Complex Predicates Research Articles

Introduction

Introduction

An acoustic exploration of event construals in Bengali language

The world is full of events, and events are construed. One of the major research questions, therefore, seeks to understand the way events are construed through language. Event construals involve the syntacto-semantic properties of certain linguistic categories such as Verb, Tense, Aspect, Modality, etc. Serial Verb Constructions (SVC) and Complex Predicate Constructions (CPC) are no exception to this very fact. In this work, we look forward to compare both linear and non linear acoustical features generated from SVC and CPC events found in sentences of Bengali language. For this, we recorded 60 common utterances of Bengali language containing SVC and CPC events individually (around 36 of them belong to SV category while the rest belong to CP category), from 1 male and 1 female native Bengali speakers. The serial verb construction in these utterances may contain two (2) serial verbs in either simultaneous or sequential order, while few sentences are such that they can be interpreted both as SV or CP events. The main objective is to look for robust acoustic features which lead to perceptual categorization of events as SV or CP with a particular linguistic background. Various linear features like MFCC (Mel frequency Cepstral Coefficients), spectral skewness/energy, pause duration, pitch profile and nonlinear features like Fractal Dimension (FD) have been employed for the classification purpose. This work is a pilot study of an ongoing project which looks to explore the concept integrating capacity of human brain in terms of Syntactic Compositionality or Semantic Combinatorics in a complex sentence. This preliminary acoustic study reveals interesting new results in terms of perceptual linguistic representation of the event construals.

On improving knowledge graph facilitated simple question answering system

Leveraging knowledge graph will benefit question answering tasks, as KG contains well-structured informative data. However, training knowledge graph-based simple question answering systems is known computationally expensive due to the complex predicate extraction and candidate pool generation. Moreover, the existing methods based on convolutional neural network (CNN) or recurrent neural network (RNN) overestimate the importance of predicate features thus reduce performance. To address these challenges, we propose a time-efficient and resource-effective framework. We use leaky n-gram to balance recall and candidate pool size in candidate pool generation. For predicate extraction, we propose a soft-histogram and self-attention (SHSA) module which serves the role of preserving the global information of questions via feature matrices. And this leads to reduce the RNN module as the simple feedforward network in predicate representation. We also designed a Hamming lower-bound label encoding algorithm to encode the label representations in lower dimensions. Experiments on benchmark datasets show that our method outperforms the competitive work for end-tasks and achieves better recall with a significantly pruned candidate space.

Heads and layers in agglutination: A case in deadjectival psych verbs with-garuin Japanese

AbstractUsing deadjectival psych verbs with-garuin Japanese, this study shows that agglutinative complex predicate formation is done by recursive application of Merge to roots and functional heads. This process creates a layered syntactic structure, with each layer providing the computational system with (i) specific semantic features, (ii) arguments, and (iii) phonetic form (PF) exponents at conceptual–intentional (CI)/sensory motor (SM) Interfaces. The whole amalgam of the root and the functional heads is interpreted as a “word” at PF. Following the general architecture of Distributed Morphology, I will show that the morpheme that derives deadjectival verbs-garuis underlyingly-k-ar-u(k-Copula-T), wherekis “little” v that originates in the verbal rootk-o“come” andar-is a copula. They are now grammaticalized functional heads that extend adjectival roots. Crucially, thiskis homophonous with “little” a, which makes-garuand the adjective-deriving morpheme-karu(k-Copula-T) parallel.kis voiced in-garudue to a structurally conditioned assimilation rule (Embick 2013). This analysis reveals the mechanisms of agglutinative predicate formation in a precise and detailed manner. Similarly, it gives natural solutions to some of the long-standing problems including how adjectives modify N such asutukusii dansaa“beautiful dancer,” which is ambiguous between attributive modification and a relative clause.

Complex predication and adverbial modification in Wagiman

ABSTRACT In many languages of northern Australia with coverb constructions, it is difficult to draw a distinction between predication and adverbial modification because coverbs appear to be both predicates and modifiers. We present evidence from Wagiman that a distinction between predication and modification can be drawn syntactically. We argue that Wagiman has two necessarily predicational positions and at least one adverbial position. One predicational position is obligatorily filled by the inflecting verb, the second can be filled by a coverb when it has a predicative function. The merger of this predicational coverb and the verb results in a complex predicate. A coverb can also be located in the adverbial position when it modifies the predicate. Using the minimalist framework, we test our proposal when two coverbs appear in the same clause, one adverbial and one predicational. This analysis has wider typological implications. We demonstrate that the novel adverbial analysis for clauses with two coverbs is also supported for other languages in the region.

Cross-linguistic influence in L1 processing of morphosyntactic variation: Evidence from L2 learners

AbstractThe current study investigates cross-linguistic influence of second language (L2) learning on native language (L1) processing of morphosyntactic variation in proficient L2 learners immersed in their L1. Despite Spanish pre- and postverbal clitic pronoun positions being grammatical in complex verb phrases, preferences of use have been well attested in naturalistic language production. To examine whether those preferences obtain for comprehension in monolinguals, as well as how those preferences might be modulated by learning an L2 with fixed pronoun positions, we administered a self-paced reading experiment to 20 Spanish monolinguals as well as 22 proficient learners English (L1 Spanish). The results of a Bayesian mixed effects regression analysis suggest that preferences in production are echoed in comprehension—but only for the monolingual group. We find support for facilitation in the bilingual group precisely where both languages overlap, as well as evidence that bilinguals may not use clitic position as a reliable cue at all. We interpret the results as evidence that learning an L2 that lacks variation for a particular feature may lead to reduced sensitivity to that feature as a cue in an analogous L1 structure. We situate these results in an experience-based, shared-syntax account of language processing.

Composite predicates in the layers of Latin

AbstractThe present study gives a comprehensive portrait of the accusatival-passival type of the composite-predicate (or support-verb) construction in the history of Latin. Taking stock of properties shows them to only minimally serve as a touchstone distinguishing between Latin composite predicates and syntagmatically identical verb + direct object constructs. The intricate issue of idiomaticity and the prospects of establishing its parameters are discussed, concluding that although characteristics such as figurativity of the verb or direct-object incorporation indicate this status, the closed corpus of Latin securely allows only for a quantitative parameter, or familiarity, for this kind of a complex predicate. We are looking from the diachronic perspective at individual support verbs, primarilycapereandhabere, in the evolvement of their function and status—grammatical or lexical-phraseological—as well as at other sets of verbs, figurative ones and verbs otherwise departing from their basic meaning. The support verb emerges as the cardinal constituent of the construct, determining the function of the complex, and as considerably autonomous, being capable of representing, in anaphoric contexts, the entire construction. A reconsideration of the opposition between composite predicates and their monolexematic counterparts brings up again their added value as to exhaustiveness, clarity, perspective, and focus of the message conveyed through a composite predicate.

Complex predicates, simple inflecting verbs, and “uninflecting verbs” in Pre-Basque

AbstractHow might Basque have looked before it came in contact with Latin? This interesting line of research may give us an idea of what the pre-Indo-European languages of Europe might have looked like, and it may help clarify how much contact-induced change Basque might have undergone during the last two millennia or so. The present paper puts forward the hypothesis that, towards the end of the Era (BC), Pre-Basque used to have a small class of verbs. These verbs were inflected for person and tense-aspect (although we know little about the specific characteristics of this inflectional system). Together with this small class of verbs, Pre-Basque had a larger group of uninflecting elements that combined with the inflecting verbs to form complex predicates. The group of uninflecting elements included bare nouns, adjectives, possibly adverbs, ideophones, and what I will call “uninflecting verbs”. The exact nature of these “uninflecting verbs” is hard to determine at this point, but they may have constituted a distinct part of speech. Certainly, this type of verbal organization is reminiscent of one common in Northern Australia. Thus, this paper also compares the reconstruction proposed for Pre-Basque with the verbal system typical in Northern Australian languages, to conclude that the similarities are remarkable and, therefore, that the verbal organization of Pre-Basque was quite different from that of the modern Western European languages, including Modern Basque.

I verbi di servizio (modali, aspettuali e causativi) e i verbi supporto nelle grammatiche descrittive d’italiano dal 1953 al 2005

The paper deals with Italian helping verbs (modal, aspectual, causative verbs) and light verbs which form complex predicates and with the rules that guide its use in Italian descriptive grammars published in Italy between 1953 and 2005. The author shows that those forms in both – traditional and contemporary Italian descriptive grammars are treated at the level of word classes whereas it could be more appropriate to discuss them as clause elements as they form complex predicates. In our opinion this way of describing such verbs is due to the tradition in Italy to focus on a form and not on function of an element. What is more, Italian grammarians tend to omit in Italian descriptive grammars noun predicationand the role of semantic predicate (n pred.). The goal of the paper is to present the advantages of an integrated approach to helping verbs (modal, aspectual, causative verbs) and light verbs in modern Italian. The author emphasizes its importance for contemporary Italian FL syntax teaching.

The syntax and semantics of Modern Standard Arabic resumptivetough-constructions

AbstractThis paper is concerned with a hitherto undiscussed type oftough-construction in Modern Standard Arabic (MSA). Our starting point is the observation that thetough-adjective in this construction invariably displays nominative masculine singular morphology, a pattern of ‘default’ agreement which does not seem to occur elsewhere in the grammar of MSA. At a semantic level, the relevant adjective is argued to form a complex predicate with a deverbal nominalization that acts as its complement: together, these two elements indirectly modify the subject noun phrase. To explain the default agreement pattern, we propose that MSAtough-constructions involve two distinct subjects, viz. a phonologically null expletive subject which controls agreement on thetough-adjective, and a Broad Subject which acts as the semantic subject of the whole construction. We show that there is independent evidence for the existence of both null expletives and Broad Subjects in MSA.

Predicate derivations in Thulung

Abstract In this article, I describe a subset of complex predicates in Thulung (Kiranti, Nepal) which I call “predicate derivations” (after Post, Mark. 2010. Predicate derivations in the Tani languages: root, suffix, both or neither? In Mark Post & Stephen Morey (eds.), North East Indian linguistics, 175–197. New Delhi: Cambridge University Press India). I argue that these constructions in Thulung are built on grammaticalized elements that can be synchronically considered to be derivational. I focus on their morphology, placing it within the context of simplex verbs, and highlighting certain features – multiple exponence, allomorphy of the derivational suffixes similar to that of simplex verbs – which suggest strongly a development from serial verb constructions. I discuss their functions, which cover valence changes, as well as the marking of associated motion and aspect/Aktionsart. Complex predicates are seen as a feature of South Asian languages, but the types found in Tibeto-Burman languages appear to be quite different, morphologically, from what is found in Indo-Aryan or Dravidian languages. This article therefore represents a language-specific description as a contribution to cross-linguistic research on the topic.

The emergence and development of complex predicates in early child Hebrew

The present study examines the development of the earliest type of complex predicates to emerge in child Hebrew – extended predicate constructions. These constructions take the form of a modal/aspectual operator followed by an infinitival verb form (e.g., roce lesaxek ‘want to.play’), and since they serve various discursive functions (e.g., intent, desire, request), their use marks a significant development in toddlers’ cognitive, linguistic, and conversational abilities. Focus here is on the three most common modal operators used in extended predicates in early child Hebrew: roce ‘want’, yaxol ‘can’, and carix ‘need/have.to’. Analyses cover all uses of these three predicates by both children and adults, in longitudinal weekly-recorded speech samples from three Hebrew-acquiring children, starting with the earliest verb forms produced by each child and lasting for a period of over a year (age range: 1;3–2;5). Findings show the following developmental trends: (a) roce ‘want’ emerges the earliest and continues to be the most frequent of the three predicates throughout the year; (b) children favor using roce, yaxol, and carix as main predicates before they start integrating them in extended-predicate contexts; (c) early adult uses of extended predicates in child-directed speech are almost exclusively in the form of Yes/No verification questions (e.g., ata roce laševet? ‘you want to.sit?’); and (d) as children gradually use more extended predicates, adults gradually use fewer Yes/No verification questions, and more prompts that elicit information from the children (e.g., ma ata roce la’asot? ‘what (do) you want to.do?’) while also expanding more on the children’s productions. These concurrent trends show that such complex predicates develop in a piecemeal fashion, promoted by both the gradually evolving relations between children’s linguistic productions and their discursive functions, and the supportive contexts provided by their adult interlocutors.

История (не)существующего наклонения: да-конструкция в македонском и сербском языках (независимый тип)

The da-construction in Macedonian and Serbian languages combines the modal conjunctive particle da with the finite form of the verb to express a wide range of modal meanings. Being a Balkan Sprachbund novelty, the da-constructions in dependent clauses take on the functions of the infinitive in a complex predicate (in all cases in Macedonian, partly in Serbian). The da-construction in independent simple sentences has a complete conjugation paradigm with the imperative meaning and competes with the synthetic imperative forms in both languages, expressing also optative meanings. The article offers an analytical review on the status of these forms based on data gathered from Macedonian, Bulgarian (being closely related to Macedonian) and Serbian languages. There are two salient extremes in it: 1) the da-construction is not an analytical form of the verb and does not have an independent status in the verb system, instead it functions as a syntactic construction; 2) it is indeed an analytical form of the subjunctive mood, with da being a grammatical indicator of modality with the general non-factual meaning. The article concludes that the first view should be regarded as definitive in determining the status of the da- constructions in the Serbian language. To support this conclusion, we are putting forward the following arguments: a) the verb form is non-adjacent to the da particle in the form; b) there is a competition between the infinitive and the da- constructions as complex predicates as well as in some tenses (future tense: ja ću ići / ja ću da idem) and negative structures (nemoj ići / nemoj da ideš); c) The conjunctive particle da does not always have the non-factual meaning, and in most cases the following verb form can take on any tense or mood. In Macedonian the da-construction should be regarded as an analytical form of the subjunctive mood on account of the following characteristics: а) the da-construction has a syntactically fixed component order, which means that it is a semantically bound or fused chain of components which lost their individual meanings; b) the da-construction has a complete conjugation paradigm, one for da+praes and another for da+imperf, on which stems a wide range of modal meanings: imperative, optative and some others; c) the particle da has an invariant meaning of non-factuality, and it marks the verb form in this construction as non-indicative.

Экспериментальное исследование приемлемости придаточных с союзом что в функции сентенциального актанта существительного в конструкциях с сочинением

Previously, čto-clause complements of nouns were found to be more acceptable in constructions with collocations as opposed to non-collocations, which was taken to support the hypothesis whereby čto-clauses are introduced by a silent preposition licensed by incorporation into a complex predicate (created in collocational constructions). The paper presents the results of an acceptability rating study testing the prediction of this account, namely, that embedding the noun in a coordinate structure should block the incorporation and hence licensing of čto-clauses. In addition, an alternative hypothesis is examined according to which čto-clauses are more acceptable in collocations due to their higher frequency.

A Complex Predicate Approach to the Caused-Motion Construction in English

It is generally assumed within the tradition of Construction Grammar that the oblique grammatical function for the English caused-motion construction directly corresponds to a prepositional phrase (PP), such as in the jar in put the spoons in the jar. Opposed to the assumption, this paper argues that the oblique function actually narrowly corresponds to the noun phrase (NP) complement of the head preposition. On the other hand, the transitive preposition is an integral part of the predicate that it forms with a lexical verb, what we shall term a complex predicate. It is thus in phrasal syntax that the preposition, together with the oblique NP, forms or corresponds to the PP. This argument is supported by the class of phrasal verb constructions with so-called unpredicated particles, such as off in wipe off the table and wipe the table off (≠*The table is off). The paper also advances an alternative analysis of the construction with theoretical advantages.

Transferência reversa e colocação clítica um estudo com bilíngues Português Brasileiro e Espanhol

Este artigo trata de transferência reversa, isto é, a influência da L2 na L1, em situação de dominância da L1. Investiga-se a colocação pronominal clítica em complexos verbais no português brasileiro (PB) e no espanhol. O espanhol privilegia a próclise ao verbo auxiliar (posição pré-verbal/subida do clítico) ou a ênclise ao verbo principal (posição pós-verbal). O PB, por sua vez, perdeu a subida do clítico no século XIX. Embora a escola tente recuperar seu uso, a fala natural privilegia a próclise ao verbo principal (posição medial). Assim, em princípio, pode-se admitir que falantes escolarizados do PB aceitem a ordem com a subida do clítico. Por outro lado, uma aceitação plena da subida do clítico pode constituir um caso de transferência reversa, no contexto de falantes bilíngues de PB/espanhol. A fim de observar esse quadro, aplicou-se uma tarefa de leitura automonitorada com escala Likert para julgamento de gramaticalidade, manipulando a posição do clítico, em sentenças em português, a monolíngues de PB com alta escolaridade e bilíngues PB/espanhol. Os resultados demonstram que ambos grupos aceitam a subida do clítico, mas os bilíngues apresentam maior aceitação e são mais rápidos na leitura das sentenças em português com subida do clítico, sugerindo uma possível facilitação do espanhol no processamento de sentenças em português.

Basque complex predicates and grammar change

Grammar change gives rise to complex predicates, which are the product of a syntax that was free in a previous period and that doesn’t function freely any longer. This paper explores the evolution of two types of Basque complex predicates that are respectively perceived as verbal and nominal in today’s syntax, although they do not clearly exhibit overt verbal or nominal syntax. We will see that what is relevant in the description of the change is sometimes the basis of the difference between nominal and verbal predication. Both type of predicates will be shown to belong to a verbal or a nominal scheme, by means of word order and determiner distribution in their arguments.

Serial Verb Construction in Etulo

In languages, different criteria are often adopted in the classification of the verb category many of which are language specific or universal. Etulo makes a distinction between simple and complex predicates. This work discusses serial verb construction (SVC) as a type of complex predicate using the typological criteria proposed in Aikhenvald (2006). Etulo SVCs have diverse semantic and grammatical functions. They express benefactive, instrumental, comparative meaning, as well as prepositional and adverbial notions indicating direction using motion verbs. Different types of serial verb constructions (SVCs) are established: the symmetric vs asymmetric type, contiguous vs non-contiguous, and the optional vs obligatory type. The SVC is further distinguished from a similar multi-verb construction known as the consecutive construction.

ESTAR + FC en el castellano peruano

En el presente artículo, describiré la construcción de tipo Ella está que canta del castellano de la variedad peruana, en donde el verbo ‘estar’ selecciona una Frase Complementante (en adelante FC) para conformar una predicación compleja. Seguiré la propuesta de Cuervo (2003) sobre la conformación de estructura eventiva para analizar los componentes de dicha construcción; y la ausencia de rasgos de chequeo de Caso en FT para licenciar el hiperascenso de Frase Determinante (FD) (Martins y Nunes 2010; Fernandez-Salgueiro 2005; Fong 2019, Pesetsky y Torrego 2001, 2004, 2007). Probaré, en primer lugar, la monoclausalidad de la estructura. Luego, mostraré que esta estructura se conforma a través de una organización eventiva por complementación: un evento estativo selecciona a otro dinámico, y ambos son interpretados desde una perspectiva aspectual de iteración; y, finalmente, explicaré el ascenso del sujeto desde una cláusula inferior a través de la inespecificación de rasgo T del verbo inferior, con el consecuente “escape” de la FD a un nivel superior, y, para ello, asumiré la posibilidad de considerar Esp,C como posición argumental o mixta A/A’, dadas las propiedades de selección de ESTAR.

Frozen structure and concessive relations: a complex predication

This paper aims at analyzing the concessive relation, as a complex predication, under the scope of the freezing concept. After having defined the three terms of the analysis, we proceed to the description of the concessive relation, which includes several predicates at a time. This complexity is fixed in the language through grammaticalization, as a form of freezing, by means of grammatical expressions used to structure the organization of the predicative relations in the utterance, whether sentencial or multi-sentencial. This fixation in the language declines the same concessive relation with semantic nuances that testify to the richness of this complex predication.