- •1. Translation theory: object and objectives
- •3. Types of equivalents
- •It is obvious that in the first sentence it should be the Russian «отношение (к работе)», in the second sentence — «позиции (обеих сторон)», and in the third sentence - «поза (угрожающая)».
- •In such cases the substitute often has a cause-and-effect relationship with the original:
- •In other cases the translator may resort to various types of grammatical substitutes.
- •It is obvious that there can be no equivalence if the original text is clear and unequivocal while its translation is obscure and hard to understand.
- •In Russian constitutions cannot be established but they can be adopted. Therefore:
- •In a transfer-based system the process of translation comprises the following processing steps:
- •Ideally a pivot language-based machine translation will comprise the following processing steps:
- •Agreement
- •7. Intellectual Property Rights.
- •It should be also noted that the syntactical pattern of this style is as important as the vocabulary.
- •Infinitive
- •In any event, members of the association should be prepared to put aside partisan interests if consensus on the abovementioned principles is to be achieved.
- •In Washington there is quiet satisfaction that the French by joining the float have indirectly acknowledged that the u.S. Was right all along.
- •Participle
- •In the carriage there was only an old man reading his paper.
- •It being too late for further discussion, the session was adjourned. Засідання було відкладено, оскільки було занадто пізно, щоб продовжувати дискусію, (adverbial modifier of cause)
- •Inversion
In Russian constitutions cannot be established but they can be adopted. Therefore:
Некоторые из этих стран приняли новые конституции, (or: В некоторых из этих стран были приняты новые конституции.)
- The AFL leaders have a corrupt alliance with the employers.
Since in Russian the usual correspondence to “corrupt” (продажный) can be applied only to human beings, we can have either «преступный союз» or «преступный сговор» or something like that. But we can also preserve the meaning of “corrupt” by referring its Russian equivalent to another word in the sentence:
Продажные лидеры АФТ вступили в преступный союз с предпринимателями .
The country had sincere and successful leaders. В стране честные руководители, добившиеся значительных успехов.
After all, successful leaders are those who have achieved good successes and the original meaning is fully preserved in the translation, though in a rather long-winded manner.
An additional way to deal with the problem of co-occurrence is through a choice of different parts of speech. “A cooperative assistance” is difficult to translate into Russian where «сотрудничающая помощь» is an unacceptable combination. But if both words were translated as nouns the problem would be solved:
We owe this success to the cooperative assistance of the Soviet Union. Мы обязаны этим успехом сотрудничеству и помощи со стороны Советского Союза.
The change in the parts of speech is a common procedure in translation. It often enables the translator to modify his variant to improve its stylistic or emotional effect. So, for ‘The wind was becoming stronger” the translator has the choice of «Ветер дул все сильнее» and «Ветер крепчал», for “I didn’t mean to be rude” he may choose between «Я не хотел быть грубым» and «Я не собирался вам грубить».
Sometimes, the use of a different part of speech is unavoidable: “He was furious” — «Он был в бешенстве»
The elements of the translator’s techniques described above give only a general idea of his professional strategy. Translation is a creative process of search and discovery and it takes much ingenuity and effort to apply the general principles of the translation theory to the practical problems.
Lecture 8. MACHINE AND COMPUTER-AIDED TRANSLATION
This Lecture: • gives a short review of basic machine translation platforms and methods: *• including direct or icon method;
transfer-based and pivot-language based engines and
AI and statistical approaches.
The idea of computer use in translation (hereinafter referred to as machine translation) appeared almost simultaneously with computers. Already in 1949 only five years after the first powerful computer had been put in operation in the USA mathematician and philosopher W. Weaver suggested the use of computers to model the process of translation.
The pioneers of the new research area were mathematicians and programmers and the first stage of computer translation development was characteristic of the so-called «encoding-decoding» approach (see Lectures 2 and 3 of this Manual)48
This approach, which still remains one of the basic methods of translation automation is usually called the direct or icon method.
•► Direct or icon method of machine translation is based on establishing a direct relationship between the source and target dictionary entries.
The target entries are regarded as regular counterparts (icon copies) of the source ones. According to the direct translation method the source and target texts are presumed to be similar both in their form and conceptual content49.
You may validate this assumption yourself having done such word-for-word translation. You will see mat rather more often than not such a non-grammatical translation makes no sense at all for the end user and, therefore, it is to be somehow rearranged and smoothed.
To improve the quality of direct translation the two following methods are usually applied:
syntactic filters and
statistical ranking of translation equivalents to select the most probable ones for the subject matter discussed in a particular document being translated.
Syntactic filters take the form of logical frames (see Lecture 17) in which the slots are filled with syntactic patterns specifying the function of an ambiguous syntactic term in the representation of the source text. Usually in machine translation systems based on the direct method quite a few filters are used to «smooth down» the raw translation.
According to the classification we suggest in this Manual, the second basic method of machine translation is the transfer-based method.
•* According to the transfer-based method of translation grammars of the source and target languages are matched in the process of translation by a set of rules called transfer.
