Search

This work investigates the processing behavior of large language models (LLMs) in sentences involving ambiguous relative clauses (RCs). We are particularly interested in unravelling attachment preferences of LLMs in disambiguating RCs (complementizer phrases CPs modifying a genitival phrase), which are either semantically (context-biased) or syntactically (structure-biased) associated to one of the preceding NP referents. A low interpretation of the RC occurs when it is joined to the local NP (low attachment). A high interpretation is provided when the RC modifies the distant NP (high attachment). We create a small dataset of parallel low- and high-attachment sentences. We use zero-shot prompting to evaluate a set of LLMs based on insights from psycholinguistic experiments. Our results show variability in the performance of some models that favor low attachment (semantically-related meanings in the CP) while other models can resolve ambiguity by choosing high-attachment (structure-biased CPs). The findings are discussed in light of directing future experimental studies to consider a comparative paradigm encompassing both multi-modal LLMs and human subjects. [ABSTRACT FROM AUTHOR]

Linguists today presuppose that for both description/analysis and comparison of languages, there is a substantial set of universal crosslinguistic categories. In the same way, in the grammaticisation of non-European languages, missionaries took the grammatical categories elaborated for Latin as a set of universal crosslinguistic categories. One could assume that they fell into the trap of transferring time-tested terminology from traditional Latin grammar adapting it to the description of a language with rather different structural properties without really capturing its 'genius'. Even if this assumption may certainly be true to some extent, it does not present the whole picture and it does not do justice to the efforts missionaries made in describing new languages. Therefore, focusing on how missionaries understood and described the Tamil relative clause and relativiser marker, this paper aims to discuss the tension arising in grammatical descriptions between language-particular categories and crosslinguistic conceptual transfer. The paper thus demonstrates how missionaries did not limit their accounts to the transfer of a priori grammatical categories, tailored for Latin and applied to Tamil, but rather how they refined their descriptions through a crosslinguistic conceptual transfer that may be considered the first step towards comparative concepts. [ABSTRACT FROM AUTHOR]

Copyright of Jezikoslovlje is the property of University of Osijek, Faculty of Philosophy and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The relative constructions in several Palaungic languages (Htanaw, Wa, Lawa, Rumai Palaung, Samlong Palaung, and Rucing Palaung), here shown to be participant nominalizations, display striking mutual similarities, while conspicuously diverging from the dominant relativization strategy within the Austroasiatic family. Instead of the common n [rel (s) v (o)] pattern, the Palaungic constructions examined exhibit the following structural features: (a) rel invariably precedes the verb complex directly; (b) internal constituent order is vs(o), with the exception of Htanaw. An unusual functional trait is additionally found in the three Palaung varieties: the construction only performs object relativization. By placing the findings in a diachronic perspective, we propose two new pathways of branch-internal syntactic change that may explain this unusual synchronic status. Among these, the lexical-to-clausal-nominalization pathway in particular offers a plausible alternative scenario to the earlier hypothesis that such verb-initial structures are inherited from Proto-Austroasiatic (Jenny 2020). Furthermore, a tentatively suggested etymological origin of the relativizer, pAA *pa 'place', provides an account of the narrow semantics in Palaung. [ABSTRACT FROM AUTHOR]

Infinitival relative clauses are untensed relative clauses which generally have a root modal value. The construction has a distinctive characteristic that makes it particularly interesting from a syntactic point of view: the impossibility to have a lexical wh-DP operator in initial position unless it is integrated in a prepositional phrase. In this paper I describe the structure and interpretation of infinitival relative clauses and discuss why such restriction, which is not present in tensed relative clauses, exists. The proposal I make is that the modal reading of these sentences and the DP/PP asymmetry they exhibit in CP are connected to the prepositional origin of particles to and for, and to the categorial features that they have retained in Present-Day English after a process of grammaticalization; the DP/PP asymmetry, in particular, will follow from the need to value the categorial features with which C is endowed here with some prepositional category under strict conditions of locality. The analysis provided thus explains the peculiarities of infinitival relative clauses with no need to resort to any construction specific mechanism, in line with the restrictive approach to linguistic facts of the Minimalist programme. [ABSTRACT FROM AUTHOR]

The well-known typological feature of Slavic possessive adjectives is their ability to control personal pronouns, relative pronouns, even modifiers, i.e. their noun-like behaviour. In Slavic languages, this feature is developed to a different extent, e.g. in Sorbian to a greater, in Croatian to a lesser extent. The paper is a follow-up of previous research and puts forth new data on the possibility of Croatian possessive adjectives to control pronouns. The data is based on a corpus of recently gathered examples from contemporary texts and on a more detailed analysis of tokens gathered from online corpora, dating from the 1850s to the 2010s. It is shown to what extent the noun-like character of Croatian adjectives can be established, in which type of texts it occurs, as well as which are its consequences on the understanding of adjectives as a word class. [ABSTRACT FROM AUTHOR]

The structures that have been referred to as relative clauses in Akan (Niger Congo, Kwa) are re-examined in light of Shibatani (2009, 2019), which shows these structures formally to be nominalizations. It is shown that Akan "relative clauses" are nominalizations in modification use. "Headless relative clauses" are shown to be argument uses of nominalizations. Evidence adduced includes the occurrence of relative structures with determiners, internal syntactic changes such as the inability to take full NPs in relevant argument positions, and external syntactic properties of occurrence as subjects and objects in clauses. Forms traditionally classed as relativizers are shown to be nominalizers or nominalization markers. The polysemous nature of these markers is seen in their occurrence with "headless genitives", which are also shown to be nominalizations. The paper highlights the value of separating form from function in syntactic analyses. [ABSTRACT FROM AUTHOR]

This study presents a cross-linguistic investigation of attributive temporal clauses encoded by a Generic Head Noun meaning ‘time’. While most studies have concentrated on the diachronic origin of this construction, no typological study has explored the synchronic properties of constructions encoded by a GHN of time (e.g. At the time I went there, I felt sick) that are used to express temporal adverbial semantic relations. The research reports on 45 languages in which attributive temporal clauses are the primary conventionalized way of expressing temporal adverbial relations. This construction is found in almost every macro-area, but especially in Africa, Papunesia, and Eurasia (particularly in Sino-Tibetan languages and Caucasian languages). In exploring the construction, the study takes into account three parameters, viz. the linear position of the Generic Head Noun of time, the encoding of Generic Head Nouns of time in comparison to other relativized temporal nouns (e.g. ʽdayʼ, ʽyearʼ), and whether languages tend to have specialized or unspecialized Generic Head Nouns of time to encode this construction. [ABSTRACT FROM AUTHOR]

The article presents the discussion on esteemed e-journal Language in India. Topics include word order encompassing the constituent order of a clause namely the relative order of subject, object, and verb, the order of modifiers; and SVO language with rigid word order likely to contribute to the study of linguistic universals as well as evaluation of existing linguistic theories.

The paper attempts to explore how a nominalized clause (relative clause) is formed and what it does in a noun phrase (NP). The relative clause can be made by nominalizing the verb of the clause or by using the quotative -haibə-. The Manipuri relative clause, like an adjective, modifies a head noun and can either precede or follow the head noun of an NP. Our findings reveal that the relative clause as a modifier can occur in an NP that functions as a subject, object or complement of a sentence. In a certain context understood by both the speaker or writer and hearer or reader, the head noun that is modified by the relative clause can be deleted and in that NP the only remaining constituent has to be the relative clause. An interesting new observation made in our analysis is that the quotative -haibə- is mainly used in relative clause formation to retain the quoted statement as it is. [ABSTRACT FROM AUTHOR]

In this article, the similarities and differences in the usage of relative clauses in Estonian, Finnish, and Russian in the overall framework of referential devices are studied. The data come from two experimental settings: "Houses" and "Narratives". In "Houses", the participants' task was to describe and compare previously defined houses to the experimenter while looking out of a window. In "Narratives", a picture-sequencebased narrative elicitation method was used to collect short and coherent spoken narratives. Collected data were coded for referential devices, including bare NPs, demonstratives, personal pronouns, zero reference, and relative clauses. In "Houses", the Russian data had the largest amount of relative clauses and the smallest amount of demonstrative pronouns; Finnish had the smallest number of relative clauses and the largest number of demonstratives. In the Estonian data, the frequencies of demonstratives and relative clauses were between those of Russian and Finnish. In "Narratives", the frequency of relative clauses was approximately the same in Estonian, Finnish, and Russian. This suggests that these three languages differ in the usage of relative clauses only in a spatial context. [ABSTRACT FROM AUTHOR]

Copyright of Contributions to Oriental Philology / Prilozi za Orijentalnu Filologiju is the property of Orijentalni institut and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

This initial study examines the development of the acquisition of relative clauses among Malay children in the Klang Valley. The objectives of the study were to determine the abilities of Malay children in using relative clause structures in Malay across two different age groups and to determine whether the acquisition of relative clauses is developmental in nature i.e. developing across the age range. In addition, this study will also analyze the error patterns produced by the subjects based on specific error categories in order to show the developmental nature of the acquisition of Malay relative clauses. Six subjects were recruited and divided into two groups: the preschool children and the primary school children aged seven. Data was collected using two main methods: (i) sentence-picture matching - to test the comprehension of relative clause structures and (ii) picture selection and picture descriptionto test the production abilities of relative clauses. Both activities test the abilities of subjects to comprehend and produce subject relative clauses (SRC) and object relative clauses (ORC). The findings showed that there was no significant difference between the preschool children and the primary school children in terms of the comprehension abilities of relative clauses,but there was a significant difference in terms of the production of relative clauses. The preschool children have limited abilities in the ORC when compared to the primary school subjects. Incidentally, the ORC structures have yet to be acquired by the primary school children. The ORC structures are acquired quite late and the development of relative clauses is still continuing among the primary school children. However, it was found that the mean score obtained increased with the increase in age. The qualitative analysis showed that subjects used strategies such as the simplification and substitution strategy. The subjects also use different language varieties: the preschool subjects tended to use colloquial Malay variety, while the primary school children used the standard Malay, and this dichotomy greatly affected their performance in the activities. [ABSTRACT FROM AUTHOR]

This paper provides an alternative viewpoint on the length and tone of the relativiser in Akan. The relativiser has been analysed as a long vowel with either a high-low tone (HL) (Saah, 2010) or a mid-low tone (McCracken, 2013). However, in this paper, we posit an underlying short vowel with a low tone (L) for the relativiser. We further show that it receives a high tone (H) from an adjacent H; making it a syllable with a contour tone at the phonetic level (i.e. â ). The contour tone then affects the length of the relativiser at the phonetic level, making it slightly longer than its original length. [ABSTRACT FROM AUTHOR]

Copyright of Concentric: Studies in Linguistics is the property of National Taiwan Normal University, Department of English and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

English sentences with relative clauses and the passive voice have long confused students when they translate from English to Chinese. In this study, a learner corpus was developed and incorporated into teaching by examining the linguistic features of relative clauses and passive voice in student translations. Students' translated sentences were analyzed using WordSmith to calculate, categorize, and annotate the common linguistic features in students' translations of English sentences that contained relative clauses and passive voice. The research findings can be used to guide students to practice translation strategies that are unfamiliar to them and enhance their ability to apply various strategies. The results can thus be used to increase students' linguistic sensitivity and reduce translation errors, thereby improving learners' ability to do translation with greater accuracy. [ABSTRACT FROM AUTHOR]

Copyright of Türk Dünyası Araştırmaları is the property of Turk Dunyasi Arastirmalari Vakfi and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

This interlanguage study, with an emphasis on two different proficiency levels of L2 learners, aims at examining Thai EFL learners' acquisition of locative adverbial relative clauses and related prepositional relatives. It was revealed that the most frequent problem with which both proficiency groups were confronted pertained to the relative adverb where functioning as a relative pronoun, possibly attributed to overgeneralisation. As regards prepositional relatives, Thai learners in the lower proficiency group had apparently acquired preposition stranding prior to pied-piping, without any use of the latter structure in their interlanguage. Additionally, preposition omission and pronoun retention, which are ungrammatical in English, were also prominent in the learners' interlanguage relative clauses. [ABSTRACT FROM AUTHOR]

Researches show that the processing asymmetry of object and subject relative clauses is a cross-linguistic phenomenon. In order to reveal whether the root cause of such an asymmetry lies in the unique syntactic feature of the relative clauses or in the cognitive capacity of human beings, researchers have done a series of studies from the perspectives of behavior, development, neuro-cognitive science and corpus linguistics. In so doing, they have established a set of research paradigms and proposed convincing theoretical explanations. This paper attempts to give a justified review of what we have achieved as well as what improvements we need to make, hoping to offer an outlook for the future studies. [ABSTRACT FROM AUTHOR]

This paper discusses the structure of two types of relative clauses, restrictive and non-restrictive, and provides an OT approach grounded in Weak Bidirectional Optimization to account for the syntactic construction of both types of clauses. The paper will show that while the Hakka basic NP structure favors the head to be positioned at the right edge, it is through the OT model of constraint interaction that different ordering patterns can be generated as grammatical outputs for a given input meaning. The first part of the analysis will explain the data by proposing constraints from the production-oriented perspective, as their ranking successfully yields the correct results in a unidirectional OT model. However, in the second part of the analysis, when the data grows more complex and requires syntactic and semantic distinctions between restrictive and non-restrictive relative clauses, the explanatory power of a unidirectional approach turns out to be limited. We find the necessity to adopt a bidirectional model, in which a form-meaning combination can be evaluated as a pair. Moreover, the fact that this innovative OT allows a recursive version of evaluation provides the possibility to generate optimal outputs successively for the two distinctive types of relative clauses under a single constraint ranking. [ABSTRACT FROM AUTHOR]

The goal of this paper is to describe some of the syntactic structures that are created through nominalization processes in Himalayan Tibeto-Burman languages and the relationships between those structures. These include both structures involving the nominalization of clauses (e.g. complement clauses, relative clauses) and structures involving the nominalization of verbs and predicates (e.g. the derivation of nouns and adjectives). We will argue that, synchronically, clausal nominalization, structurally represented as [clause]NP, is the basic structure underlying many of the nominalizing constructions in these languages, even though individual constructions embed and alter this structure in interesting ways. In addition to clausal nominalization, we will illustrate the presence of derivational nominalization, represented as [V-NOM]N and [V-NOM]ADJ, although some nominal derivations target the predicate, not the verb root as their domain. We will also demonstrate that derivational nominalization can be seen as having developed from clausal nominalization, at least for some forms in some languages, and that the opposite direction of development, from derivational to clausal structures, is also attested. We will conclude with some syntactic observations pertinent to recent claims made on the historical relationship between nominalization and relativization, demonstrating that there are various ways that these structures can be related. This study is based on data from five Tibeto- Burman languages of the Himalayan area: Manange, Dolakha Newar, Mongsen Ao, Dongwang Tibetan, and Zhuokeji rGyalrong. [ABSTRACT FROM AUTHOR]