Proceedings of the International Conference on Head-Driven Phrase Structure Grammar
Not a member yet
468 research outputs found
Sort by
The morphology of trapping and climbing in French causatives
Full text linkIn this paper, we shall discuss how the treatment of
morphological periphrasis (Vincent & Börjars, 1996; Popova & Spencer,
2013; Bonami, 2015) can be integrated into the framework of
Information-based Morphology (IbM; Crysmann & Bonami, 2015), an
inferential-realisational theory of inflection couched in terms of
typed feature structures.
French verb morphology provides a rich set of synthetic and analytic
forms. Among the latter we find the so-called composed tenses
(e.g. passé composé) and the near tenses, such as the passé récent
(=‘recent past’) and the futur proche (=‘near future’). Recently, we
have argued that clitic climbing can equally be understood as
morphological periphrasis (Aguila-Multner & Crysmann, 2020a,b;
Aguila-Multner, 2023). Thus, the morphosyntax of French verbs provides
an ideal testing ground to study the interaction of different
periphrastic dependencies, which we shall use to illustrate our
treatment of periphrasis in IbM
An HPSG account for German numeral classifiers
Full text linkThis paper analyzes German numeral classifier phrases within HPSG, focusing on their morphosyntax and semantics. I differentiate between sortal and measure classifiers, which, despite being structurally identical, contribute differently semantically. In a German classifier phrase ([Num CL NP]), the classifier serves as the head, with the counted NP as its complement and requires a specifier to form a complete classifier phrase. This specifier can be realized by a determiner in the traditional sense but also by a numeral. Additionally, numerals are treated as underspecified for a specifier or a modifier, allowing for a flexible combination of determiner, numeral, and other modifier elements in the structure
Hybrid Agreement in Bosnian/Croatian/Serbian
Full text linkThis paper examines the hybrid agreement patterns in Bosnian/Croatian/Serbian (BCS). Building on the previous work of the analysis demonstrates that by adopting a default unification mechanism and extending the existing type hierarchy, a feasible analysis of BCS hybrid agreement can be accomplished. The paper also explores a functor analysis, which delivers successful results with minor changes to the type hierarchy and lexical entries. The findings also refute the claim that the phenomenon of hybrid agreement in BCS gives evidence for the DP-hypothesis, thereby rendering the argument unable to favour either the NP or the DP hypothesis
Gradient HPSG
Full text linkPrevailing grammatical frameworks treat grammaticality as a binary concept, despite strong experimental evidence suggesting it is better understood as a gradient notion. This highlights a serious disconnect between linguistic theory and empirical data. While a few truly gradient frameworks have been proposed to bridge the gap, none have been developed within a constraint-based formalism – an approach particularly well-suited for modeling gradient grammaticality. This work formally introduces a gradient version of HPSG and subsequently employs it to analyze acceptability judgment data on unlike coordination phenomena in Turkish, which display distinctly gradient patterns
The morphological causative in Panãra: A Grammar Matrix implementation
Full text linkIn this paper, we describe the modeling of the causative morpheme in an implemented HPSG grammar for the Panãra language. The Panãra causative morpheme appears on the verb between the agreement markers for the ergative and absolutive arguments, increasing the verb\u27s valency to two core arguments. Due to the linear ordering of inflectional morphemes (position classes) in Grammar Matrix grammars, the position of the causative morpheme would require the verb to have information about its valence before the morpheme is attached. We model the descriptive data with a new lexical rule that introduces the causative morpheme and changes the verbal inflection path from intransitive to transitive using the existing Valence Change library. This analysis is validated in an implemented grammar fragment for Panãra
Examining Delayed Complements in Norwegian within an Incremental Left-Branching Grammar Framework
Full text linkIn this paper, I will present an analysis of complement clauses in Norwegian that are licensed by the five adverbs så/såpass ‘so’ and slik/sånn/sådan ‘such’. It will be assumed that the licensed complement clause, although it is licensed by the adverb, is not a complement within the constituent with the adverb, but rather a complement of the clause. This opens for a uniform analysis of complement clauses licensed by adverbs, irrespective of their position with regard to the licensing adverb. The analysis will be conducted within the framework of an HPSG-inspired incremental typed feature structure grammar of Norwegian
Topic drop in German: Grammar and usage
Full text linkGerman topic drop clauses are a subtype of declarative
clauses where the initial position (usually filled by an overt
constituent) is left empty. It is often noted that topic drop appears
mainly in specific registers (e.g. spoken dialogues), but this claim has
neither been previously experimentally validated, nor formally
implemented. In this paper, we report the results of a matched-guise
study which indicate that the syntactic variation between topic drop and
regular V2 declaratives in fact correlates with different social
meanings, leading to the register variation postulated in the
literature. In order to model German speakers\u27 grammatical and register
knowledge about topic drop in HPSG we propose, (i) a unified grammatical
constraint that licenses topic drop structures, (ii) a formal theory of
register that treats social meanings as a type of use-conditional
content subject to compositional rules
Horn clauses and strict NPIs under negated matrix clauses
Full text linkThe present paper makes four main contributions: First, it argues for a surface-scope oriented approach to phenomena that have been considered strong arguments against a surface scope of negation: the licensing of embedded strict Negative Polarity Items (NPIs) and embedded Negative Inversion (i.e. Horn clauses). Second, this surface-scope analysis is expressed within existing HPSG analyses of NPI licensing and Negative Inversion. Third, the strict/non-strict distinction of NPIs is reduced to universal vs. existential quantification over the licensing requirement within a semantic representation. Fourth, Negative Inversion is analyzed as a constructional NPI. The existence of constructional NPIs should not be surprising, but no such example has been previously discussed in HPSG to my knowledge
Between shared expectations and possibilites: The discourse particle sempre in European Portuguese
Full text linkThis paper examines the syntactic and semantic properties of the confirmative use of sempre in European Portuguese. Unlike its temporal counterpart which carries the meaning of `always\u27, confirmative sempre is restricted to the pre-verbal position, disqualifying it as a prototypical adverb. In terms of its semantic contribution, the confirmative discourse particle sempre marks the proposition as given by suspending the stress on phonological constituents in the clause. Consequently, the nuclear stress falls on sempre. Contrary to the analysis proposed by Amaral & DelPrete\u27s (2014), givenness is not equated with being part of the shared knowledge/common ground. Instead, there could have been disagreement about the validity of the embedded proposition between the speaker and the addressee at some earlier stage. Givenness, therefore, will be understood as being part of at least one party\u27s discourse commitments, following the framework of Farkas & Bruce (2010)
A Construction-based Approach to Cantonese Classifiers
Full text linkThis paper focuses on the structure and interpretation of Cantonese NPs. We first map different cognitive statuses (Gundel et al. 1993) to different forms of Cantonese noun phrases, following the hierarchy proposed in Borthen & Haugereid (2005). We then provide an HPSG analysis for Cantonese noun phrases. We account for the differences between classifiers appearing with and without a numeral, where classifiers with no numeral are interpreted as having a cardinality of ‘one’. We propose a Classifier Head Rule where the noun first takes a specifier containing the classifier, and the output further takes a determiner as its specifier. The analysis is implemented in an open-source Cantonese HPSG