USING NATIONAL CORPORA IN TRANSLATION Текст научной статьи по специальности «Языкознание и литературоведение»

ФИЛОЛОГИЧЕСКИЕ НАУКИ

USING NATIONAL CORPORA IN TRANSLATION

Mamedzade Sevinj Salim

Odlar Yurdu University, Azerbaijan, K.Rahimov street, 13, 1072 Baku

Abstract. Nowadays, English language is one of the most important languages in the world. It is spoken not only in English speaking countries, but also in other countries of the world. It is the official language of huge countries. People use this language more often than we can imagine. Day after day need for good translation process is increasing. We live in the world of up-to-date technology with their various areas helping for our private and business lives. Using corpora have a great help and significant role for translators. This article deals with problems in translation process. The main idea of this article is to show the benefits of corpora as translation aides. The aim here is to illustrate the ways that translator may use to improve the quality of translation: finding a lot of collocations, adjectives, nouns and other terms.

Keywords: corpus linguistics, machine translation, parallel corpora, bilingual dictionaries.

Ключевые слова: корпусная лингвистика, машинный перевод, параллельные корпуса, двуязычные словари.

In recent years corpus linguistics is a lively subject and there are many links between corpus linguistics and translation practice. Nowadays modern computers are the most important tools for translator regardless he or she is a technical translator, legal translator or literary translator. Monolingual and bilingual dictionaries are the main helpers as well as encyclopedias in electronic format. Machine-readable corpaora have a number of advantages over other forms of storage. Firstly, machine readable corpora may be searched which is simple and possible. Secondly, it can be enlarged and enriched with additional information.

National language corpora can be named as one of the significant modern sphere in linguistics during the last decade. Nowadays without the information from national corpora it's difficult to imagine any linguistic research. Using the corpus data lets us to enrich the process of studiing the foreign landuages, creating new methodological materials and, of course, the process of translation. At the same time, it's a fact that lots of young translators neglect so valuable linguistical information with knowing a little about national linguistic corpora. [4]

First of all, it's important to define what is national corpora. In broad context- it's the compact of texts connecting with the same characteristics; genre, authority and so on. What about the language corpora in narrow context, it's the unified massive of language data in computer.[5]

Within many features we can underline the following features of corpora.

• Electronic- corpus should be in electronic form.

• Representative- it should represent the object well.

• Marked up- the main difference of corpus from text collection.

• Pragmetically oriented- it should be formed under the certain issue.

We can classify the corpora according to different signs: according to their formation, types of language data, genre, size of texts and so on. According to their parallel criteria corpora may de divided into: monolingual, bilingual,morelingual corpora.

V. Rikov underlines that the safest results during the research can be given only by universal corpora of texts, which the natrional corpus is. [3] National corpus is the collection of texts in computer presenting the language in the stage of its existing period. [2] The searching system of national corpora lets us to vary the options of request; to choose the grammatical or semantical characteristics and to define the chronological borders. It's important that many of contemporary corpora (as BNC -British National Corpus, COCA- Corpus of Contemporary American English) marked up according to the genres of texts, so the request of one word or expression may be limited with concrete type of discourse. Of course, it will be valuable to treat with corpus material if you need the meaning of word which is not found in dictionary. However, the data of national corpora may be usefull to itemize the meaning that the words fixed in dictionaries. Practice shows that even with rich dictionaries young translators have difficulties with choosing the right equivalent of word within the line of given equivalents. Another important aspect of using the corpus data during the translation is clarifiing the compatibility of words in target language. Even the authoritative dictionaries are not able to include all the lexems of the language. They can only give the general recommendation for combining lexical words and they don't give full contemporary condition of the language. Lots of contexts made up national corpora give more chances to estimate the real patterns of compatibility of words in modern speech. Considering this, for getting the exact imagination of position of words in contemporary English language, the datas of both languages are recommended to use in combination. Today translators can investigate the information in corpora. All the information in corpora may help the translator to search exact word or phrase by clicking the screen. One of the best known corpus of British English is BNC

(British National Corpora), 100 million word collection of samples. Another important point of corpora is that they should be open for texts can be added.

For choosing the equivalent during the translation parallel corpora are also more usable, for ex: text corpora in English language with parallel translation. These corpora give us chance to use the results of other translators. They consist of published translations, as well as english or russian translations. That's why the information may be the key factor for choosing the equivalent. Let's analyze the word "сало" (Russian word). The translation of this word may cause difficulties for translator. Dictionaries give us the following equivalents:

• Fat (soft flesh on the body)

• Suet (fat on loins of cattle)

• Lard (pig fat)

• Tallow ( for candles)

The most important fact is ignored in this situation, that in Russia the word "сало" is used for sliced eatable raw fat which is used in meals. In requesting to parellel english-russian corpora, we can get a line of contexts where we can meet this lexem with parallel english variations. Analyzing this list we can see that for translation of the word "сало", sliced and used as raw for meals, is used the equivalent "bacon". At the same time these contexts from English language into russian demonstrates the cultural difference between these 2 words: the word "bacon" english people use for roast form , however russian people use its raw form. These datas prove that lots of words need more sensitivity in different contexts.[4] Modern corpora are divided into subcorpora with the limited genres. (COCA is subdivided into the following genres: spoken, fiction, magazine, newspaper and so on.)

Parallel corpora can be compared to bilingual dictionaries with differnces. Parallel corpora can provide information that bilingual dictionaries usually don't contain. Translation corpora or parallel corpora are widely used in the field of computational linguistics, especially in machine translation.

Paralell texts are the main helpers of translation process. Transferring the parallel texts to machine form and organizing them as a form of electronic corpora carries advantages (huge amount of materials, their variety according to their style and genre and so on).

Using parallel corpora influences well to translation process. We can meet different variations of a lexem except of one and the same word form. Besides, traslators need resources that they can use for perfect translation. Translators don't waste their time by searching the word forms in dictionaries. They can save 80% of their time. Parallel corpora and computer technology minimize the time for translators in order to get professional results.

As the searching system of corpora let's analyze the usage of word or phrase in each of the genres. It helps us to get the full imagine about the stylistical characteristics of language forms and may help the translator in that case when the dictionaries don't give us the stylistical charcter of this or that word form. For ex: the word form "well-nigh" is given as "почти, приблизительно" [6]

Overall, we may say that the right usage of corpora may help the translator to qualify the meaning of the word and its combinations. There are number of ways in which specialised corpora can help the translator. We can use the word list in corpora to learn about the subject we are translating, and about the most common lexical and grammatical patterns through collocates and clusters. It involves a great development in the documentary sources for the translator. We believe that corpora help the students to develop their own capabilities on translation, and they use corpora perfectly to become specialised translators.[1]

References:

1. Bermudez M. B.The importance of corpora in translation studies, 2016 (pdf) https://files.eric.ed.gov

2. Podlesskaya V.I. Modern computer methods in the study and the teaching of linguistic disciplines: corpus linguistics. Electron.( In Russian)

http://www.rsuh.ru/article.

3. Rykov V.V. Corpus linguistics and lexicography - the problem of representativeness // Transactions of the conference on lexicography. Grodno, 2002. Electron. (In Russian)

http://rykov-cl.narod.ru.

4. Semina O.Y. On the use of data from the national linguistic corpus in translation. Electron.(In Russian)

https://cyberleninka.ru/article

5. Zakharov V.P. Web space as a language corpus. Electron. (In Russian)

http://www.dialog21.ru/Archive/2005/

6. https.//www.lingvolive.com/translate