Search

This experimental study tests the usability of a collocation retrieval tool that combines corpus and dictionary features in order to help learners of Spanish as a Foreign Language to improve their production of collocations. More specifically, the study tests the extent to which such a tool is effective without prior training, and evaluates which of its features might be improved. Two experiments were conducted. The first experiment involved Japanese-L1 learners of Spanish. One subgroup completed a gap-fill test both with and without the collocation tool, while another subgroup used electronic bilingual dictionaries instead of the collocation tool. In a second experiment, we chose a more heterogeneous group of participants with less experience in using lexicographic tools. Results show that learners do not need previous training to benefit from the collocation tool, but also that they do not check corpus examples as often as they might.

This article compares the use of collocations in texts written by native speakers and advanced learners of Spanish. The collocations studied were first identified in the sample texts on the grounds of phraseological criteria and subsequently assigned frequency information corresponding to their occurrence in a reference corpus of Spanish. The distribution of collocations with different frequency of co-occurrence and mutual information scores in the texts of the two samples was then compared, as was their proportion in the collocation repertoire of native speakers and learners. The study revealed significant differences both in terms of frequency and mutual information.

This paper studies the linguistic change whereby Spanish dar prevailed as the default verb to express causation when taking a state noun as its object, displacing alternatives such as hacer vergüenza ‘to cause embarrassment’ or poner miedo ‘to frighten’. In spite of the seeming regularity of the dar-plus-state-noun pattern, it is argued that this collocational pattern has an idiosyncratic character, since not all combinations of lexical units conveying an equivalent meaning are acceptable in present day Spanish, whilst some alternatives that are currently possible in related languages and were commonly found in previous stages of this language’s history (hacer miedo lit. ‘to make fear’) are no longer available. The study tries to show that the idiosyncratic character of these combinations in present-day Spanish can be explained in diachronic terms and that they are the result of two frequency effects that have been observed in other instances of linguistic change: on the one hand, the prevalence as productive patterns of linguistic structures with a high type frequency and, on the other, the resilience of structures with a high token frequency.

ResumenEste trabajo se centra en la diacronia de cuatro formas verbales que, a lo largo de la historia del espanol y hasta el presente, han funcionado como colocativos causativos con nombres del tipo de miedo. La eleccion de estos colocativos obedece a que, en contraste con otras alternativas a dar en el contexto senalado, como hacer o poner, los colocativos en cuestion no han cedido frente a la primera forma, sino que se conservan en espanol contemporaneo y, en ciertos casos, han aumentado su productividad considerablemente. Para explicar este hecho, se argumenta que meter ha sobrevivido en el contexto estudiado como consecuencia de la repeticion de unas pocas secuencias concretas. El caso de los otros tres verbos es distinto y se propone que su productividad esta ligada al hecho de que, a diferencia de otros colocativos causativos, su significacion no esta contextualmente condicionada.

espanolEl objetivo del presente trabajo es determinar si existe alguna relacion entre la correccion de las colocaciones producidas por aprendices de espanol en textos escritos y la asociacion que sus miembros presentan en un corpus representativo de esta lengua. Para ello se partira de una muestra de colocaciones extraidas de un corpus de aprendices, las que han sido identificadas desde presupuestos fraseologicos, y se estudiara su frecuencia en un corpus representativo del espanol. A partir de estos datos, se comprobara si existe alguna relacion entre la asociacion de los miembros de una colocacion en terminos de frecuencia de coaparicion e informacion mutua y la probabilidad con la que aparece producida correctamente en los textos de aprendices. EnglishThis study aims to establish whether there is any relation between the correctness of collocations present in the writings of learners of Spanish and the association their members display in a representative corpus of this language. In order to do that, starting from a sample of collocations extracted from a learner corpus where they have previously been identified by means of phraseological criteria, I study their frequency in a representative corpus of Spanish. Based on these data, I try to establish if the association between the constituents of a collocation, as measured in terms of frequency of co-occurrence and mutual information, has any influence on the probability wherewith learners produce such collocation correctly

This article studies the use of support verb constructions (SVCs) in the written production of learners of Spanish. SVCs are lexical combinations whose content is similar to verbal predicates but is distributed between a verb and a noun, the noun being the carrier of the core lexical meaning of the predicate. Although there is considerable agreement on the importance of these constructions in the learning process, their use in the production of learners of Spanish has so far attracted little attention. This study examines the difficulties posed to learners by this construction by means of a qualitative analysis of the errors registered in 3 samples consisting of essays by learners with 3 different mother tongues (English, Swedish, and Japanese). It focuses on 3 types of error, 2 of which-the support verb choice and the determiner choice-seem to be especially problematic due to the unpredictability of the units involved. The third type-using an SVC instead of a more idiomatic 1-word verb-is regularly found only in the samples of the Japanese speakers, which suggests the influence of a particular mother tongue in its production. [ABSTRACT FROM AUTHOR]

Resumen: Este artículo propone un método para nivelar las colocaciones del Diccionario de Colocaciones del Español de acuerdo con los niveles propuestos en el MCER. Como criterio nivelador se ha usado la frecuencia que las colocaciones del diccionario presentan en un corpus. Mediante el análisis de una muestra de colocaciones contenidas en el diccionario y en el Plan Curricular del Instituto Cervantes se comprueba una correlación negativa entre la nivelación propuesta para esas colocaciones en el Plan Curricular y la frecuencia de corpus. Tal correlación justifica la pertinencia de la frecuencia como criterio nivelador. Finalmente, se lleva a cabo un examen cualitativo de los resultados para comprobar la validez del método propuesto.

This paper studies the role of two different types of motivation that have been proposed to explain the use of subject personal pronouns in Spanish, namely their function as indications for the addressee to identify the subject's referent, and their suitability for expressing informational values such as contrastiveness or focus. This study focuses exclusively on third-person forms and relies on conversational data. The distribution of third-person pronouns is analysed combining qualitative and quantitative approaches. It will be argued that the informational and referential properties of subject personal pronouns are by themselves insufficient to account for their expression, their occurrence depending crucially on their activation through the previous use of units of the same type.

This paper presents a new multilingual corpus with semantic annotation of collocations in English, Portuguese, and Spanish. The whole resource contains 155k tokens and 1,526 collocations labeled in context. The annotated examples belong to three syntactic relations (adjective-noun, verb-object, and nominal compounds), and represent 58 lexical functions in the Meaning-Text Theory (e.g., Oper, Magn, Bon, etc.). Each collocation was annotated by three linguists and the final resource was revised by a team of experts. The resulting corpus can serve as a basis to evaluate different approaches for collocation identification, which in turn can be useful for different NLP tasks such as natural language understanding or natural language generation.