Developing expertise

According to Sirén and Hakkarainen (2002: 75), another way of examining translation expertise is investigating the process of becoming an expert, and thus discovering why some people achieve this status while others do not. One way to look at this is to inspect how people resolve problems they encounter.

Bereiter and Scardamalia (1993: 98–100) introduce the terms progressive problem solving and problem reduction to help in this approach. The first term is usually linked with expertise and the latter the exact opposite of it. However, it should be mentioned that these terms are mainly applied to study expertise in relation to writing, which is an essential skill when translation is concerned, but this is not necessarily the case with other fields. According to Bereiter and Scardmalia (1993: 98–100, 156–157) the problem reduction approach is based on the idea that problems are something that need to be eliminated, which often ignores the complexities of the task at hand. In a translational context, such an approach usually results in a mediocre translation at best. However, the more progressive approach utilised by the experts does not ignore these complexities and the ones practising it do not insist on using only familiar routines unsuitable for the task. When facing a problem, they are working at the edge of their competence, and it is in this window, where their expertise develops. This type of working is undoubtedly taxing, but it often yields admirable results as well, since the new skills combine

13

with the old ones, thus resulting in better skills which can be utilised to perform even more complex tasks, since the edge of their competence has been pushed even further. If the problems are merely removed, the person will not have a chance to learn from them, and their learning process will not be as effective as it is for someone who has decided to solve the problem in a more progressive way. Furthermore, Bereiter and Scardamalia (1993: 156–157) add that in addition to a more effective learning process, the results are usually better as well, since the person has recognised the unique qualities of the problem and adjusted their process accordingly. The results are usually poorer if the person merely uses the knowledge he or she already possesses and forces the problem into that mould, creating a certain ‘close enough’

approach. However, as Bereiter and Scardamalia (1993: 100) mention, it is of great importance to realise that working at the edge of one’s competence all the time, in every situation, is virtually impossible, since it takes a lot of energy. Therefore, it is beneficial to learn when to use problem reduction and rely on routines, and when to invest in progressive problem solving.

Both problem reduction and progressive problem solving are strongly associated with learning.

Translation is a learned skill, and therefore, many of the skills related to it are teachable, such as searching for information and applying it accordingly, according to Jääskeläinen (1999: 242).

She explains that a difference between professionals and professionals is that non-professional and non-professional translators use different types of knowledge when they are working with a translation. Non-professionals tend to concentrate on linguistic matters, while professionals focus more on the cultural implications and the qualities of the text. In other words, they use a wider array of knowledge than their non-professional counterparts. However, with training, students learn to find and use different kinds of information, thus expanding their mental toolkit.

According to Whyatt (2017: 54–55, 61), turning the initial translation potential into a trained skill requires conscious effort from the person, including both formal translator trainees and people learning translation by themselves. In addition to treating translation as an educational learning instrument, external factors that support the development are required as well, such as a demand for translations. In addition, the skill can be developed further into translation competence with repeated practice and self-reflection. It is a slow process and can be acquired with assistance in different contexts such as institutionalised translator training or mentoring from a more experienced colleague.

14

Shreve (2006: 28) is also of the opinion that deliberate practice is essential in developing expertise. According to Shreve (2006: 29) deliberate practice can be defined as “regular engagement in specific activities directed at performance enhancement in a particular domain, where domain is some sort of skilled activity”. If this objective towards better performance, along with experience, is absent, the cognitive processes will not evolve towards expertise.

However, if the individual engages in deliberate practice, the cognitive changes will not only enable him or her to draw solutions from the patterns stored in his or her episodic memory but also develop more efficient knowledge structures. Shreve (2006: 37) lists several different changes in cognitive processes: greater episodic memory; pattern recognition in relation to the domain-specific problems and their solutions; continuing to learn new solution methods and applying them to classes of problems, which have been organised together in a more abstract level of thinking; and finally, storing the solutions in the long-term memory in an organised way for optimal retrieval. Shreve (2006: 29) also emphasises that translation expertise is an acquired skill that does not correlate with other cognitive abilities such as intelligence or memory, and that the most significant factor in its development is deliberate practice. However, merely translating a mass of text cannot be called deliberate practice, even if translation experience is vital for translation expertise.

According to Shreve (2006: 29–30), deliberate practice needs to meet several conditions. The first condition is that the practice should have a well-defined task, but due to the nature of translation, such a task is usually difficult to determine. However, there are smaller sub-task components that can be identified, such as text comprehension, translation strategies and target text production. Nevertheless, practicing these underlying basic tasks does not lead to expertise, and superior performance stems more from the experiential aspects, such as limiting oneself in regards of subject, type and the skopos of the translated text. Shreve (2006: 30, 38) states that a well-defined task in translation is therefore based on the assumption that the translator works on similar texts related to a certain domain over a long period of time. Therefore, in Shreve’s model, it is required that a translator also specialises in some other domain besides translation, and general translation expertise is not even achievable. If the translator who is an expert in one domain is told to work on some other domain, their superior performance will expectedly decline.

The second of Shreve’s (2006: 32) conditions is that the task is of reasonable difficulty but still challenges the translator. Shreve (2006: 32) defines the difficulty of a translation task as: ‘-- an

15

intersection of the specific characteristics of a given text with the existing (possibly deficient) cognitive resources of the translator.’ Therefore, the difficulty arises from the discrepancy between the skills of the translator and the aspects of the text, which the translator may not have the direct tools to solve. The difficulties may occur during different phases of the translation process. For example, in texts where the density of field specific vocabulary is high, and the sentence structures are complex, the translator is required to re-read the text multiple times and effectively utilise both their short- and long-term memory to be able to solve the translation problems. Shreve (2006: 31, 35) states that an expert translator can identify more potential translation problems and have more solutions to the problems than their non-expert counterparts.

He also suggests that a translator expert is able to consistently pick the best translation solutions most suitable for the situation, even when faced with a wide array of difficulties. Furthermore, Shreve adds that expert translation solutions are usually effective and do not require excessive amounts of effort, which results in greater speed and accuracy. The efficiency is a result of proceduralisation of the domain knowledge, which means that the translators recognise certain patterns in the text that they have seen before and know how to approach the situation. Shreve (2006: 32–33) suggests that a translator who wishes to improve their skill level should select texts which include challenging aspects regarding the skill in need of improvement, and then use those texts to deliberately practise. When translators are no longer a part of an educational environment, such as a university, it is likely that they will no longer receive constructive feedback on their translations, unless their workplace has deliberately opted to take feedback as a part of their workflow. Therefore, they need to be aware of the nature of the task and the translation themselves, which requires extensive utilisation of metacognitive abilities.

According to Shreve (2006: 38–39), the increase of expertise increases metacognition (also called self-regulation or self-monitoring, see Section 2.3) as well, which means that the translators are consciously aware of mental processes and how they work, in addition to controlling the said processes. Metacognitive regulation is one aspect of metacognition and it includes, for example, planning and monitoring of the working process.

The last of Shreve’s (2006: 34) conditions are repetition and error correction. The latter is usually a part of the revision process, but in the case of practice translations, there may not be possibilities to adjust the text, since the translator is not aware of the errors. In such cases, it would be best to develop a system which would allow error corrections, for example, with the assistance of a mentor. Since translation tasks are rarely identical, repetition of the task cannot be achieved in the same sense as in some different activities. However, there are usually certain

16

patterns, which will become familiar when the translator continues to translate texts in the same domain. According to Shreve (2006: 35), with time, the domain-specific problem-solving procedures will override the more general approaches, thus further accentuating the idea that expertise can only be achieved in quite a narrow field.

However, despite all the practice an expert translator has gone through, according to Gerloff (1990: 54), translation expertise does not always guarantee ease. In fact, it seems that if one achieves the status of an expert and has more experience, translating becomes harder. On one hand, some processes will undoubtedly become more automatised and therefore easier, but on the other hand, at the same time, this seems to allow more complex thought processes to take place instead. Therefore, it appears that there will always be problems to solve, whether one is an expert or a novice, but the problems the experts identify are more complex in nature and therefore take more time to solve. Sirén’s study (in Sirén and Hakkarainen 2002: 79–80) showed that the translation experts considered translation tasks harder than the non-experts did.

In the study, six participants translated two texts each: one literary text and another related to medicine. Two of the participants were translators specialised in medical translation, the other two in literary translation and the last two were medical professionals without background in translation. In the end, the translational experts took longer to translate the texts and did not consider their translations as finished work, unlike the non-experts who would not want to edit their texts apart from a few separate details if given the chance. The experts insisted that especially the literary text translation could not be done during only one sitting, and that they would need to know more about the context and the novel itself, which was not possible during the short time given to the research participants. It is of interest to note that one of the expert literary translators produced a very rough draft of the literary text, while one of the non-experts produced a text that was evaluated to be of good quality. However, since the expert considered their text a draft, in an authentic situation, the expert translator would have most likely worked on the text significantly more, which would have presumably yielded better results. The non-expert, however, considered the work already done and did not want to make any additional changes, and therefore the quality would remain the same.

Bereiter and Scardamalia (1993: 35) agree that experts in writing work harder than novices as well. According to them, in a think-aloud study, the word count of the experts’ thinking process was five or six times higher than the word count of their final essays. This results from the fact that they spend a lot of time evaluating and processing their work, so that their main point will

17

come across clearly. Contrastingly, non-experts appear to put their thoughts almost straight to the paper, deciding on their main points nearly immediately. Similar results were observed in Jääskeläinen’s (1999: 242) study, where the non-professional translators performed quickly and effortlessly, since they ignored the problems present in the text, whereas more experienced translation students spent a lot more time on the task, since their training had taught them to notice potential problems. However, the professionals performed either quickly or slowly depending on the context. In routine tasks, their automatised processes helped them to speed up their process, but in more complex tasks, they needed to exert more effort and utilise more time-consuming strategies, which made their process slower.