There are some criteria for classifying translation:

The first one is based on who does the translation. These days translation may be done by a human translator or by computer.

Form of speech: according to this criterion, translation as a written form, sight translation (or translation-at-sight, on-sight translation) as the oral translation of written text, and interpreting as oral translation of oral discourse are differentiated. This criterion also involves subtitling, that is visual translation involving the superimposition of written text onto the screen, and dubbing, or the replacement of the original speech by a voice track which attempts to follow as closely as possible the timing, phrasing and lip movements of the original dialogue.

Source text perception: a translator can see or hear the text.

Time lapse between the source text perception and translation: consecutive and simultaneous interpreting.

Number of languages in translation situation: one-way or two-way translation.

Direction of translation: direct translation, that is, translation into the mother-tongue, and inverse translation, or translation into a foreign language.

Methods of interpreting: note-taking interpretation, phrase-by-phrase interpretation

Functional style and genre of the text: literary works and informative texts.

§ 2. MACHINE TRANSLATION

The first idea of machine translation is known to have been expressed in 1933 by the Soviet engineer Petr Smirnov-Troyansky but it is not he but Warren Weaver who is credited as the founding father of Machine Translation (MT) research.¹¹⁷ R o u t l e d g e E n c y c l o p e d i a… - P.140.⁷ The first demonstration of an MT system took place in 1954 in Georgetown University, U.S.A., where the experiment of making a computer translate words from Russian into English was conducted.

Machine translation is based on analysis and synthesis operations and has required many years of hard work and frustrations. Sometimes the end-product of the machine translation was so ridiculous (like Out of sight, out of mind. - Слепой идиот), that in the 1960s there happened a machine translation `recession'. However, with third-generation computer systems emerging in the 1970s, interest in machine translation was revived. Word-processors appeared and today's translators cannot imagine their lives without them.

Today, machine translation is often called computer-aided translation (CAT). CAT systems are divided into two groups: machine-aided human translation (MAHT) and human-aided machine translation (HAMT). The difference between the two lies in the roles of computer and human translator.

In MAHT, a translator makes the translation, then uses the computer as a tool for typing, checking spelling, grammar, style; for printing the target text, for looking up words in electronic dictionaries and data bases, for getting references on CD-ROMs and other sources, for consulting about contexts, for discussing problems in the web, for seaching a job, etc.

In HAMT, the translation is automated, done by a computer but requiring the assistance of a human editor. There are two phases of human help: pre-editing and post-editing. In pre-editing, an operator (or a customer) prepares the text for input. A special computer translation program transfers the text from one language to another. Then a translator does the post-editing, mostly by correcting the word usage.

Machine translation has a number of advantages and disadvantages. The advantage is, first and foremost, its fast speed, which saves time, so important these days. The computer is tireless; it can work day and night. Now that there are lap-tops, a computer is a very flexible and convenient tool: it can accompany a translator anywhere. Computers are also of great help to disabled people, especially computers working with a human voice.

On the other hand, computers are restricted to the materials. They can translate only clichйd texts. They cannot translate unpredictable texts, like fiction, for example. Usually they provide `raw translation'. Another disadvantage is that they are still rather expensive. They require constant upgrading, which is usually not cheap. Computer viruses are a serious danger to work. And computers are not absolutely safe for human health, either.

§ 3. TRANSLATION AND INTERPRETING

Difference in written translation and interpreting has been fixed by two international professional associations: F.I.T. (Fйdйration Internationale des Traducteurs) or the International Federation of Translators, the association of written translators; and A.I.I.C. (Association Internationale des Interprиtes de Conferйnce), or the International Association of Conference Interpreters, dealing with oral translation.

As is seen from the name of the professional association, interpreters are often called conference interpreters, though their functions can be much broader. Conference interpreting is known to have started after World War I, at the Conference on the Preliminaries of Peace in 1919. Until then all international meetings had been held in French, the language of 19^th century diplomacy.

The first conference interpreters did consecutive interpreting, i.e. they delivered their translation after listening to the speaker so that there was some time between the source language text and the translation. The interpreters worked in teams of two, each into his mother tongue. At the League of Nations, interpreters went to the rostrum to deliver their translation as soon as the speaker had finished. Occasionally speeches lasted well over an hour, so the interpreters, considering it bad taste to interrupt a speaker, developed a technique of consecutive interpreting with note-taking.

Two Geneva conference interpreters, J.-F. Rozan and J. Herbert, after having reviewed their own as well as their colleagues' writing pads, came to the conclusion that although each interpreter had his or her own manner of writing, there was something common to all the notes reviewed. This brought to life recommendations to would-be interpreters on how to take notes in order to memorize the message and not to interrupt the speaker.¹¹⁸ Ч е р н я х о в с к а я Э. А., С л а д к о в с к а я Е. Н. Учебное пособие по теории перевода английского языка. - М.: МГПИИЯ им. М.Тореза, 1985. - С.64-65.⁸

Unlike shorthand, an interpreter's system of note-taking or speedwriting is not a word-for-word recording of speech. It is based on the conceptual representation of the message utterance by utterance and helps to single out the main idea of the speaker. The main principles of note-taking are as follows:

only key-words and the so called `precision' words (i.e. words conveying unique information, e.g., proper names, statistics, etc.) are put down;

words are contracted (vowels are omitted, the so-called Arabic approach);

special symbols are used;

the syntactic structure has a vertical progression:

Subject group

Predicate

Object

Object (homogeneous parts of the sentence are written one under the other).¹¹⁹ Ч у ж а к и н А., П а л а ж ч е н к о П. Мир перевода, или Вечный поиск взаимопонимания. - М.: Валент, 1997. - С.97-119.⁹

The `sentence-by-sentence' interpreting often found in liaison and community interpretingб is not regarded now as `true consecutive'.²²⁰ R o u t l e d g e E n c y c l o p e d i a ... - P.41.⁰ Liaison interpreting takes place in spontaneous conversational settings,²²¹ H a t i m B., M a s o n I. The Translator as Communicator. - London & New York: Routledge, 1997. - P.219.¹ while community interpreting is typical of the public service sphere.²²² R o u t l e d g e E n c y c l o p e d i a… - P.33.²

These days consecutive interpreting is used mostly in bilateral contacts, to serve only two languages.

Interpreting may take place in two directions when the interpreter has to work for both language participants. This is a two-way, or bidirectional, translation (interpretation) and it requires a special skill of switching the languages to speak to, suppose, a Russian participant in Russian and to an English participant in English and not vice versa. A one-way interpreting means translation from one language only and is usually employed for summit meetings.

There is a sub-variety of the consecutive interpreting, known as postponed consecutive interpreting. This is a translation which is not performed in the presence of the participants, but which is dictated from the interpreter's notes into a dictating machine or typed, in case the participants have understood the speaker but want to think over the discourse to take appropriate decisions on it.

Consecutive interpreters are also called linear interpreters, for their translation is in line with the source text unlike simultaneous translation that overlaps the original speech.

Simultaneous interpreting, i.e. interpreting almost immediately as the speaker produces the text (the interpreter can lag behind the speaker not more than 2 or 3 seconds), came into life much later, at the Nuremberg trials (1945-1946) and Tokyo trials (1946-1948) of war criminals, though some attempts had been made in the late 1920s and the early 1930s. In the USSR, simultaneous interpreting was first introduced at the 6^th Congress of the Communist International in 1928, with the interpreters sitting in the front row of the conference hall trying to catch speakers' words coming from the rostrum, and talking into heavy microphones hanging on strings from their necks.²²³ Ч е р н я х о в с к а я Э. А., С л а д к о в с к а я Е. Н. Указ. соч. - С.34.³ Isolated booths for interpreters appeared five years later, in 1933.

Simultaneous interpreting gained ground at the United Nations Organization that began the era of multilateral diplomacy. Today's simultaneous interpreters, unlike their predecessors, are provided with special equipment. They work in a special booth, listening through a headset to the speaker in the conference room and interpreting into a microphone, while at the same time watching what is going on in the meeting room through the booth window or viewing projections on the TV screen. Delegates in the conference room listen to the target-language version through a headset.

Simultaneous translation is usually employed at multilanguage (multilateral) meetings, so that conference participants can switch their headphones to the appropriate language channel.

Simultaneous interpreting is very exhausting work. It requires extremely concentrated attention. The interpreter should adjust his/her own speech tempo to that of the speaker. Several skills are simultaneously featured: listening, speaking, switching to another language, compressing information. Simultaneous interpreting is possible due to the human ability to anticipate and forecast what will be said in some minutes (вероятностное прогнозирование).²²⁴ Ч е р н о в Г. В. Основы синхронного перевода. - М.: Высшая школа, 1987. ⁴ To do it, one must have a good command of the subject matter under discussion. Since the simultaneous interpreter's work is so intense and the conditions are extreme, interpreters are usually changed at the microphone every 20 or 30 minutes.

Simultaneous translation may take place not only in the special booth. There is also whispered interpreting (or chuchotage) where the interpreter sits between the participants and whispers his/her translation to them. This type of translation is often used in a business meeting.

The simultaneous interpreter can get the source text in written form, which does not make his/her job easier, since the interpreter has to do simultaneously three jobs: read, listen and interpret. It is a most strenuous task, for the interpreter has to be watchful of the speaker deviating from the text.

Written translation is also divided into sub-varieties. It may be a visual translation (a written text is before the translator's eyes), translation by ear (in this case the translator listens to the text and writes the translation: dictation-translation), sight translation, (i.e. translation of the written text without preliminary reading, usually done orally).

The most obvious differences between written translation and interpreting are as follows:

translators have time to polish their work, while interpreters have no time to refine their output

any supplementary knowledge, for example terminological or world knowledge, can be acquired during written translation but has to be acquired prior to interpreting

translators can re-read their texts, they do not have to memorize big segments, while interpreters are able to listen to the text but once

interpreters have to make decisions much faster than translators

unlike written translation, interpreting requires attention sharing and involves severe time constraints. Following the United Nations norms of six to eight pages of written translation per day, the professional translator typically produces about five words per minute or 300 words per hour. The simultaneous interpreter, in contrast, has to respond instantly at a rate of 150 words per minute or 9000 words per hour.²²⁵ R o u t l e d g e E n c y c l o p e d i a… - P.186.⁵

§ 4. FUNCTIONAL CLASSIFICATION

According to the dominating function of the source text, translations are divided into literary and informative groups.

In literary translation, the poetic function of the text prevails. It is the translation of fiction prose, drama, and poetry. To translate a literary work, a translator should apply for the copyright.

Informative translation is the translation of texts on science, technology, official writings, business messages, newspaper and magazine articles, etc. These texts can also have an expressive function, but it is not dominating in the text. The prevailing function here is informative.

Chapter 3. EVALUATIVE CLASSIFICATION OF TRANSLATION

§ 1. ADEQUATE AND EQUIVALENT TRANSLATION

Translation theorists have long disputed the interrelation of the two terms.²²⁶ В а н н и к о в Ю. В. Проблемы адекватности перевода. Типы адекватности, втды перевода и переводческой деятельности. // Текст и перевод / В.Н.Комиссаров, Л.А. Черняховская, Л.К. Латышев и др. - М.: Наука, 1988. - С.34-39.⁶

V. Komissarov considers them to denote non-identical but closely related notions. He claims that adequate translation is broader in meaning than equivalent translation. Adequate translation is good translation, as it provides communication in full. Equivalent translation is the translation providing the semantic identity of the target and source texts.²²⁷ К о м и с с а р о в В. Н. Теория перевода. - М.: Высшая школа, 1990.⁷ Two texts may be equivalent in meaning but not adequate, for example:

Никита грозил: «Покажу тебе кузькину мать.» - Nikita threatened , “I'll put the fear of God into you!” The Russian sentence is low colloquial, whereas the English one, though it describes a similar situation, has another stylistic overtone, a rather pious one.

A. Shveitser refers the two terms to two aspects of translation: translation as result and translation as process. We can speak of equivalent translation when we characterize the end-point (result) of translation, as we compare whether the translated text corresponds to the source text. Adequacy characterizes the process of translation. The translator aims at choosing the dominant text function, decides what s/he can sacrifice.²²⁸ Ш в е й ц е р А. Д. Теория перевода: Статус, аспекты, проблемы. - М.: Наука, 1988.⁸ Thus, adequate translation is the translation corresponding to the communicative situation. For example, Здравствуйте, я ваша тетя! can be inadequate to Hello, I'm your aunt!, when the Russian sentence is used not in its phatic (i.e. contact supporting) function but in the expressive function (as an interjection) to express the speaker's amazement.

Close to this understanding of translation adequacy is E. Nida's concept of dynamic equivalence, “aimed at complete naturalness of expression” and trying “to relate the receptor to modes of behavior relevant within the context of his own culture.”²²⁹ N i d a E. Towards a Science of Translating. - Leiden: Brill, 1964.⁹ Nida's principle of dynamic equivalence is widely referred to as the principle of similar or equivalent response or effect.³³⁰ N e w m a r k P. Approaches …- P.10.⁰

Y. Retsker states that the notion of adequate translation comprises that of equivalent³³¹ Р е ц к е р Я. И. Теория перевода и переводческая практика. - М.: Междунар. отнош., 1974.¹. According to him, an adequate target text describes the same reality as does the source text and at the same time it produces the same effect upon the receptor. Translation adequacy is achieved by three types of regular correlations:

equivalents, that is regular translation forms not depending upon the context (they include geographical names, proper names, terms): the Pacific Ocean - Тихий океан, Chiang Kai-shek - Чан Кайши, hydrogen - водород.

analogs, or variable, contextual correspondence, when the target language possesses several words to express the same meaning of the source language word: soldier - солдат, рядовой, военнослужащий, военный.

transformations, or adequate substitutions: She cooks a hot meal in the evening. - На ужин она всегда готовит горячее.

§ 2. LITERAL TRANSLATION

Literal translation is the translation that reproduces communicatively irrelevant elements of the source text, This usually happens when the translator copies the source language form on this or that level of the language.

According to the language level, there exist various types of literal translation:

on the sound level: this type of literal translation results in the so called “translator's false friends”, that is words similar in sounds but different in meaning: conductor - not кондуктор, but дирижер; herb - not герб, but лекарственная трава; computer silicon chips - not компьютерные силиконовые чипсы, but кремниевые чипы компьютера.

on the syntactic level: copying the structure of the source language. Sometimes an inexperienced translator is hypnotized by the source language, and, to translate “accurately”, he tries to render the meaning word for word, thus breaking combination rules of his/her own language. As an example, We often heard his name mentioned. - *Мы часто слышали его имя упомянутым.

on the semantic level: giving the primary meaning of the word or its part, whereas a semantic transformation is required: But outside it kept on raining. - *Но снаружи шел дождь, which is incorrect. Or подполковник - *subcolonel, the word not existing in English.

etymological errors: disregarding language changes. Words acquire new meanings over time and use: There, there, don't cry. - *Там, там, не плачь.

following the style of the source text: different registers require different language means. Thus, to use the example by V. Komissarov³³² К о м и с с а р о в В. Н. Теория ...², to a Russian, who got accustomed to brief and abrupt structures in the weather forecast, an English weatherman's sentence can sound like a poem line: Mist covered a calm sea in the Strait of Dover last night. - Туман покрывал спокойное море в Па-де-Кале прошлой ночью. Therefore, to produce the same impact upon the receptor as does the original, the translator has to partition the English sentence and make it more adaptable to a Russian: Прошлой ночью в проливе Па-де-Кале стоял туман. Море было спокойно.³³³ К о м и с с а р о в В. Н., Р е ц к е р Я. И., Т а р х о в В. И. Пособие по переводу с английского языка на русский. - М.: Высшая школа, 1965. - С.45.³

We can see that very often literal translation is not necessarily a word-for-word translation, although it is often associated with a rather negative evaluation of the translation.

Literal translation is sometimes referred to as formal, or grammar translation, though it is not the same.

However, sometimes literal translation on this or that level is a must. The translator cannot do without it when rendering proper and geographical names (Khabarov, Nakhodka); some borrowings (Red Guards - хунвэйбины is a literal translation (on a semantic level), into English of the Chinese hong (Red) wei bing (Guard), while the Russian word is a literal reproduction of the Chinese word on a sound level.

In some works, literal translation is called `faithful' translation - this term does not necessarily imply the negative connotation of slavish literalism.

§ 3. FREE TRANSLATION

Free translation is the reproduction of the source form and content in a loose way. This concept means adding extra elements of information or losing some essential ones.

Of course, it is not very accomplished of a translator to add details not described by the author, as was often done by a well-known (sometimes notorious) Russian translator I. Vvedenski. Neither is it proficient to contract the source text like A. Houdar de la Motte who reduced the twenty-four books of the Iliad to twelve in his translation, leaving out all the “anatomical details of wounds” and some other information.³³⁴ T r a n s l a t i o n / H i s t o r y / C u l t u r e / Ed. By A.Lefevere. - London & New York: Routledge, 1992. - P.26.⁴ Scholars of translation usually take a negative view of this type of free translation, known as adaptation in history of translation.³³⁵ R o u t l e d g e E n c y c l o p e d i a … - P.5-8.⁵

Nevertheless, free translation is appropriate in some cases: poetry translations are done with a certain degree of freedom. A translator is also free to modernize a classic text in order to subvert established target-language reader-response. Free translation is also admitted in the titles of novels, movies, etc. For instance, the outstanding Russian novel by Ilf and Petrov «Двенадцать стульев» is known in the United States as “Diamonds to Sit On”, which is accounted for by the bookselling advertising policies. The British movie “Square Peg” was translated into Russian as «Мистер Питкин в тылу врага», since the film translators did not find the adequate Russian idiom to convey the meaning “a person unsuitable for the place in which he works or lives” expressed by the English phrase “a square peg in a round hole”.

Recently translation theorists have begun to relate free translation to communicative translation, depending on the purpose of the translation, and literal translation to the so-called semantic translation. Communicative translation tends to undertranslate, i.e. to use more generic, catch-all terms in difficult passages. A semantic translation tends to overtranslate, i.e. to be more detailed, more direct, and more awkward.³³⁶ N e w m a r k P. A p p r o a c h e s …- P.39.⁶ P. Newmark, however, distinguishes semantic translation - as the attempt to render as closely as possible the semantic and syntactic structures of the target language, from literal translation, when the primary senses of the lexical words of the original are translated as though out of context. He defines communicative translation as that which produces on its receptors an effect similar to that on the receptors of the original.³³⁷ Ibid. - P.39.⁷⁶