Interaction, Satisfaction, and the PCC

The person-case constraint (PCC) is a family of restrictions on the relative person of the two objects of a ditransitive. PCC effects offer a testing ground for theories of the Agree operation and of syntactic features, both those on nominals and (of special interest here) those found on agreement probes. In this paper, I offer a new theory of PCC effects in an interaction/satisfaction theory of Agree (Deal 2015a) and show the advantages of this framework in capturing PCC typology. On this model, probes are specified for interaction features, determining which features will be copied to them, and satisfaction features, determining which features will cause probing to stop. Applied to PCC, this theory (i) captures all four types of PCC effect recognized by Nevins (2007) under a unified notion of Agree; (ii) captures the restriction of PCC effects to contexts of “Double Weakness” in many prominent examples, e.g. in Italian, Greek, and Basque, where PCC effects hold only in cases where both the direct and indirect object are expressed with clitics; (iii) naturally extends to PCC effects in syntactic environments without visible clitics or agreement for one or both objects, as well as the absence of PCC effects in some languages with clitics or agreement for both the direct and indirect object. Two refinements of the interaction/satisfaction theory are offered. The first is a new notation for probes’ interaction and satisfaction specifications, clarifying the absence from this theory of uninterpretable/unvalued features as drivers of Agree. The second is a proposal for the way that probes’ behavior may change over the course of a derivation, dubbed dynamic interaction.

Extending the Person-Case Constraint to Gender: Agreement, Locality, and the Syntax of Pronouns

In many languages with clitic or other weak pronouns, a Person-Case Constraint (Perlmutter 1971, Bonet 1991) prohibits certain combinations of these pronouns on the basis of their person features. This article explores the crosslinguistic variation in such constraints, starting with several closely related Zapotec varieties. These restrict combinations of clitics not just on the basis of person, but also on the basis of a finely articulated, largely animacy-based gender system. Operating within a larger combinatorial space, these constraints offer a new perspective on the typology of Phi-Case Constraints (ΦCCs) more generally. This typology has an overall asymmetrical shape correlating with the underlying syntactic position of pronominal arguments. We develop a principled theory of this typology that incorporates three hypotheses: (a) ΦCCs arise from how a functional head Agrees with clitic pronouns, subject to intervention-based locality (Anagnostopoulou 2003, Béjar and Rezac 2003, 2009); (b) the variation in these constraints arises from variation in the relativization of probes (Anagnostopoulou 2005, Nevins 2007, 2011); and (c) clitic and other weak pronouns have no inherent need to be licensed via Agree with a functional head. Under this account, the crosslinguistic typology of ΦCCs has the potential to shed light on the grammatical representation of person and gender.

Feature Gluttony

This article develops a new approach to a family of hierarchy-effect inducing configurations, with a focus on Person Case Constraint effects, dative-nominative configurations, and copula constructions. The main line of approach in the recent literature is to attribute these effects to failures of φ-Agree or, more specifically, failures of nominal licensing or case checking. We propose that the problem in these configurations is unrelated to nominal licensing, but is instead the result of a probe participating in more than one Agree dependency, a configuration we refer to as feature gluttony. Feature gluttony does not in and of itself lead to ungrammaticality; rather, it can create irresolvably conflicting requirements for subsequent operations. We argue that in the case of clitic configurations, a probe that agrees with more than one DP creates an intervention problem for clitic doubling. In violations involving morphological agreement, gluttony in features may result in a configuration with no available morphological output.

Dissociating the Person Case Constraint from its “repair”

In French ditransitive sentences, certain person combinations of the two internal arguments cannot be expressed with two co-occurring clitics (a phenomenon referred to as the Person Case Constraint or PCC). To fill the interpretational gap created by this restriction, there is an alternative construction characterized as a “repair”, where the goal is realized as an independent phrase. The fact that the double-clitic construction and the repair construction are in complementary distribution led to a proposal of an interface algorithm that provides a way to repair a non-convergent structure. This article proposes an alternative account of the PCC, and claims that the complementarity between the PCC and its repair is instead accidental and is an artefact of the feature structure of arguments. The proposed account explains the unavailability of certain clitic combinations and some repairs independently, without resorting to a trans-derivational device like the previously proposed algorithm.

Person complementarity and (pseudo) Person Case Constraint effects: Evidence from Inuktitut

This paper examines the nature of person complementarity in Eastern Canadian Inuktitut (Eskimo-Aleut), arguing that despite its apparent patterning as a Person Case Constraint (PCC) effect, it is not due to the presence of a defective intervener blocking person agreement with a lower argument, as is often the case in other languages. Instead, the observed effect is caused by a defective or missing person probe on C that cannot value local person features on absolutive arguments. Given the use of the PCC as a diagnostic for differentiating clitics and agreement, this result has implications for the proper identification of φ-marking in Inuktitut.

Clitics and Clitic Clusters in Morphology

Clitics can be defined as prosodically defective function words. They can belong to a number of syntactic categories, such as articles, pronouns, prepositions, complementizers, negative adverbs, or auxiliaries. They do not generally belong to open classes, like verbs, nouns, or adjectives. Their prosodically defective character is most often manifested by the absence of stress, which in turn correlates with vowel reduction in those languages that have it independently; sometimes the clitic can be just a consonant or a consonant cluster, with no vowel. This same prosodically defective character forces them to attach either to the word that follows them (proclisis) or to the word that precedes them (enclisis); in some cases they even appear inside a word (mesoclisis or endoclisis). The word to which a clitic attaches is called the host. In some languages (like some dialects of Italian or Catalan) enclitics can surface as stressed, but the presence of stress can be argued to be the result of assignment of stress to the host-clitic complex, not to the clitic itself. One consequence of clitics being prosodically defective is that they cannot be the sole element of an utterance, for instance as an answer to some question; they need to always appear with a host. A useful distinction is that between simple clitics and special clitics. Simple clitics often have a nonclitic variant and appear in the expected syntactic position for nonclitics of their syntactic category. Much more attention has been paid in the literature to special clitics. Special clitics appear in a designated position within the clause or within the noun phrase (or determiner phrase). In several languages certain clitics must appear in second position, within the clause, as in most South Slavic languages, or within the noun phrase, as in Kwakw'ala. The pronominal clitics of Romance languages or Greek must have the verb as a host and appear in a position different from the full noun phrase. A much debated question is whether the position of special clitics is the result of syntactic movement, or whether other factors, morphological or phonological, intervene as well or are the sole motivation for their position. Clitics can also cluster, with some languages allowing only sequences of two clitics, and other languages allowing longer sequences. Here one relevant question is what determines the order of the clitics, with the main avenues of analysis being approaches based on syntactic movement, approaches based on the types of morphosyntactic features each clitic has, and approaches based on templates. An additional issue concerning clitic clusters is the incompatibility between specific clitics when combined and the changes that this incompatibility can provoke in the form of one or more of the clitics. Combinations of identical or nearly identical clitics are often disallowed, and the constraint known as the Person-Case Constraint (PCC) disallows combinations of clitics with a first or second person accusative clitic (a direct object, DO, clitic) and a third person (and sometimes also first or second person) dative clitic (an indirect object, IO, clitic). In all these cases either one of the clitics surfaces with the form of another clitic or one of the clitics does not surface; sometimes there is no possible output. Here again both syntactic and morphological approaches have been proposed.

dom and datives in Basque

This paper analyzes the syntax of dom and causee, experiencer, goal and possessor datives in Basque. It presents novel criteria distinguishing their categorical status: the possibility (i) to license Depictive Secondary Predication (DSP) and (ii) to appear as non-agreeing in contexts affected by the Person Case Constraint (PCC). It argues that, contrary to the rest of the datives, goals are generated as PPs, since they are unable to license DSP, but able to occur as non-agreeing in PCC-affected contexts. Besides, despite exhibiting the same categorical status as causee, experiencer and possessor datives, it claims that dom objects are syntactically identical to canonical absolutives, as they show the same configurational as well as Case licensing pattern, which is based on v-Agree.

Nominal licensing is driven by valued (phi-)features

This short paper lays out the components of a new model of nominal licensing, motivated by novel observations about parallels between the Person Case Constraint and Differential Object Marking. The model revolves around the idea that valued features on nominals---namely, phi-features and features related to definiteness and animacy---are the sorts of features that need abstract licensing, rather than an abstract Case feature. This model helps us understand where differential marking and featural restrictions occur, and in particular, why it is that subjects and indirect objects, in contrast to direct objects, tend not to be differentially marked or featurally restricted.

Person Feature Geometry and Pronominal Clitic Ordering Preferences

Summary This paper examines so-called “Person Case Constraint” (PCC) effects. These are ordering restrictions on co-occurring clitic pronouns, where only certain person combinations are felicitous but the possibilities vary cross-linguistically. Taking the South Slavic languages as a point of departure, an account is developed in terms of person feature spreading from a high Appl(icative) node to underspecified clitic pronouns. It is argued that 3rd person is the absence of person features, hence there is no PERS(on) node per se, and that person can be characterized in terms of PART(icipant) and AUTH(or). It is further argued that languages may differ in terms of how these two features are arranged, accounting for additional systems. Finally, speculations are offered on how feature spreading enlightens PCC violation repair strategies.