1 2 3 ... 12 13

1. Translation definition

Translation means both a process and a result. In order to explain translation we need to compare the original (source) text and the resulting (target) one.

The formation of the source and target texts is governed by the rules characteristic of the source and target languages. Hence the systems of the two languages are included in our sphere of interest. These systems consist of grammar units and rules, morphological and word-building elements and rules, stylistic variations, and lexical distribution patterns (lexico-semantic paradigms). Language itself is a formal model of thinking, i.e. of mental concepts we use when thinking.

In translation we deal with two languages (two codes) and to verify the information they give us about the extralinguistic objects (and concepts) we should consider extralinguistic situation, and background information.

As an object of linguistic study translation is a complex entity consisting of the following interrelated components:

a) elements and structures of the source text;

b) elements and structures of the target language;

c) transformation rules to transform the elements and structures of the source text into those of the target text;

d) systems of the languages involved in translation;

e) conceptual content and organization of the source text;

f) conceptual content and organization of the target text;

g) interrelation of the conceptual contents of the source and target texts.

In short, translation is functional interaction of languages and to study this process we should study both the interacting elements and the rules of interaction.

Among interacting elements we must distinguish between the observable and those deducible from the observables. The observable elements in translation are parts of words, words, and word combinations of the source text.

However, translation process involves parts of words, words, and word combinations of the target language (not of the target text, because when we start translating or, to be more exact, when we begin to build a model of future translation, the target text is yet to be generated). These translation components are deducible from observable elements of the source text.

In other words, one may draw the following conclusion:

During translation one intuitively fulfills the following operations:

a. deduces the target language elements and rules of equivalent selection and substitution on the basis of observed source text elements;

b. builds a model consisting of the target language elements selected for substitution;

c. verifies the model of the target text against context, situation and background information;

d. generates the target text on the basis of the verified model.

Thus, the process of translation may be represented as consisting of three stages:

1) analysis of the source text, situation and background information,

2) synthesis of the translation model, and

3) verification of the model against the source and target context (semantic, grammatical, stylistic), situation, and background information resulting in the generation of the final target text.

Let us illustrate this process using a simple assumption that you receive for translation one sentence at a time (by the way this assumption is a reality of consecutive translation).

For example, if you received:

At the first stage the chips are put on the conveyer” as the source sentence. Unless you observe or know the situation your model of the target text will be: “На першому етапі стружку (щебінку) (смажену картоплю) (нарізану сиру картоплю) (чіпси) кладуть на конвеєр”.

Having verified this model against the context provided in the next sentence (verification against semantic context):

Then they are transferred to the frying oven you will obtain: “На першому етапі нарізану сиру картоплю кладуть на конвеєр”.

It looks easy and self-evident, but it is important, indeed, for understanding the way translation is done. In the case we have just discussed the translation model is verified against the relevance of the concepts corresponding to the word chips in all its meanings to the concept of the word frying (Is it usually fried? Or Is it worth frying?).

Verification against semantic and grammatical context is performed either simultaneously (if the grammatical and semantic references are available within a syntagma) or the verification against semantic context is delayed until the availability of a relevant semantic reference which may be available in one of the following rather than in one and the same sentence. Cases when the grammatical, semantic or situational references are delayed or missing present serious problems for translation.

The examples of specifying contexts are given in Table below.

Long stick – long run

grammatical and semantic context in one syntagma

The results are shown in the table

Put this book on the table

grammatical and semantic context in one sentence

The tanks were positioned in specially built shelters and the tank operating proved successful. The enemy could not detect them from the air.

Semantic context in different sentences

With these examples we want to stress a very important fact for translation: the co-occurring words of the words situated close to each other in a source text have invisible pointers indicating various kinds of grammatical, semantic and stylistic information. This information is stored in human memory, and the principal task of a translator is to visualize all of this information.

In the examples with chips we used so called deduction modeling, that is, we built our translation on the basis of our knowledge about the languages involved in translation and the knowledge of “the way things are in life” (e.g. that it is hardly reasonable to fry fried potatoes or fragmented stones). We intuitively formulated hypotheses about translation of certain words and phrases and then verified them.

So, speaking very generally, when we translate the first thing we do is analyzing the source text trying to extract from it all available information necessary for generating the target text (build the intermediate model of the target text), then verify this information against situation and background knowledge and generate the target text.

For example, let the source text be:

Europe’s leaders trust that these criticisms will pale into insignificance when the full import of expansion begins to grip the public mind.

Then, omitting the grammatical context which seems evident (though, of course, we have already analyzed it intuitively) we may suggest the following intermediate model of the target text that takes into account only semantic ambiguities:

Європейські лідери/лідери європейської інтеграції/ вважають/вірять/, що ця критика вщухне/поступово зійде нанівець/, коли важливість поширення (Євросоюзу) почне завойовувати громадську думку/ коли суспільство почне краще усвідомлювати важливість розширення Євросоюзу/.

On the basis of this model we may already suggest a final target text alternative:

Лідери європейської інтеграції вважають, що ця критика поступово зійде нанівець, коли суспільство почне краще усвідомлювати важливість розширення Євросоюзу. [It goes without saying that this target text alternative is not the only one – many other alternatives are possible].

It is important to bear in mind that in human translation (unlike automatic) the intermediate representation of the target text will comprise on the conscious level only the most problematic variations of translation which one cannot resolve immediately.

We seldom notice this mental work of ours but always do it when translating. However, the way we do it is very much dependent on general approach, i.e., on translation theories which are our next subject.

2. Basic translation theories

Here we shall discuss the most common theoretical approaches to human translation paying special attention to their limitations and ability to explain the translation process.

Roughly, the human translation theories may be divided into three main groups which quite conventionally may be called transformational approach, denotative approach, and communicational approach.

2.1. transformational approach

The transformational theories consist of many varieties which may have different names but they all have one common feature: the process of translation is regarded as transformation.

According to the transformational approach translation is viewed as the transformation of objects and structures of the source language into those of the target.

Within the group of theories which we include in the transformational approach a dividing line is sometimes drawn between transformations and equivalences.

According to this interpretation a transformation starts at the syntactic level when there is a change, i.e. when we alter, say, the word order during translation. Substitutions at other levels are regarded as equivalences, for instance, when we substitute words of the target language for those of the source, this is considered as equivalence.

In the transformational approach we shall distinguish three levels of substitutions: morphological equivalences, lexical equivalences, and syntactic equivalences and/or transformations.

In the process of translation:

  • at the morphological level morphemes (both word-building and word-changing) of the source language are substituted for those of the target;

  • at lexical level words and word combinations of the source language are substituted for those of the target;

  • at the syntactic level syntactic structures of the source language are substituted for those of the target.

For example, in the process of translation, the English word room is transformed into Ukrainian words кімната or простір.

The syntactic transformations in translation comprise a broad range of structural changes in the target text, starting from the reversal of the word order in a sentence and finishing with division of the source sentence into two and more target ones.

The most common example of structural equivalences at the syntactic level is that of some Verb Tense patterns. Real translation transformations are complex and often at different levels of languages. This kind of transformation is especially frequent when translation involves an analytical and a synthetic language, e.g. English and Ukrainian.

Thus, according to the transformational approach translation is a set of multi-level replacements of a text in one language by a text in another, governed by specific transformation rules.

2.2. denotative approach

The transformational approach is insufficient when the original text corresponds to one invisible concept which is rendered by the translator as a text in another language also corresponding to the relevant invisible concept.

For instance, the translation of almost any piece of poetry cannot be explained by simple substitution of source language words and word combinations for those of target language.

This type of translation is characteristic of any text, written or spoken, rather than only for poetry or high-style prose and the denotative approach is an attempt to explain such translation cases.

Though denotative approach to translation is based on the idea of denotatum (see above the relationship of signs, concepts and denotata), it has more relevance to that of a concept.

According to denotative approach the process of translation is not just mere substitution but consists of the following mental operations:

  • translator reads (hears) a message in the source language;

  • translator finds a denotatum and concept that correspond to this message;

  • translator formulates a message in the target language relevant to the above denotatum and concept.

According to this approach during translation we deal with similar word forms of the matching languages and concepts deduced from these forms, however, as opposed to the transformational approach, the relationship between the source and target word forms is occasional rather than regular.

To illustrate this difference let us consider the following two examples:

(1) The sea is warm tonight – Сьогодні ввечері море тепле.

(2) Staff only – Службове приміщення.

In the first instance the equivalences are regular and the concept, pertaining to the whole sentence may be divided into those relating to its individual components (words and word combinations): sea – море, tonight – сьогодні ввечері is warm – тепле.

In the second instance, however, equivalence between the original sentence and its translation is occasional (i.e. worth only for this case) and the concept, pertaining to the whole sentence cannot be divided into individual components.

2.3. communicational approach

The communicational theory of translation was suggested by O.Kade and is based on the notions of communication and thesaurus. So, it is worthwhile to define the principal terms first.

Communication may be defined as an act of sending and receiving some information, which is called a message.

It should go without saying that this definition is oversimplified and not all communication terms used here are standard terms of communication and information theories. Our purpose, however, is to describe the act of communication in the simplest possible terms and to show translation as part of this act.

Information, which is sent and receives (communicated) may be of any kind (e.g. gestures, say, thumbs up), but we shall limit ourselves to verbal communication only, i.e. when we send and receive information in the form of a written or spoken text.

Naturally enough when communicating we inform others about something we know. That is in order to formulate a message, we use our system of interrelated data, which is called a thesaurus.

We shall distinguish between two kinds of thesauruses in verbal communication: language thesaurus and subject thesaurus.

Language thesaurus is a system of our knowledge about the language which we use to formulate a message, whereas subject thesaurus is a system of our knowledge about the content of the message.

Thus, in order to communicate, the message sender formulates the mental content of his or her message using subject thesaurus, encodes it using the verbal forms of language thesaurus, and conveys it to the message recipient, who decodes the message also using language thesaurus and interprets the message using subject thesaurus as well. This is a simple description of monolingual communication.

It is very important to understand that the thesauruses of message sender and recipient may be different to a greater or lesser degree, and that is why we sometimes do not understand each other even when we think we are speaking one and the same language.

So, in regular communication there are two actors, sender and recipient, and each of them uses two thesauruses.

In special bilingual communication (i.e. translation), we have three actors: sender, recipient, and intermediary (translator). The translator has two language thesauruses (source and target one) and perform two functions: decodes the source message and encodes the target one to be received by the recipient.

O.Kade’s communicational theory of translation describes the process of translation as an act of special bilingual communication in which the translator acts as a special communication intermediary, making it possible to understand a message sent in a different language.

One may note that the communicational approach pays special attention to the aspects of translation relating to the act of communication, whereas the translation process as such remains unspecified, and one may only presume that it proceeds, either by a transformational or denotative path.

However, it is difficult to overestimate the importance of the communicational aspect in the success of translation.

To understand this better let us consider an example of message formulation (encoding), message translation (encoding/decoding), and message receipt (decoding).

Let the original message expressed by a native speaker of English (encoded using the English language as a code to convey the mental content of the message) be:

Several new schools appeared in the area.

Let us assume then that the message sender, being a fisherman and using relevant subject thesaurus, by schools meant large number of fish swimming together rather than institutions for educating children, and the correct translation then had to be:

У районі з’явились нові косяки риби

whereas the translator who presumably did not have relevant information in his subject thesaurus translated schools as institutions for educating children:

У районі з’явились нові школи,

which naturally lead to misunderstanding (miscommunication).

The above example shows a case of miscommunication based on the insufficiency of extralinguistic information. However, there are also cases of miscommunication caused by the insufficiency of linguistic information. The example clearly illustrates a dividing line between linguistic and extralinguistic information in translation as visualized by the communicational approach to translation.

Thus, the communicational approach to translation, though saying little about translation as such, highlights a very important aspect of translation.

According to communicational approach translation is a message sent by a translator to a particular user and the adequacy of translation depends on similarity of their background information rather than only on linguistic correctness.

3. Translation ranking

Even in routine translation practice one can see that there are different ranks of translation, that one rank of translation consists of rather simple substitutions whereas another involves relatively sophisticated and not just purely linguistic analysis.

Several attempts have been made to develop a translation theory based on different translation ranks or levels as they are sometimes called. Among those one of the most popular in the former Soviet Union was the “theory of translation equivalence level (TEL)” developed by V.Komissarov.

According to this theory the translation process fluctuates passing from formal inter-language transformations to the domain of conceptual interrelations.

V.Komissarov’s approach seems to be a realistic interpretation of the translation process; however, this approach fails to demonstrate when and why one translation equivalence level becomes no longer appropriate and why, to get a correct translation, you have to pass to a higher TEL.

Y.Retsker maintains that any two languages are related by regular correspondences (words, word-building patterns, syntactical structures) and “irregular ones”. The irregular correspondences cannot be formally represented and only the translator’s knowledge and intuition can help to find the matching formal expression in the target language for a concept expressed in the source language.

According to J.Firth, in order to bridge languages in the process of translation, one must use the whole complex of linguistic and extralinguistic information rather than limit oneself to purely linguistic objects and structures.

J.Catford, similar to V.Komissarov and J.Firth, interprets translation as a multi-level process. According to Catford a certain set of translation tools characteristic of a certain level constitutes a rank of translation and a translation performed using that or another set of tools is called rank bound.

All these theories try to explain the process of translation to a degree of precision required for practical application, but no explanation is complete so far.

The transformational approach quite convincingly suggests that in any language there are certain regular syntactic, morphological, and word-building structures which may be successfully matched with their analogies in another language during translation.

The transformational approach forms the basis of machine translation design – almost any machine translation system uses the principle of matching forms of the languages involved in translation. The difference is only in the forms that are matched and the rules of matching.

The denotative approach treats different languages as closed systems with specific relationships between formal and conceptual aspects; hence in the process of translation links between the forms of different languages are established via conceptual equivalence.

The communicational approach highlights a very important aspect of translation – the matching of thesauruses. Translation may achieve its ultimate target of rendering a piece of information only if the translator knows the users’ language and the subject matter of the translation well enough (i.e. if the translator’s language and subject thesauruses are sufficiently complete). This may self-evident, but should always be kept in mind, because all translation mistakes result from the insufficiencies in the thesauruses.

Moreover, wholly complete thesauruses are the ideal case and it is still virtually impossible to know everything about any possible subject matter related to the translation.

Different approaches differ only in the accents placed on this or that component but all theories discussed recognize the following three basic components of translation:

Meaning of a word or word combination in the source language (concept or concepts corresponding to this word or word combination in the minds of the source language speakers).

Equivalence of this meaning expressed in a word or word combination of the target language (concept or concepts corresponding to this word or word combination in the minds of the target language speakers).

Extralinguistic information pertaining to the original meaning and/or its conceptual equivalent after the translation.

So, to put it differently, what you can do in translation is either match individual words and combinations of the two languages directly (transformational approach), or understand the content of the source message and render it using the formal means of the target language (denotative approach) with due regard of the translation recipient and background information (communicational approach).

The hierarchy if these methods may be different depending on the type of translation. Approach priorities depending on the type of translation are given in Table below.

Translation Type

Translation Method Priorities

Oral Consecutive

Denotative, Communicational

Oral Simultaneous

Transformational, Communicational

Written (general & technical)


Written (fiction & poetry)


Thus, in oral translation priority is given to denotative method, because a translator is first listening to the speaker and only after some time formulates the translation, which is very seldom a structural copy of the source speech.

In simultaneous translation as opposed to consecutive priority is given to direct transformations since a simultaneous interpreter simply has no time for conceptual analysis.

In written translation, when you seem to have time for everything, priority is also given to simple transformations (perhaps, with exception of poetic translation). This is no contradiction, just the path of least resistance in action – it is not worthwhile to resort to complex methods unless simple ones fail.

It should be born in mind, however, that in any translation we observe a combination of different methods.

4. Translation equivalence and equivalents

Translation equivalence is the key idea of translation. According to A.S.Hornby equivalent means equal in value, amount, volume, etc. The principle of equivalence is based on the mathematical law of transitivity. As applied to translation, equivalence means that if a word or word combination of one language (A) corresponds to certain concept (C) and a word or word combination of another language (B) corresponds to the same concept (C) these words or word combinations are considered equivalent (connected by the equivalence relation).

In other words, in translation equivalent means indirectly equal, that is equal by the similarity of meanings. For example, words table and стіл are equivalent through the similarity of the meanings of the Ukrainian word стіл and one! Of the meanings of the English word table. In general sense and in general case words table and стіл are not equal or equivalent – they are equivalent only under specific translation conditions.

This simple idea is very important for the understanding of translation: the words that you find in a dictionary as translations of the given foreign language words are not the universal substitutes of this word in your language. These translations (equivalents) are worth for specific cases which are yet to be determined by the translator.

As we know, the relation between a language sign (word or word combination) and the fragment of the real world it denotes is indirect and intermediated by the mental concept. We also remember that the mental concept of a given language sign is usually rather broad and complex, consisting of a lexical meaning or meanings, a grammatical meaning or meanings, connotations and associations. It is also worth reminding that the mental concept of a word (and word combination) is almost never precisely outlined and may be defined even in the minds of different speakers of the same language, not to mention the speakers of different languages.

All this naturally speaks for the complexity of finding the proper and only translation equivalent of the given word. Translation equivalence never means the sameness of the meaning for the signs of different languages.

Translation equivalents in a dictionary are just the prompts for the translator. One may find a proper equivalent only in speech due to the context, situation and background knowledge.

The idea of translation equivalence is strongly related to that of the unit of translation, i.e. the text length required to obtain proper equivalent.

It is generally known that one word is hardly a common unit of translation, especially in analytical languages with usually polysemantic words. Their meaning strongly depends on the environment. One is more likely to find a universal equivalent for a word combination, in particular for a clichéd one (e.g. hands up, ready made), because a word combination is already a small context and the clichéd expressions are commonly used in similar situations. The general rule of translation reads: the longer is the source text, the bigger is a chance to find proper and correct translation equivalent.

Traditionally and from practical viewpoint the optimal length of text for translation is a sentence. Being a self-sustained syntactic entity a sentence usually contains enough syntactic and semantic information for translation. However, there are cases when a broader stretch of the source text is required. It supplies additional information necessary for translation.

Thus, put with certain degree of simplification, equivalence is a similarity of meaning observed in the units of different languages and used for translation. The units of the target language with meanings similar to the relevant units of the source language are called translation equivalents. Modern translation theory suggests two basic grades of translation equivalents.

a) Full Translation Equivalents

As it was previously mentioned, one can hardly find truly full and universal equivalents for a word. However, practical translation dates back to ancient times and since then translations are commonly regarded and used as full-pledged substitutes of the relevant source text. That is why despite contradicting theoretical evidence full equivalence is commonly accepted as a convenient makeshift.

For practical purpose full equivalence is presumed when there is complete coincidence of pragmatic meanings of the source and target language units.

This rule applies both to individual words and their regular combinations. Speaking generally, translation equivalents of all words and word combinations one finds in a good dictionary are full because the translation practice reflected in dictionaries shows them as complete substitutes universally accepted by the speakers’ community of the target language (i.e. pragmatically equivalent).

Of them the stylistically neutral words with reference meanings (terms, geographical and proper names, words denoting physical objects and processes) are more likely to have full translation equivalents because semantic and pragmatic parts of their meaning are less ambiguous.

b) Partial Translation Equivalents

To understand the partiality and the completeness of translation equivalence let us consider the syntactic, semantic and pragmatic aspects of equivalence, because the partiality of equivalence is, as a matter of fact, the absence of one or more of these aspects.

Let us start from examples. Книга as an equivalent of the English word book is full in all equivalence aspects because it has similar syntactic functions (those of a Noun), its lexical meaning is also generally similar, and the pragmatic aspect of this equivalent (the message intent and target audience reaction) coincides with that of the English word. Thus, книга is conventionally regarded as a full equivalent of the word book.

Strictly saying, however, the Ukrainian word протестувати, for example, is a partial equivalent of the English word protesting (say, in the sentence Protesting is a risk – Протестувати ризиковано) because of different grammatical meanings (a Gerund and a Verb), the semantic and pragmatic aspects being similar.

To take another example of partial equivalence let us consider the English saying Carry coal to Newcastle. If one translates it as Возити вугілля до Ньюкасла it would lack the pragmatic aspect of equivalence (The intent of this message Bring something that is readily available locally would be lost, because the Ukrainian audience could be unaware of the fact that Newcastle is the center of a coal-mining area). If, however one translates it Їхати до Тули з власним самоваром it would lose the semantic similarity, but preserve the pragmatic intent of the message, which, in our opinion, is the first priority of translation. Anyway, both suggested translation equivalents of this saying are considered partial.

<< предыдущая страница   следующая страница >>