Corpus Search Research Articles

Kollokationen in der Übersetzerausbildung: ein korpuslinguistischer Ansatz

AbstractStudents usually learn very quickly to deal with pragmatic and cultural translation problems, but very frequently lapse into repeating linguistic translation errors. A considerable amount of these errors are related to the use of collocations. This paper focuses on how to solve translation problems related to collocations by using corpus search techniques. Firstly, it gives an insight into the achievements obtained by using corpus linguistics methods in the teaching of a foreign language and in the training of translators. Secondly, different definitions of collocations will be presented and the strengths and limitations of different methods for the extraction of collocations from corpora will be commented. To round off this article, different parallel and comparable corpora will be presented and some corpus-based and corpus-driven exercises with comparable and parallel corpora will be designed to solve translation problems related to collocations.

Copy Raising Construction in English: A Usage-based Perspective

In accounting for the so-called copy raising (CR) in English, the movement perspective has assumed that the embedded subject of the CR verb`s sentential complement is raised to the matrix subject, leaving behind its pronominal copy. This kind of movement-based analysis raises both empirical and analytical issues, when considering variations in the pronominal copy constraint. This paper investigates the actual uses of the construction, using online-available corpora. Based on this corpus search, we classify two different types of copy raising predicates (genuine and perception), and discuss their grammatical properties in detail. We suggest that the simple copying rule couched upon movement operations is not enough to capture great variations in the uses of the construction, and show that interpretive constraints, e.g., perceptual characterization condition, play an important role in licensing the construction.

Learners’ search patterns during corpus-based focus-on-form activities

Our research explores the search behaviour of EFL learners (n=24) by tracking their interaction with corpus-based materials during focus-on-form activities (Observe, Search the corpus, Rewriting). One set of learners made no use of web services other than the BNC during the central Search the corpus activity while the other set resorted to other web services and/or consultation guidelines. The performance of the second group was higher, the learners’ formulation of corpus queries on the BNC was unsophisticated and the students tended to use the BNC search interface to a great extent in the same way as they used Google or similar services. Our findings suggest that careful consideration should be given to the cognitive aspects concerning the initiation of corpus searches, the role of computer search interfaces, as well as the implementation of corpus-based language learning. Our study offers a taxonomy of learner searches that may be of interest in future research.

Foreword

Foreword

Pedagogic Corpora for Content and Language Integrated Learning. Insights from the BACKBONE Project

BACKBONE is a European LLP/Languages project (1) (Jan 2009 - Feb 2011), whose overall objective is to provide foreign language teachers in CLIL settings with innovative language learning solutions. To achieve this goal, pedagogic corpora of spoken interviews are combined with corpus-related e-learning activities in blended learning scenarios. The seven BACKBONE corpora contain video interviews in English, German, French, Polish, Spanish and Turkish as well as in European manifestations of English as a Lingua Franca (ELF). The interviews have been transcribed and pedagogically annotated with regard to thematic and linguistic features; additional enrichment resources include ready-made language learning modules as well as suggestions and instructions for exploratory and communicative learning activities. The BACKBONE search interface provides free online access to the interviews and enrichment resources. It supports pedagogically motivated searches using thematic and linguistic categories as well as lexical searches with words and phrases. 24 CLIL-related pilot courses have been implemented and evaluated in secondary, higher and vocational education; they demonstrate how BACKBONE search results can be used to facilitate individual and collaborative learning in Moodle-based blended learning activities. A suite of pedagogic corpus tools covering transcription, annotation, management of enrichment resources and corpus search is available under a GNU General Public License. The customization and flexibility these tools offer enables teachers to cater to diverse language learning and teaching needs in CLIL contexts or in connection with lesser taught languages and varieties. To facilitate exploitation, the BACKBONE website serves as a ‘one-stop-shop’ for an ensemble of teacher support facilities including web support for the development and hosting of “guest” corpora and courses.

Using concept mapping and the web as corpus to develop terminological competence among translators and interpreters

“Übersetzen ist die Verwendung des Verstandenen,” wrote the late Hans Vermeer in 1986. Seen within the context of Vermeer’s Skopos theory, translation is at its core a dual process of understanding and applying this understanding in the production of a target text. I intend to discuss the essence of Vermeer’s quote against the background of specialized translation and interpreting. In particular, I will be focusing on the two main aspects of translation highlighted in the quote, the cognitive dimension of translation as information processing on the one hand and the pragmatic dimension of the purposeful application of that information in the transfer and production phases of the translation process on the other hand. At the heart of my paper lies a didactic model for the development of information processing and information application competence in the university training of future translators and interpreters. The main focus of the discussion will be on terminology in general and, following Teresa Cabré’s “theory of doors,” on the cognitive and pragmatic dimensions of terminological units in particular. The objective is to combine terminological (concept-oriented) approaches with textographical (corpus-driven) ones, so as to show, first, how terminological units can be organized in mono and multilingual knowledge structures and, second, how, based on a digital corpus search, these units can be implemented into the norm-guided production of the target text.

Learner–Corpus Interaction: A Locus of Microgenesis in Corpus-assisted L2 Writing

This paper examines the processes through which learners interact with a corpus system and microgenetic development emerges from the interaction. The corpus system described in this paper is capable of retrieving highly relevant textual examples tailored to individual needs. Data were collected from an undergraduate ESL composition course in North America for one semester while students interacted with the system. Using exploratory techniques that triangulate on real-time screen recordings, corpus queries, and oral/written reflections, this paper documents these interactions as evidencing each learner’s efforts to resolve issues by retrieving, evaluating, and appropriating the corpus search results. The findings show that the learner’s achievements depend both on their ability to interpret and exploit the search results and on the corpus system’s ability to respond to the learner’s particular needs. Based on these findings, this paper suggests that corpus-based research should focus on maximizing the learner’s developmental potential by developing more advanced corpus systems and enhancing the learner’s textual awareness and analytical ability.

The grammaticalization of the have perfect in Dutch – a corpus study of contextual extension and semantic generalization

The article investigates the grammaticalization of the have perfect in Dutch by means of a corpus study of historical legal texts dating from the middle of the thirteenth century until the end of the eighteenth century. The focus of the investigation is on the gradual extension of the have+past participle construction in contexts that were not attested before. The study of the status of the subject, the direct object and the past participle in the corpus shows that the construction is increasingly used in a wider array of contexts. Moreover, the corpus search indicates that meaning components of the have+past participle construction are lost in the process of contextual extension. More specifically, the construction is increasingly used in the background of the discourse in order to expand on events that happened before the time of reference.

Poniendo de manifiesto: Spanish light verb constructions in Latin American newspapers

This study investigates Spanish light verb constructions (LVCs), or construcciones con verbo soporte, such as dar respuesta and tomar en cuenta in a seven-million-word corpus of contemporary newspaper texts from seven Latin American countries. I provide a detailed discussion of the procedure of compiling the corpus to ensure a balanced representation of newspaper registers. This is followed by a description of how the targeted verbs, dar, poner, tener and tomar, were identified and tagged using corpus analysis software. The results of the corpus search are analysed in terms of frequency, and selected morphosyntactic and textual features, such as the distance between components, tense, number and mood of the light verb, adjectival modification, article use, complex LVCs, and the occurrence of multiple LVCs. The results, supported by numerous examples in Spanish from the corpus, provide further evidence of the wide variation in form and use found among many LVCs.

Poniendo de manifiesto: Spanish light verb constructions in Latin American newspapers

This study investigates Spanish light verb constructions (LVCs), or construcciones con verbo soporte, such as dar respuesta and tomar en cuenta in a seven-million-word corpus of contemporary newspaper texts from seven Latin American countries. I provide a detailed discussion of the procedure of compiling the corpus to ensure a balanced representation of newspaper registers. This is followed by a description of how the targeted verbs, dar, poner, tener and tomar, were identified and tagged using corpus analysis software. The results of the corpus search are analysed in terms of frequency, and selected morphosyntactic and textual features, such as the distance between components, tense, number and mood of the light verb, adjectival modification, article use, complex LVCs, and the occurrence of multiple LVCs. The results, supported by numerous examples in Spanish from the corpus, provide further evidence of the wide variation in form and use found among many LVCs.

Grammatische Illusionen – Lokal wohlgeformt – global deviant

Abstract Grammatical illusions are linguistic equivalents of optical illusions in visual processing. A grammatical illusion amounts to an expression that is judged acceptable in spite of its grammatical deviance. It is the hitherto insufficiently appreciated fourth “aggregate state”, that completes the set of four logical pairings of the properties “(un-)grammatical” and “(un-)acceptable”. A typical source of syntactic illusions is the mental processor's attempt to repair grammatical inconsistencies resulting from contradicting requirements of syntactic regularities. The resulting pseudo-solutions tend to be appraised as acceptable, although they remain patently ungrammatical. The consequences of this phenomenon, when acknowledged, are evident: Acceptability turns out to be neither a necessary nor a sufficient property of a grammatical construction. The attempt of modeling any construction that is (introspectively) judged acceptable as grammatically well-formed is tantamount to the attempt of devising a grammatical derivation for constructions of arbitrary (un)grammaticality. Informant consultation results are contaminated and of limited value. Syntactic investigations (and semantic ones as well) deserve more reliable standards of data assessment than mere sampling of self-perception testimonies and corpus search.

Polish aspectual prefixes, their order and functions: A study based on the National Corpus of Polish

The paper presents the results of a study of Polish perfectivizing prefixes and their combinations based on bilingual electronic dictionaries and data from the National Corpus of Polish. A dedicated corpus search engine used in the study makes it possible to find verbal prefixes within a near context of homonymous prepositions and thus to determine the extent to which the prefix meaning is still connected to the meaning of the cognate preposition. By comparing the general frequency of prefixes in aspectual pairs with the number of possible prefix combinations it is possible to identify the most productive 1 prefixes in the Polish aspect system and compare their role as perfectivizers with their cognates in other Slavic languages. The possibilities for combining verbal prefixes evident in a recent Polish dictionary are presented along with the corpus analysis. Both analyses show that the verbal prefixes that combine with others most freely in verbal derivatives are usually only weakly lexicalized, and function as pure perfectivizers.

English Conditional Inversion: A Construction-Based Approach

Conditional sentences also can be formed by inversion of subject and auxiliary, but it happens only in a limited environment. This paper addresses grammatical constraints in conditional inversion and how they behave differently from the regular conditional clauses based on corpus investigations. Our corpus search reveals many different types of conditional inversion constructions, indicating the difficulties of deriving inverted conditionals from movement operations. In this paper, we provide a construction-based approach to the inverted conditional construction. The paper shows that the most optimal way of describing the general as well as idiosyncratic properties of the inverted conditional constructions is an account in the spirit of construction grammar in which a grammar is a repertory of constructions forming a network connected by links of inheritance.

TPEMatcher: A tool for searching in parsed text corpora

Recently, due to the widespread on-line availability of syntactically annotated text corpora, some automated tools for searching in such text corpora have gained great attention. Generally, those conventional corpus search tools use a decomposition-matching-merging method based on relational predicates for matching a tree pattern query to the desired parts of text corpora. Thus, their query formulation and expressivity are often complicated due to poorly understood query formalisms, and their searching tasks may require a big computational overhead due to a large number of repeated trials of matching tree patterns. To overcome these difficulties, we present TPEMatcher, a tool for searching in parsed text corpora. TPEMatcher provides not only an efficient way of query formulation and searching but also a good query expressivity based on concise syntax and semantics of tree pattern query. We also demonstrate that TPEMatcher can be effectively used for a text mining in practice with its useful interface providing in-depth details of search results.

Non-coordination-based ellipsis from a Construction Grammar perspective: The case of the coffee construction

Abstract This paper focuses on English yes-no interrogatives of the type Would you like/Do you want [NP]? and their elliptical variants You like/You want [NP]?, Like/Want [NP]? and [NP]?. The central question is what type of theoretical relationship can be assumed between the different forms. A theoretical discussion of different traditional approaches and their limitations is followed by the presentation of an explorative corpus search in the BNC, which can reveal interesting differences between the different stages of initial reduction, in particular regarding the semantic class of the NP head nouns: While the possible lexemes found in the full forms and intermediate reductions seem only restricted by pragmatic constraints, the highly reduced [NP]? question displays only lexemes related to the semantic class of consumption goods. Construction Grammar (CxG) is suggested as a useful approach to explain the internal relationships between the different forms, especially because it offers the option of lexically open idiomatic structures, assumes a usage-based, hierarchical network with different degrees of independence between single constructions, and allows for an integration of pragmatic features into a grammatical model.

La Base de Datos Terminológica Jurídica Bilingüe Diseñada Para el Traductor

The majority of legal dictionaries or databases, both explanatory and translation, do not specify their target audience. Moreover, the scarcity of bilingual legal dictionaries and databases is striking –in particular when the language in question is a minority language such as Dutch– even though it is the optimum tool for the translator. Hence the need for bilingual legal tools, in this case Spanish/Dutch tools, specifically designed for the translator. The translator must find equivalents for all the units in the source text that are of interest from a contrastive standpoint. This article deals with the problem of such units starting from a well defined geographic context (Spain/Dutch speaking part of Belgium) and a specific type of text (articles of association), on the basis of a corpus and electronic search tools. In the end we distinguish three types of units for our electronic database JuriGenT: 1) terminological units, which are units with clear juridical contents but completed with units that border common language use 2) phraseological units, which are units selected on the basis of morphosyntactic and syntactico-semantic criteria and which always contain a juridical term, and, 3) connecting units, which are units that only gain juridical meaning in combination with paradigms of terminological units. The selection of units for our database JuriGenT containing these three types, in particular the phraseological and connecting units, which often demonstrate lexical syntactic and/or lexico, syntactic assymetries between source and target unit, contribute to an idiomatic translation.

Exploring English with Online Corpora. An Introduction. Wendy Anderson and John Corbett.

This textbook by Wendy Anderson and John Corbett is presented as a low threshold introduction into the main linguistic properties of English based on the use of online corpora. It addresses a non-specialist readership. The book is subdivided into eight chapters that successively deal with the nature and analysis of English corpus data (Chapters 1 and 2); lexical, grammatical, discourse, and phonetic properties of English that can be explored through online corpora (Chapters 3, 4, 5, and 6), the contextualization of corpus data and their use in education and research (Chapters 7 and 8). There is an appendix with a brief description of 18 online corpora, a term list explaining the main technical terms used in the book, a reference list, and an index. Each chapter clearly defines its objectives and identifies a number of interesting research questions, which are then answered on the basis of results yielded by corpus searches. Readers are also set a number of tasks, which are mainly searches they have to carry out themselves to reach a conclusion about the use of particular linguistic items and structures. Each chapter suggests further readings on the topics it discusses.

A corpus search methodology for focus realization.

We describe a methodology for investigating the semantic‐grammatical conditioning and phonetic realization of contrastive intonation using a web harvest of particular word strings followed by grammatical and acoustic analysis. A commercial audio web search engine using speech recognition retrieved 179 MP3 files purportedly containing a token of the string “than I did.” In this comparative clause fragment, contrastive focus commonly falls on the subject (she did more than IF did), on “did” (I wish I had done more than I didF), or following (I said more now than I did beforeF). The 96 true tokens of “than I did” were classified into the categories “subject,” “did,” and “following” by grammatical and semantic criteria. For each token, 5 segment intervals were hand‐annotated and more than 300 acoustic parameters extracted using a Praat script. SVM machine learning classifiers were trained that identify focus classes by acoustic criteria. On a 10‐fold cross‐validation test, the classifier achieves 90.2% accuracy in discriminating the dominant “subject” and “following” classes. In a listening task, human subjects achieved comparable accuracy of 90.3‘ given only the acoustic target “than I did.” Stepwise logistic regression revealed measures of duration, f0, intensity, formants, and formant bandwidths among the significant factors.

Using a Corpus‐Based Lexicogrammatical Approach to Grammar Instruction in EFL and ESL Contexts

This article reports on a study that examined the effects of integrating corpus and contextualized lexicogrammar in foreign and second language teaching. The study was conducted in English as a foreign language (EFL) and English as a second language (ESL) courses at 1 Chinese university and 2 U.S. universities, involving 244 participants (236 EFL/ESL students and 8 instructors). A variety of data was collected, including students' corpus search projects and reflection papers, teachers' lesson plans and teaching journals, and a poststudy assessment survey. A close analysis of the data reveals several positive effects of the approach, such as improved command of lexicogrammar, increased critical understanding of grammar, and enhanced discovery learning skills. It also reveals some challenges of corpus‐based lexicogrammar learning, including the daunting difficulty many students feel in corpus analysis. The study also identifies some variables influencing learners' experience in using the approach, such as course content, student learning styles, and learning settings. Implications for pedagogy and further research are also discussed.

Datibo komunztadura beti zaindu, inoiz zaindu ez eta batzuetan baino zaintzen ez denean. Hiru ahoko aldagaia, datu iturri bi, eta erreminta bat : Corsintax

In this paper, we analyze dative agreement drop, a phenomenon attested in North-Eastern varieties of Basque. More precisely, we distinguish three main settings of this syntactic variable with respect to agreement obligatoriness. In the first setting, dative agreement is optional. This setting is found with (i.) quirky dative objects in absolutive/dative alternating verbs, (ii) dative-postpositions in postposition/dative alternating verbs, and (iii) aspectual datives. In the second setting, dative agreement is obligatory rather than optional, and here we distinguish (iv.) dative-experiencers, (v.) dative-possessors in inalienable possession and ethical datives, and (vi.) causees. The third and last setting corresponds to datives which may never agree, that is, adjunct type datives. We’ll obtain the main data from the historical dictionary Mitxelena and Sarasola´s (1989-2005) Orotariko Euskal Hiztegia (OEH) and the modern literature corpus Ereduzko Prosa Gaur (EPG). We’ll also present a corpus search tool which retrieves syntactic information from this corpus : Corsintax.