Traitement de la coordination à l’intérieur des groupes nominaux

The article deals with some principles that regulate the order of the elements inside the noun phrase that refers to the grammatical subject. The position of the words inside the noun phrase is variable by their grammatical and semantic functions in Lithuanian and fixed in English. The English and Lithuanian languages differ as to whether their syntactic organisation in the noun phrase reflects grammatical relations of a head noun with an article, with a noun, with an adjective, with a numeral, with a clause, etc. in a head-marked structure, or the noun phrase comprises terms or phraseological units bearing figurative sense. In the English noun phrase, the position of the dependent elements is before the head noun they modify or they are expressed by means of postposition. In Lithuanian the grammatical means of indicating grammatical relations in the noun phrase are endings and inflexional suffixes. Generally the flexibility of the word order within the noun phrase in Lithuanian is presented on the basis of morphological case marking. The dependent-marked structure of the determiners inside the noun phrase in English and Lithuanian sentences is a complex of interrelations between its elements. Thus, the semantic meaning of the sentence of the same content in the both languages can be expressed by the different order of its constituents although the order of the elements in the noun phrase can be different too. Generally this study is concerned with the noun phrase, i.e., the grammatical subject that occurs only before the main verb in a clause in English, though the position of the corresponding noun phrase translated into Lithuanian can appear before or after the main verb, or the clause may correspond to the word order with the retained object. The contrastive analysis of the noun phrase structure is based on the restrictive effects of grammatical and semantic meaning in reference to the syntactic interpretation. DOI: http://dx.doi.org/10.5755/j01.sal.0.21.3047

This paper focuses on the capacity of the case markers in the Mongolian language, as a relative element, to generate any finite noun phrase or relative clause based on their syntactic function or relationship. In Mongolian, there are two different approaches to generate noun phrases: parataxis and hypotaxis. According to my early observation, if the noun phrase generated through the parataxis, is the complement of the postpositional phrase, the head word of the relevant noun phrase can be truncated. In other words, since this head noun is governed by case marker in its null form to generate the postpositional phrase, the head noun can be encoded. The second approach generates two different types of noun phrases in their structures: free structured and non-free structured noun phrases. Of them, the free structured noun phrase allows any syntactic transformations in their internal structure based on the senses of the case markers which denote a relation. That is to say, the null constituents in this type of noun phrases can be encoded to generate an extended alternative of the noun phrase and a relative clause.

Grammatically, the simplest kinds of clause in English usually consist of nouns and verbs, for example: However, both in spoken and written languages, the positions before and after the verb are rarely occupied by just a single noun. More usually, they are filled by groups of two or more words. These groups of words are called noun phrases (NP), for example: NPs consist of a head noun plus one or two of the optional elements. These optional elements fit into four predetermined slots in the NP: 1= determiner and/or enumerator (e.g., the, a, first, his) 2= pre-head modifier (e.g., red, washed, painting, steel) 3= head noun 4= post-head modifier (e.g., in Leeds, which I showed you) Because of the optional nature of the slots 1, 2 and 4, NPs have a highly variable length, ranging from just single words to passages of text (1). Specifically in biomedical research papers, NPs are commonly long containing large quantities of information, which are also the common reason for long and complex sentences, a prominent linguistic feature of research papers in English. Some examplar sentences are as the following: Example 1 NP (The age-matched LETO rats that are developed from the same colony of OLETF rats but do not show DM and pancreatic fibrosis) were used as a normal control (2). The subject of this sentence is a long NP, with the head noun of rats and slots 1, 2 and 4. The slot 4, as the post-head modifier, consists of two clauses, which is the main reason for the length and complexity of the sentence. Example 2 These rats were maintained in NP (a temperature- and humidity- controlled room with a 12:12h light-dark cycle) (2). After the preposition in, there is an NP with the head noun of room. The slots 1, 2 and 4 define the number and conditions of the room. The comparatively complex NP enlarges the quantity of information contained in just one simple sentence. Such long and complex NPs are common in biomedical research papers, whose use constitutes challenge to Chinese scholars due to the differences in habitual way of thinking and language expressing between Chinese and English. Linguistically, it is known as the negative transfer of the mother tongue (3). In this paper, I categorize the inappropriate use of NP in biomedical research papers written by Chinese scholars into four types as the following.

This article discusses two relatively novel properties of possessive noun phrases in Arabic and explores their implications for existing analyses. First, in some dialects of Moroccan Arabic the genitive preposition shows agreement in number and gender with the possessum to its left in analytic (Free State) noun phrases. Preposition-possessum agreement appears to parallel verb-subject agreement and is to all intent and purposes an instance of predicate-subject agreement. Secondly, in many Arabic varieties inalienable noun phrases are incompatible with the analytic pattern and can only have the synthetic (Construct State) pattern. These two properties are difficult to accommodate under an analysis that assigns a uniform syntactic structure to both alienable and inalienable noun phrases, where the possessum is the head noun and the possessor is its specifier. Instead, they point to an analysis where the two types of noun phrases derive from radically different syntactic structures. While inalienable noun phrases derive from a structure where the inalienable possessum is the head noun and the possessor is its argument, alienable noun phrases derive from a structure along the lines outlined in Den Dikken (2006), where the possessum is the subject and the possessor a predicate complement, and the relationship between the two is necessarily mediated by a relational category that corresponds to the genitive preposition in the analytic pattern. A version of this analysis is adopted here that accounts for additional known properties of analytic and synthetic noun phrases, including their PF interface properties. The latter part builds on the analysis outlined in Ouhalla (2009b) and addresses questions left open there.

Open Knowledge Graphs (OpenKG) refer to a set of (head noun phrase, relation phrase, tail noun phrase) triples such as (tesla, return to, new york) extracted from a corpus using OpenIE tools. While OpenKGs are easy to bootstrap for a domain, they are very sparse and far from being directly usable in an end task. Therefore, the task of predicting new facts, i.e., link prediction, becomes an important step while using these graphs in downstream tasks such as text comprehension, question answering, and web search query recommendation. Learning embeddings for OpenKGs is one approach for link prediction that has received some attention lately. However, on careful examination, we found that current OpenKG link prediction algorithms often predict noun phrases (NPs) with incompatible types for given noun and relation phrases. We address this problem in this work and propose OKGIT that improves OpenKG link prediction using novel type compatibility score and type regularization. With extensive experiments on multiple datasets, we show that the proposed method achieves state-of-the-art performance while producing type compatible NPs in the link prediction task.

Open Knowledge Graphs (OpenKG) refer to a set of (head noun phrase, relation phrase, tail noun phrase) triples such as (tesla, return to, new york) extracted from a corpus using OpenIE tools. While OpenKGs are easy to bootstrap for a domain, they are very sparse and far from being directly usable in an end task. Therefore, the task of predicting new facts, i.e., link prediction, becomes an important step while using these graphs in downstream tasks such as text comprehension, question answering, and web search query recommendation. Learning embeddings for OpenKGs is one approach for link prediction that has received some attention lately. However, on careful examination, we found that current OpenKG link prediction algorithms often predict noun phrases (NPs) with incompatible types for given noun and relation phrases. We address this problem in this work and propose OKGIT that improves OpenKG link prediction using novel type compatibility score and type regularization. With extensive experiments on multiple datasets, we show that the proposed method achieves state-of-the-art performance while producing type compatible NPs in the link prediction task.

The fact that concepts are grammaticalized differently in different languages is a major problem for translation, especially for machine translation. Two major examples of this are syntactic number, and the use of (in)definite articles (a, some, the). In languages such as English, nouns are marked for number and the choice of article (or of no article) must be made for every noun phrase. In contrast, for languages such as Japanese, number distinctions are not normally made, and there are no articles. This means that whenever a noun phrase is translated from Japanese to English, even if the denotation is perfectly understood and a good translation equivalent found, generating the noun phrase still requires two difficult choices: should the head noun be singular or plural, and which article, if any, should be generated. This thesis proposes a semantic representation and a series of three heuristic algorithms that make possible the appropriate generation of articles and number when translating from Japanese to English. The semantic representation provides a tractable set of features to represent (1) the referential use of a noun phrase, as either referential, generic, ascriptive or idiomatic; (2) the interpretation of the noun phrase's referent as either a countable individual or a mass, with seven detailed subtypes; (3) the definiteness of the noun phrase, as either definite, indefinite, definite and extensive, or possessed. The three algorithms automatically acquire values for these features from the analysis of the Japanese text and the lexical properties of the English translation equivalents, and then use them to generate English. The first algorithm determines the referential use of Japanese noun phrases, based on a defeasible hierarchy of pragmatic rules that are applied top-down, from the clause to the noun phrase. The second algorithm determines the appropriate interpretation for English noun phrases, while the third determines which determiner, if any, should be generated. These algorithms use rules based on the different referential uses of the noun phrase. The proposed algorithms are implemented in a Japanese-to-English machine translation system, and the detailed lexical information is entered into its lexicon. The use of the algorithms improves the percentage of noun phrases generated with correct use of articles and number from 65% to 85%.

This paper discusses on how a head modifier of Noun Phrases (NPs) in Malay sentence can be detected from the four combination of phrases, such as a noun phrase (frasa nama) and noun phrase (frasa nama), a noun phrase and verb phrase (frasa kerja), a noun phrase and adjective phrase (frasa adjektif) and, a noun phrase and prepositional phrase (frasa sendi). Most of the sentences in Malay have compound nouns. The position of the head within a compound often depends on the order of the word, i.e. the most common order of constituents in Malay phrases, where nouns are modified by adjectives, verbs, other nouns and prepositional. We also investigated the relative contribution of the modifier and the head noun in noun compounds of different other related examples in Malay sentence.

This study examines English syntax, which is concerned with the study of the ordering of components or constituents in phrases and sentences. The spotlight of investigation was directed toward the ordering of the constituents of noun phrases found in the tale Cinderella. The noun phrases found in the tale were broken down into their constituent parts for two reasons. The first is to reveal the sequential ordering of constituents in noun phrases. Do all noun phrases in the tale follow the following sequential ordering: pre-modifiers + head noun + post=modifiers? Do pre-modifiers consist of determiners and adjectives? What constituents can these determinants be broken into? Do the adjectives consist of one single adjective or a series of adjectives? Do post-modifiers consist of prepositional phrases and relative clauses? How many prepositional phrases did the writer use to post-modify a head noun? How many relative clauses did the writer use to post modify the head noun? Did the writer place prepositional phrases before or after the relative clauses? Are relative clauses reduced or non-reduced in nature? Second, to reveal the types of noun phrases used by the writer of the tale—basic or complex noun phrases. Are most noun phrases used by the writer basic or complex in nature? To attain the research objectives mentioned above, a descriptive method was used. This method attempts to describe and interpret the types and sequential ordering of the constituent parts of noun phrases found in the tale Cinderella.

Aspects of Plurality in ≠Hoan Chris Collins In ≠Hoan, nouns and verbs can be pluralized. This article investigates various syntactic and semantic aspects of plurality in ≠Hoan: first, the formation of plurals of inalienable nouns, then, how the plurals of verbs (termed pluractional verbs) are formed in much the same way as the plurals of inalienable nouns. The phenomenon of pluractional verbs strongly supports the event argument analysis of verbal semantics.* Introduction The parallel between noun phrases and verb phrases has been a longstanding issue in generative grammar (Chomsky 1970). For example, it is generally held that both noun phrases and verb phrases fall under X′-Theory, and that similar types of movement can occur in noun phrases and verb phrases. But in spite of the many similarities between noun phrases and verb phrases, there are a number of recalcitrant differences. Introductory linguistics textbooks often note that verbs cannot be pluralized, and that this distinguishes nouns and verbs. I will argue here that in ≠Hoan,1 nouns and verbs can be pluralized. This will add a further dimension to the parallel that is often postulated between noun phrases and clauses (Abney 1987, Bernstein 2000, Szabolcsi 1984, 1994). In particular, I will argue that the morpheme kí is an agreement morpheme (which I will gloss kí[pl]). In the nominal domain, it agrees with a plural inalienable noun. In the verbal domain, it agrees with a plural verb. A plural verb (often called a pluractional verb) is interpreted as repetition of the event (and so could be called iterative aspect). This article owes much to the ideas and data in Gruber 1973, 1975a,b,c. Although this work builds on Gruber’s in many ways, much of the data here were elicited during my own field work in Botswana in 1996–97, and the summer of 1999. My analysis differs from Gruber’s (1975a) in the analysis of plural deletion, in the analysis of the kí[pl] agreement morpheme, and in adopting the Davidsonian treatment for the semantics of verbs. 1. Inalienable versus alienable possession Many languages have a distinction between alienable and inalienable possession. Alienable possession in ≠Hoan is illustrated in 1.2 [End Page 456] (1) The main morphological indicator of alienable possession is the use of the morpheme ci. In 1, I give examples containing both a singular head noun (1a) and a plural head noun (1b). The issue of the plurality of the head noun will be fully discussed in §§1.3 and 2. Inalienable possession is distinguished by the absence of ci; I give examples in 2 and 3. (2) gya”m-si (*ci) !o child-dim poss stomach ‘the child’s stomach’ (3) gya”m-si (*ci) gye child-dim poss mother ‘the child’s mother’ In ≠Hoan the class of inalienable nouns includes all body part terms, most kinship terms, some spatial relation terms, and some cultural artifact terms. This class corresponds roughly with the crosslinguistic characterization in the literature. I assume here (following Vergnaud & Zubizarreta 1992:596 and Gruber 1975a:40) that inalienable nouns differ from alienable nouns in that they take their possessor as an argument. Gruber says that ‘inalienable nouns are themselves underlyingly predicates’ (1975a: 40). In terms of theta-roles, kyeama ‘dog’ does not assign a theta-role to ≠’amkoe ‘person’ in 1a, but !o ‘stomach’ in 2 does assign a theta-role to gya”msi ‘child’. In the rest of this section, I will give an overview of the different classes of inalienable nouns in ≠Hoan; these are, basically, the Many-class, the Kin-class and the kí[pl]-class. 1.1. The many-class The Many-class consists of nouns that refer to sets of objects that are naturally in a many-to-one relationship with their possessor. This class is illustrated by gyo”ba ‘leaf’ in ex. 4. (4) a. |’on kí- gyo”ba tree kí leaf ‘the leaf of the tree’ b. |’on-qa kí- gyo”ba-(qa) tree-pl kí leaf-pl ‘the leaves of the trees’ Other nouns in this group include: n≠u ‘feather’, kya”a ‘bone’, gu ‘flower’, n|u ‘vein’. Mass nouns are also in the Many-class...

ABSTRACTThe present study investigates the impact of first language (L1) structural frequency and L1 lexical accessibility, manipulated via cognate status, on second language (L2) speech production. L1 German–L2 English speakers and L1 English speakers completed a production task containing pre- and post-modified possessive noun phrase (NP) constructions (e.g. The actress’s sofa vs. The sofa of the actress) in which the head nouns were English-German cognates (e.g. sofa) or noncognates. While English prefers pre-modified NPs, German has a strong frequency bias for post-modified NPs. L2 English speakers exhibited higher production accuracy than L1 English speakers on post-modified NP sentences. However, facilitative L1 effects in production latencies were restricted to post-modified NP sentences containing cognates and only developed cumulatively after repeated exposure to post-modified NP sentences. We discuss how cognate status and L1 structural frequency differentially influence the accuracy and timing of choosing between different structural alternatives during L2 production.

Mandarin Chinese: A Functional Reference Grammar

This paper discusses on the rules of Noun Phrase (NP) in Mandailing language. The objectives are firstly to structurally examine the phrase structure rules (PSR) of NP in Mandailing language and to investigate the constituents of each rules in NP. The method is descriptive qualitative with a case study. The data taken from the native speakers of Mandailing language through interviewing. The data are the utterances of Mandailing language in daily conversation. The analysis is based on the use of X-Bar theory, to see how NPs in Mandailing language are hierarchically governed and presented in tree diagram. Thus, the findings showed that there are five major rules of NP in Mandailing language, (a) single noun as headword, (b) pre-modifier preceedes head noun, (c) post-modifiers follows head noun, (d) head noun between pre-modifier and post-modifier, and (e) complex NP with coordinative conjunction. Also, there is found particle na to relate NP with the complement such as AP. To summarize, Mandailing language has similar patterns of word order with Indonesia language. Overall, this research gives contribution to provide a deep understanding in studying morphosyntactic in linguistics especially phrase structures beyond the effort on exploring and revitalizing a regional language in Indonesia.

The objective of this study is to examine the syntactic structure of Awgni Noun Phrases. The assumption of Labeling Algorithm {XP, H} is holding on, and a descriptive research design was employed to explore the intended objective. Data for this research were enriched by interviewing 12 native speakers of Awgni specializing in the proposed language. Through expert samplings, 20 Noun Phrases were selected and illustrated. Results showed that the Noun Phrases in Awgni could be formed out of the head Nouns all along through other lexical categories reminiscent of the Noun Phrases, Adjective Phrases, Verb Phrases, Determiner Phrases, and Adverb Phrases. These grammatical items were serving as dependents to the head Nouns. The head Nouns in Awgni are for all time right-headed. These heads are the only obligatory constituents, while the Phrasal categories are optional elements which could be either modifiers or complements to the head Nouns. In this regard, Labeling Algorithm explicitly chooses the contiguous Noun heads that are the label of the complete Syntactic Objects (SOs) anticipated for all Noun Phrase structures.

The structure of the noun phrase (NP) is demonstrated to differ among languages. Albeit studies that paid attention on Southern Cushitic languages in general and Iraqw in particular, their contribution is selective to the general grammar of the language while the structure of the NP is scarcely described. This study contributes to the description of the language through an empirical explanation of the elements and the morphosyntactic properties of the NP in the language. It describes the orders of the elements, their co-occurrence, and constraints to illuminate the structure of the NP of the language. It draws on the data collected through elicitation and text collection. The qualitative analysis of the data obtained revealed that Iraqw NP demonstrates four different peculiar characteristics (i) the Iraqw NP is constituted of the head noun with or without dependents. The dependents of the NP are possessives, demonstratives, indefinite markers, the distributive determiner, adjectives, quantifiers, numerals, genitive constructions, and relative clauses. The dependents follow the head noun, except the distributive determiner, which precedes the head noun. (ii) The attested order of the dependents is N>[Poss>Dem/Indef]>[Gen>Adj>Num/Quant]>[Rel]. (iii) Possessive, demonstrative, and indefinite markers are non-lexical words that occur immediately after the head noun. (iv) In spontaneous speech, the NP with up to three dependents is preferred.