillian Ramchand, Nathan Sanders, Peter Svenonius, and Jie Zhang. Any remaining mistakes or (perish the thought!) misrepresentations are my responsibility.
Asudeh, Ash and Frank Keller. 2001. Experimental evidence for a predication-based Binding Theory. In Mary Andronis, Chris Ball, Heidi Elston, and Sylvain Neuvel (eds.), CLS 37: The main session. Papers from the 37th. meeting of the Chicago Linguistic Society, vol. 1. Chicago, IL: Chicago Linguistic Society.
Beck, Sigrid. 1996. Quantified structures as barriers for LF movement. Natural Language Semantics 4:1 (1-56).
Büring, Daniel. 1998. Identity, modality, and the candidate behind the wall. In Proceedings of SALT 8. Ithaca, NY: Cornell University Linguistic Publications.
Cinque, Guglielmo. 1999. Adverbs and Functional Heads. Oxford, England. Oxford University Press.
Fox, Danny. 2000. Economy and Semantic Interpretation. Cambridge, MA: MIT Press.
Giannakidou, Anastasia. 1999. Affective dependencies. Linguistics and Philosophy 22:4 (367-421).
Graff, Delia. 2000. Shifting sands: an interest-relative theory of vagueness. Philosophical Topics 28:1 (45-81).
Graff, Delia. 2001. Descriptions as predicates. Philosophical Studies 102:1 (1-42).
Hale, Kenneth and Samuel Jay Keyser. 1993. On argument structure and the lexical expression of synactic relations. In Kenneth Hale and Samuel Jay Kesyer (eds.), The View from Building 20, 53-109. Cambridge, MA: MIT Press.
Heim, Irene. 1982. The Semantics of Definite and Indefinite Noun Phrases. University of Massachusetts, Amherst, MA.
Heim, Irene. 1983. On the projection problem for presuppositions. In Michael Barlow, Daniel P. Flickinger, and Michael T. Wescoat (eds.), Proceedings of WCCFL-2, 114-125. Stanford, CA: Stanford Linguistics Department.
Heim, Irene and Angelika Kratzer. 1998. Semantics in Generative Grammar. Malden, MA: Blackwell.
Jacobson, Pauline. 1997. Where (if anywhere) is transderivationality located?. In Peter W. Culicover and Louise McNally (eds.) The Limits of Syntax, 303-336. New York, NY:Academic Press.
Johnson, David E. and Shalom Lappin. 1999. Local Constraints vs. Economy. Stanford, CA: CSLI Publications.
Kayne, Richard. 1994. The Antisymmetry of Syntax. Cambridge, MA: MIT Press.
Kennedy, Christopher. 1999. Projecting the Adjective: the syntax and semantics of gradability and comparison. New York, NY: Garland Press. [1997 UCSC Ph.D. dissertation].
Kennedy, Christopher and Jason Merchant. 2000. Attributive comparative deletion. Natural Language and Linguistic Theory 18:1 (89-146).
Ladusaw, William A. 1982. Semantic constraints on the English partitive construction. In Daniel P. Flickinger, Marlys Macken, Nancy Wiegand, Proceedings of WCCFL-1, 231-42. Stanford, CA: Stanford Linguistics Department.
Linebarger, Marcia C. 1987. Negative polarity and grammatical representation. Linguistics and Philosophy 10:3 (325-387).
McCloskey, James. 1991. Clause structure, ellipsis, and proper government in Irish. In James McCloskey (ed.), The Syntax of Verb-Initial Languages, 259-302. Lingua special edition.
Merchant, Jason. 2001. The Syntax of Silence. Oxford, England: Oxford University Press.
Mikkelsen, Line Hove. 2001. Specification is not inverted predication. In Masako Hirotani (ed.), Proceedings of NELS 32, 403-422. University of Massachusetts, Amherst: GLSA
Müller, Gereon. 1996. Incomplete category fronting: a derivational approach to remnant movement in German. Dordrecht: Kluwer Academic Publishers.
Partee, Barbara. 1997. Genitives — a case study. In Johan van Benthem and Alice ter Meulen (eds.), The Handbook of Logic and Language, Appendix to Theo M.V. Janssen: Compositionality. Cambridge, MA: MIT Press.
Pfau, Roland and Markus Steinbach. 2001. A constraint-based analysis of reciprocals in German Sign Language. To appear in Proceedings of WECOL 2001.
Seuren, Pieter. 1978. The structure and selection of positive and negative gradable adjectives. In Papers from the Para-session on the Lexicon. 14th Regional Meeting of the Chicago Linguistic Society, 336-346. Chicago, IL: CLS.
Sharvit, Yael. 1999. Connectivity in specificational sentences. Natural Language Semantics 7:2 (299-346).
Steedman, Mark. 2000. Information structure and the syntax-phonology interface. Linguistic Inquiry 31:4 (649-690).
Svenonius, Peter. 1994. Dependent Nexus: Subordinate Predication Structures in English and the Scandinavian Languages. Ph.D. dissertation, UC Santa Cruz.
Vikner, Carl and Per Anker Jensen. 2001. A semantic analysis of the English genitive: interaction of lexical and formal semantics. Manuscript Copenhagen Business School and the University of Southern Denmark, Kolding, Denmark.
von Stechow, Arnim. 1984. Comparing semantic theories of comparison. Journal of Semantics 3:1 (1-77).