SYSTEMIC APPROACH TO TRANSLATION: EXPERIMENTAL RESEARCH

The article deals with the problem of translation as a system and analysis of its systemic features in both structuralist and processoriented paradigms. In its structuralist dimension, systemic nature of translation is revealed through that of a text (as the main object of translation analysis) and language (as a source of its means). In its process-oriented dimension, translation is revealed in translator’s systemic thinking. Researching the specifics of systemic thinking as a cognitive activity requires a new methodology. The article proposes the exploration into translator’s decision-making on the basis of experimental method of autointrospection, which is defined as exteriorisation of translator’s considerations as to different aspects of the activities he or she is being engaged in. For this purpose, the eccentric tale by S. Milligan was translated and notes were taken as to the specifics of overcoming selected translation difficulties of the lexical level (proper names, onomatopoeias, culturally-bound lexemes, nonce words, and puns). Special attention was given to formulating a general strategy of translation of the text in general and local strategies for reproduction of the selected lexical items. The research allowed to draw some conclusions as to: correlation between structuralist and process-oriented vision of translation as a system; ways and methods of correlating methods of systemic translating lexical items both within their categories and among these categories on the general textual level. Systemic thinking is thus presented as a skill whose development is the essential part of translators’ professional training.

It's also quite important to make difference between the systemic nature of language and that of translation which have both common and distinct features. Since translation implies operations with linguistic units of different hierarchy of the source and target languages, it assumes (at least, partially) systemic features of these two languages that seem to be necessary for interlinguistic cooperation. At the same time, translation possesses systemic features of its own. In my opinion, systemic specificity of translation is connected with translator's systemic thinking and characterises it as an activity rather than result (i.e. translated text). Systemic thinking allows to get beyond the borders of what seems like isolated and independent acts and to see the structures lying at their foundation. Due to this, we can identify the ties between and among these acts and to perfect our ability to understand and influence them. Thus emerges the necessity of a principally new research methodology reflecting the shift of Translation Studies into the cognitive sphere.

Method
Systemic nature of translation is to be placed among its ontological characteristics as it reveals itself in each translation act and is explained differently in every possible paradigm of Translation Studies. In the previous section of this article I characterised systemic specifics of translation in terms of structuralist paradigm where it is treated similarly to that of the language or text. Since the aim of my research is to find and expose peculiarities of translator's systemic thinking, descriptive methods (i.e. comparative analysis) become irrelevant which makes us look for some new ways of scientific exploration.
The best way to unlock the "black box" of translation and get inside the translator's mind is to apply one of the experimental methods that are currently gaining more and more popularity among scholars. Experiment can be broadly defined as an empirical method of cognition which helps obtain under controlled and guided conditions some new knowledge about cause-effect relations among the phenomena and objects and to determine new properties of these phenomena and objects. This definition proves that experimental methods ideally suit the purpose of exposing systemic nature of translation in its process-oriented dimension. If you want to find out how the system works and to perceive its emergent properties you should see it in action. Proceeding from this notion, I decided to turn to a rather simple but very efficient experimental method known as autointrospection.
What is autointrospection? The essence of this method lies in translator's attempt to exteriorise (verbalise) his or her considerations as to different aspect of the activities he or she is being engaged in. Introspection can be general (when no particular task is set) or pointed (when some particular task is set). General introspection processes are considered faster and more efficient than processes which are under conscious control, however, they are also less flexible and more difficult to modify at need (Bernardini, 1999, p. 180). In my case, I conducted the translation myself (hence "autointrospection") and, following the course of my research, tried to "pull out" all ideas and thoughts connected with systemic aspects of my work.
In many respects, autointrospection is close to the popular research method known as Think-Aloud-Protocol (TAP) which is usually defined as "an experiment where one subject talks to himself or herself while translating a text" (Kussmaul, Tirkkonen-Condit, 1995, p. 179). TAP also employs special equipment and /or computer software for recording translator's speaking and movements (like keystrokes, cursor movements, etc) (Rambaek, 2004). In contrast to this, I neither pronounced nor recorded my considerations, giving preference to note-taking.

Results and Discussion
I chose for the experiment a well-known eccentric tale by Spike Milligan "The Sad Happy Ending Story of the Bald Twit Lion" (Milligan, 1968). My considerations were as follows: 1) the tale hadn't been previously translated into Ukrainian which allowed me to avoid secondary influence; 2) the tale is maximally infused with all possible kinds of translation difficulties which provide a very interesting material for analysing both creative and systematic aspects of translator's thinking.
Since the translated text is aimed at children's audience, general (superordinate) strategy of moderate domestication was chosen. In my opinion, it helps overcome potential ambiguity of the original created by abundant cultural and ludic elements. Further, for each particular type of translation difficulties I developed a suitable local strategy subordinate to the general one adopted for the text as a whole. I also followed the general rule of translator's strategic thinking: there's one general strategy for the text and it is determined at the pre-translation stage, while there may be several different local strategies which are determined directly during translation process.
Proper Names. Most proper names in the story are of allusive character, that's why the local strategy for this type of elements was to reproduce their "hidden" allusive meanings. Within this strategy, I invented my own correspondences for the proper names coined by the author, while real names were either preserved (if they seemed familiar to the potential Ukrainian reader) or substituted for some functional equivalents with similar cultural colouring. The following example features the name of the protagonist and the name of its location: One day in the middle of the Jungle, near a village called Pongoland, a big lion called Mr Gronk had an attack of strongness.
Pongoland is a toponym coined by the way of combining two root morphemes Pongo (which is a colloquial form for "orangutan") and land. Since the context clearly indicates that the toponym is the name of a fictional village, I coined a similar lexeme following the rules and traditions of the Ukrainian toponymy -"Орангутангове". The next onym Mr Gronk was created through the method of anthroponymisation of a slang word gronk with the derogatory meaning "dumb person" or "idiot". Since I decided to avoid direct insult not so typical of the Ukrainian children's literature, my way out was to combine two root morphemes "дурень" ("fool") and "лев" ("lion") in a nonce anthroponym "Дуролев": Одного чудового дня в нетрях Джунглів поблизу селища Орангутангове зі здоровенним левом на ім'я Пан Дуролев стався напад всесильності.
Another example: Once, twice and thrice upon a time there lived a Jungle. <…> It was a really good Jungle: great scarlet lilies, yellow irises, thousands of grasses all grew very happily, and this Jungle was always on time. Some people are always late, like the late King George V. But not this Jungle.
Here, in addition to an anthroponym, we have some other cases of translation difficulties, namely, a modified phraseological unit (Once, twice and thrice upon a time from "Once upon a time") and a pun (late -1) occurring or arriving after the correct or expected time; 2) having died, esp. recently). This made me look for the translation variant that would allow coordinating all three elements and thus required a systematic approach.
By substituting potentially ambiguous onym King George V for colloquial Ukrainian "Царь Панько" I could concurrently solve the problem of modified phraseological expression at the beginning of the sentence by substituting it for the Ukrainian "За часів царя Панька".
Onomatopoeias. The text of the tale includes so many onomatopoetic words and expressions that they can justly be considered a token of S. Milligan's individual style. It also means that onomatopoeias carry out important stylistic and narrative functions that are to be reproduced in the translated text. Correspondingly, my local strategy of dealing with onomatopoeias was determined by the intention to preserve (reproduce) them by analogues means.
Take a look at the following example: He was twenty-one that day and had been given the key to the Jungle, so he put on a fierce look and then, leaping in the air, he gave the biggest, loudest roar in the world. "ROAR -ROAR -ROAR!! ROAR!!!" he went; in fact he roared so loud that it loosened all the roots of his hair and tinkle tinkle all his lovely mane fell off, and landed on the ground PLIP-PLAP-PLOP 200,000 times, one for every hair. <…> A Bald Lion? "Oh dearie me, I'll be the laughing stock of the hyenas," he said. So he un-roared, "RAOR! RAOR! RAOR!", but his hairs didn't go back in.
Culturally-bound words. Culturally-bound words known in the Ukrainian Translation Studies under the term "realia" are used by S. Milligan to make a bridge between fairy quasi-reality and modern life of Great Britain. This narrative device is also typical of the author's works in general and thus deserves my special attention translation-wise. Taking into account the general course towards domestication, I developed a local strategy according to which most culturally-bound items were substituted for those not necessarily associated with any particular culture or country (including Ukraine). Such a decision was partly taken in an attempt to help resolve the puns integrated by S. Milligan into culture-specific words. Consider the next example: Suddenly Mr Gronk saw himself in the Daily Mirror and, oh! he saw that he was now bald! The main difficulty in translating the name of the popular British newspaper "The Daily Mirror" is that it goes hand in hand with the play of words based on the direct meaning of the word "mirror". Thus, the character looked simultaneously at the newspaper and at its own reflection. At the same time, the symbol of the mirror as the name of a newspaper is so typical all over the world that I could easily find the Ukrainian equivalent denoting the real newspaper: У цей момент Пан Дуролев, Здоровенний Лев із Джунглів, побачив себе у "Дзеркалі Тижня" абсолютно лисим! Another example demonstrates how realia is employed for the play of words:

One night when he was having tea (Lyons) he said, "I can't go on being bald."
In this sentence the author plays with the name of a popular brand of English tea "Lyons" which is homophonic with the word "lion". Thus, speaking about "Lyons tea" S. Milligan simultaneously hints that it is also the tea of the protagonist (lion's tea). In this particular case, I had to resort to the compensation technique since I could not find among Ukrainian brands any connected with lions. Instead, I found a local coffee brand which served pretty well for achieving in translation the ludic effect similar to that of the original: Одного вечора, коли Лев пив каву "Золотий Лев", він сказав сам собі: "Я не можу більше залишатися лисим".
Nonce formations. I've already analysed the translation of some items belonging to this category of translation difficulties but only in connection with other categories. Local strategy of reproducing nonce words is determined by the necessity to preserve their functional, stylistic and pragmatic load which is only possible on condition of coining similar nonce words with the means of the target language. Both form and meaning are essential since losing either of these components will inevitably lead to dramatic decrease in the level of equivalency. Let's take the following example: His head got into the Top Ten Baldies; he out-balded Yul Brynner and Bing Crosby.
The nonce word to out-bald is once again "accompanied" with other units that fall into the focus of my interest: proper names with strong cultural connotations (Yul Brynner, Bing Crosby) and slang word (baldie). Taking into account stylistic and narrative ties among them, systemic principle should dominate the choice of equivalents for all these items.
For the nonce word, I coined a nonce correspondence on the basis of the same method of wordformation -prefixation. The meaning of Ukrainian prefix "пере-" perfectly matches that of the original one -"out-". In addition, the newly coined verb "перелисити" reminds the verb "пересилити" ("to overpower") which, in my opinion, creates additional ludic effect. Dealing with proper names, I had to take into consideration that they were chosen by S. Milligan for one particular purpose: to represent famous personalities with bald spots. Being confident that the original names are most likely not familiar to the potential Ukrainian readers of the tale, I once again followed the abovementioned strategy of substitution and proposed instead the names of Ukrainian folk heroes associated with the round form (like the bald head). Finally, dealing with derogatory slang word baldie, I opted for a descriptive translation because I could not find any Ukrainian slang words with the similar meaning: Невдовзі левова лисина потрапила до Десяти Найгламурніших Лисин Світу. Пан Дуролев навіть пере-лисив Колобка та Котигорошка! Another very interesting example deals with the nonce words whose ambiguous meanings are almost impossible to decipher in a more or less definite way: One monkey laughed so much he fell out of his tree and krupled his blutzon.
The nonce words krupled and blutzon fall under the category of lexical nonsense which is defined as "an integrated, non-conventional combination of conventional elements of phonemic and/or morphemic levels of a language system, whose ontological characteristics are: non-conventionality, referential-significational indeterminateness and semantic ambiguity" (Voronina, 2016, p. 167). These characteristics determine the impossibility of ascribing any conventional meaning to nonsensical lexemes; instead, their interpretation by any possible recipient will be unique and based not on morphological analysis but rather on personal associations. It's even more difficult for the translator to embody his or her associations into a coherent linguistic form (unless transcoding is chosen as a method of translation). That is why, I opted for a descriptive variant that matched the situational context pretty well: Одна поважна мавпа так реготала, що аж з дерева впала та хвоста зламала.
Puns. As I've already mentioned, word play is one of the most prominent individual features of S. Milligan's works turning them into intricate objects of linguistic play. Puns always present a real challenge for translators but, as M. Cronin put it, "If there is no play, the game of translation is likely to lose its appeal" (Cronin, 2005, p. 93). Bearing in mind this wise piece of advice, I developed a local strategy of puns' parallel reproduction, meaning that every case of the word play should be substituted for another case of a word play. Of course, the implementation of this principle is always limited due to the linguistic asymmetry and my translation was no exception. In my previous examples, I've already discussed the successful resolution of puns based on the words late, mirror and Lyons, so now it's turn of a not so successful solution: "Listen," he said. "I was once locked in a den of lions, and none of them bit me, and the audience asked for their money back, so it's my turn to do you all a good turn." So he did twenty good turns and became giddy.
In this example the pun is based on the expression "to do smb a good turn" with the meaning "to do smb a favour". S. Milligan factually employs the technique of desemantisation of a phraseological unit returning it its literal meaning: "to do a turn" -"to turn around". In translation the pun is lost. In order to partially compensate this loss I added in the first sentence "і як я не крутився" (literally, "no matter how hard I tried to turn around", thus creating a new pun based on two meanings of the Ukrainian verb "крутитися": 1) "робити оберти навколо себе" and 2) "намагатися": "Слухайте-но сюди, якось мене посадили у клітку з левами, і як я не крутився, але жоден з них мене так і не вкусив, через це глядачі зажадали назад свої гроші. Тож зараз я спробую покрутитися перед вами ще раз, може це вас трохи розважить». Святий Даниїло двадцять разів зробив повне сальто, аж голова пішла обертом The total number of different categories of lexical difficulties in the translated tale by S. Milligan was 40. The qualitative analysis of the experiment's results demonstrated that in 11 cases (28%) I resorted to the method of contextual substitution of the original unit which led to the loss of some formal parameters of the source element but (what is more important) allowed to preserve its functional load and pragmatic effect. At the same time, in 9 cases (22%) contextual substitution led to the loss of the original element and, consequently, its functional and pragmatic properties. The losses mainly occurred due to the restrictions that arise from asymmetry between source and target languages. Deviation from the declared local strategies once again proved that linguistic restrictions are the main barriers on the way of both systemic and creative thinking and corresponding decision-making in translation.
In 18 cases (45%) I managed to reproduce the original element preserving not only its functional and pragmatic but also formal (structural and semantic) parameters, thus providing the highest possible level of equivalence. In 2 cases (5%) literal translation resulted in the loss of the word play and nonce word. In total, in 73% of cases the application of systemic approach through coordinated local and general strategies proved successful in overcoming translation difficulties, while in only 27% of cases I was forced to apply translation methods leading to the loss of the original element with its formal, functional and pragmatic parameters. Such a correlation between successful and unsuccessful reproduction of translation difficulties not only testifies in favour of a systemic approach but also demonstrates how translator's creative potential is revealed as a result of translator's systemic thinking.
In a theoretical sense, the conducted experiment allowed me to trace how translator's aspiration to resolve systematically different types of translation difficulties resulted in the realisation of the following principles common for all complex systems including those of language, text and translation:  integrity which allows to view the system as simultaneously the whole and a subsystem of another system. In respect to translation, integrity may be revealed in different aspects. Translated text as a system, on the one hand, integrates different objects of linguistic and/or speech character, but, on the other hand, is subordinate to other higher systems such as genre, style, literature and culture. Translator's decisions should be of integrated character as well. Taking decisions on the basic level of linguistic units, the translator should make sure they are integrated into the text genre, literary and cultural traditions of a target society. For instance, dealing with a literary tale, I had to bear in mind potential limitations set by the Ukrainian tradition of children's literature;  structural/organisational hierarchy which presupposes the existence of a set of elements located according to the principle of subordination of lower level elements to higher level elements. The realisation of this principle is easily seen in the structural organisation of both language and textual systems. In respect to translation, it should be noted that equivalence on the lower level can be sacrificed for the sake of equivalence on the higher one;  structurisation which allows analysing systemic elements and ties between/among them within a particular structure because the process of system's functioning is, as a rule, determined not only by the properties of its separate components but rather by the properties of the structure itself. Here we deal with dynamic complexity which emerges when the elements of a system can form different relations between / among them. Since each element can exist in multiple states (like morphological or syntactic categories), they can be (inter)connected in a literally innumerable number of ways. This statement is of great importance for translation as it points at the necessity of multivector thinking in order to coordinate textual elements in their different states in many different aspects: phonologically, lexically, morphologically, syntactically, pragmatically, stylistically, logically, compositionally, etc.
Interconnectivity of system's elements gives rise to another important regularity of their behaviour. When you change a certain element (or its state) the consequences of your act can be compared to the circles on the water from the stone. In other words, the response to this act may potentially involve and influence other elements of the system; -multiplicity which allows employing multiple models and methods for the description of both separate elements and a system as a whole. Theoretically, multiplicity implies the use of different approaches, methods and paradigms for enlightening some particular aspects of translation as well as the notion of translation in general. Practically, multiplicity deals with the problem of decision-making which is always connected with choosing from a row of potential equivalents. Multiplicity thus lies at the point of intersection of systemic and creative aspects of translation.

Conclusion
The employment of experimental methods opens new ways for shedding light on those aspects of translation which so far have remained hidden from scholars' attention. The cognitive shift of Translation Studies makes it possible to explore the systemic nature of translation not only as a text but also as a process. Through the method of autointrospection, according to which I verbalised my considerations as to the systemic ways of resolving translation difficulties, I managed to collect some data concerning, firstly, the coordination of my decisions within local strategies as well as, secondly, coordination of local strategies within a general strategy. The prevalence of successful cases of resolving translation difficulties over failures in proportion 3 to 1 is a conclusive proof to the necessity of developing the skills of systemic thinking among professional translators.