The articles assembled in this issue of the Canadian Journal ofLinguistics/Revue canadienne de linguistique contribute to ourunderstanding of the role of asymmetric relations at the interfaces. Asymmetricrelations have privileged status in the syntactic, phonological, andmorphological derivation of linguistic expressions (see for example the articlesin Di Sciullo 2003).

Interfaces are representations that must meet legibility conditions imposed byexternal systems. According to the Strong Minimalist Thesis (Chomsky 2001),language is an optimal solution to interface conditions, in that language is anoptimal way to link sound and meaning. Questions arise regarding the propertiesof the interface representations that make them optimally legible by externalsystems. These properties could very well be abstract, and remote from theperceptual systems, and could bear on the form of interface representations,rather than on the interpretation of their parts. A strong hypothesis in thisregard is that asymmetric relations are core properties of the relations derivedby the grammar (Chomsky 1981, 1995, 2001; Kayne 1994; Moro 2000; Di Sciullo2005; Zwart 2006). From this perspective, asymmetry is a pervasive property ofderivations and interface representations; it is thus expected to be a propertyof different structural relations, such as the relation between a displacedconstituent and its copy, the relation between an anaphor and its antecedent,the relation between a head and its dependent, and more generally, the relationbetween the constituents of a configuration.

For some adverbs, European and Brazilian Portuguese sharply differ, withconsequences for the syntax of tense, quantification, word order, quantifierfloating, wh-extraction, and clause structure. Evidence is provided for thearchitecture of the left, for remnant IP movement, and for the claim that“micro-” lexical differences between European and BrazilianPortuguese reduce to syntactic “macro-” variation: absence vs.presence of V-movement.

We argue that Agree is an asymmetric relation. According to Asymmetric Agree, ifx Agrees with y, then the features ofy must be properly included in the features ofx. If the links of a chain created by movement are relatedby Agree, this correctly predicts the properties of syntactic movement chains.And if the links of a chain created by movement are subject to the ProperInclusion Condition, that accounts for the (un)availability of extraction out ofa moved XP. The following distinctions interact with Asymmetric Agree: whethermovement is A or A-bar, whether or not the constituent extracted out of themoved phrase is selected by the head of that phrase, and whether movement isovert or covert. Syntactic movement is feature-structure preserving, in that theProper Inclusion Condition applies to sets of features in a particularconfiguration.

I examine scope interaction between a wh-phrase and a quantified phrase (QP).Sloan-type sentences in Japanese are also discussed, based on Motomura’sanalysis on zibun ‘self’. I show thatSloan’s original proposal is empirically inadequate, and propose ananalysis based on the Hornstein/Kayne-type analysis of bound pronouns. Oneconsequence is that scope determination must be understood to occurderivationally, based on the asymmetrical relationship between a wh-phrase and aQP at one point in the derivation. Also, only QPs collocated with an uncheckedCase feature can be a probe for Q-absorption.

I examine computational issues in the processing of SOV languages in theprobe-goal theory of the Minimalist Program. A theory that minimizes search,such as the probe-goal theory, provides a strong linguistic basis for theinvestigation of efficient parsing architecture. For parsing, two main designchallenges are presented: (i) how to limit search while incrementally recoveringstructure from input without the benefit of a pre-determined lexical array, and(ii) how to come up with a system that not only correctly resolves parsingambiguities, but does so with mechanisms that are architecturally justified. Itake as the starting point an existing probe-goal parser with features thatallow it to compute syntactic representation without recourse to derivationhistory search. I extend this parser to handle pre-nominal relative clauses ofthe sort found in SOV languages. I provide a unified computational account offacts on possessor (and non-possessor) relativization and processing preferencesin Turkish, Japanese, and Korean.

I consider the empirical domain constituted by word order alternations in theverbal domain in English and Japanese. Based on the observation of prosodicproperties of these alternations, a derivational analysis of the syntax-prosodyinterface is proposed in the context of a model of grammar in whichlinearization is determined by core syntax and the syntax-prosody interface.This approach crucially posits, and hence gives independent support for, twoauxiliary assumptions: incremental structure building in the grammar andmultiple spell-out into the phonological component.

I advance the thesis that the narrow syntactic component of CHL ispurely symmetrical, with all apparent asymmetries reducing to interfaceconditions. Properties of Germanic object shift and scrambling, in particularthe VO order-preservation (Holmberg’s Generalization), derive fromsymmetry-breaking strategies which dispose of superfluous (symmetric) syntacticinformation and ensure the PF-legibility (linearization). These interfacedesymmetrization strategies provide a Minimalist analysis of non-lexical(macro-)parameters, such as the head-directionality parameter.

The Linear Correspondence Axiom is reconciled with the VO/OV head parameter suchthat linear “shape” is preserved across the derivation. Thisallows object shift and scrambling to emerge as two sides of a single parametriccoin. The implications of this system for the A/A-bar movement typology areexplored, and a Minimalist reconceptualization is offered.

Recently, there has been a surge of psycholinguistic research on theinterpretation of type-shifting constructions. We evaluate some of thetheoretical and empirical research in this area and propose a reassessment ofthe nature of type-shifting operations.

We argue, in particular, for a structural gap in type-shifting constructions,motivated mainly by the restructuring nature of the verbs employed. We alsoargue that the putative underspecification (or indeterminacy) of type-shiftingconstructions is resolved via pragmatic inferences.

I present a derivational analysis of the construal of the focus structure of thesentence. I propose that the [+FOCUS] feature is an optional formal feature thatcan be assigned to multiple tokens of the numeration. Hence, the focus structureis derived via Merge (creation of a set) in a strictly compositional way. Thistheory circumvents the empirical and theoretical shortcomings of Nuclear StressRule-based approaches, and accounts for split focus constructions (answers tomultiple wh-questions).

I argue that syntactic structure encodes three types of asymmetries. The firstcorresponds to the asymmetry between mother and daughters nodes that is calleddominance, that is, syntactic hierarchy. The second is the selectional asymmetrybetween sister nodes, which is translated into the precedence relation in thephonological component. The third, called behindance, is analternative to dominance, and corresponds to parataxis. Parenthesis,coordination, and apposition are analyzed on the basis of behindance. In ourderivational model of grammar it is defined as a special type of inclusion thatblocks c-command. It follows that parenthetic material can neither move towardthe matrix, nor be bound by a constituent from the matrix.

This squib presents an exploration of how Exemplar Dynamics (Hintzman 1986;Goldinger 1996; Pierrehumbert 2001) can be used to model the increased phoneticvariability observed in language shift situations. It is based on a study oflaryngealized resonants, as pronounced by three fluent speakers ofSt’át’imcets, a Northern Interior Salish language ofBritish Columbia.