The Impact of the Source-Text Syntactic Characteristics on the Sight-Translation Strategies and Quality

Introduction. The aim of this first research in the Slovak-English language pair, is to establish the impact of the source-text (ST) syntactic structure on the target-text (TT) quality and its delivery in sight translation (SiT), as well as to identify the problems the interpreter students have in it, and the strategies they apply to solve them


Introduction
In this paper, we are going to focus on sight translation (SiT) in the meaning of the 'transposition of a message written in one language into a message delivered orally in another language' as opposed to sight interpretation, i. e. 'simultaneous interpretation with text' (Lambert, 2004: 294).Until recently, SiT, a hybrid form of translation and interpreting (Cao, 2020), or hybrid interpreting (Keníž, 1980), has mostly been viewed as a pedagogical exercise, for example, for testing consecutive interpreting (Sitnic, 2020) or developing oral language transfer skills through the source-text (ST) syntactical restructuring and paraphrasing (Ilg & Lambert 1996;Ho, 2022), and consequently, students have been rarely trained in this task (Agrifoglio, 2004: 43).
On the other hand, this type of language transfer is frequently used at meetings, where written documentation is delivered in the SiT mode.It is also often practiced at press conferences, in press reports, letters of apology for absence or congratulations (Čeňková, 2010;Čeňková, 2015;Gile, 2009), as well as in community interpreting (Li, 2014).
However, despite its importance in the field and in interpreter training, SiT is still often confused with sight interpretation, it has not attracted sufficient scholarly focus in the specialized literature and remains the poor cousin of other interpreting modalities (Čeňková, 2010) in terms of the attention devoted to it in research (Fang et al., 2022), in interpreter training and in practice (but see e.g., Weber, 1990;Martin, 1993;Moser-Mercer, 1995;Ilg & Lambert 1996;Agrifoglio, 2004;Shreve et al., 2010;Lee, 2012;Li, 2014;Putranti, 2017;Chmiel et al., 2019;Su et al., 2020;Lijewska et al., 2022).
Thus, taking into account the lack of sufficient knowledge on the SiT underlying mechanisms, the relevant research may provide valuable information for SiT theory and practice (Lee, 2012;Fang et al., 2022).It seems reasonable to agree that currently we have only a general picture of what SiT is and how it is done, and thus, more research is needed, especially, related to the comparisons across various language pairs (Ho, 2022).This work has already started with the investigations involving such pairs as German-English (McDonald & Carpenter 1981).Italian-English, Italian-French, Italian-German (Agrifoglio, 2004), Polish-English (Chmiel & Mazur, 2013;Chmiel et al., 2019;Chmiel et al., 2020;Lijewska et al., 2022), Kurdish-English (Fraidoon, 2021), Romanian-English (Sitnic, 2020), Indonesian-English (Putranti, 2017), Chinese-English (Cao, 2020;Su & Li 2020), Persian-English (Akbari, 2017) and others.However, the Slovak-English pair has not been researched so far.Thus, in this paper, which is the first one in the planned series of articles on the problem, we attempt to start filling in this gap.
Background One of the main problems in SiT is the permanent availability of the source text (ST), which provokes the interlanguage interference and literal translation (Putranti 2017), when the students have problems with deverbalization, i.e. they cannot free themselves from the impact of the ST structure (Akbari, 2017).The greater visual interference (Shreve et al., 2010) in SiT as compared to the auditory one in simultaneous interpreting, even in case of the high linguistic competence of the interpreters, has been proven experimentally (Chmiel et al., 2020).It is explained by the high cognitive demands on interpreters (Chmiel et al., 2020), specifically imposed by the limits on the reallocation of cognitive effort (Gile, 2009) during SiT, where comprehension problems may be more serious (Shreve et al., 2010).
It may seem that in SiT, the longer-term retention of the ST information is not a problem, since the ST continuously remains in front of the interpreters' eyes.In addition, the target-text (TT) delivery in SiT does not depend on the source-language (SL) speaker's delivery rate.It might lead to the conclusion about the relative ease of SiT compared to other modes of interpreting.However, research (Agrifoglio, 2004;Gile, 2009) has shown that SiT is a complex and unique technique, which is no less demanding than the other modalities of interpreting (consecutive and simultaneous).
First, SiT delivery speed may not be entirely at the discretion of the interpreter (Brady, 1989: 142).The SiT TT should sound as if the interpreter were reading a document in the target language (TL), which implies smooth delivery devoid of hesitations and pauses (Angelelli, 1999: 27).
Second, in the reception phase, the sight translator cannot rely on the speaker's vocal indications (intonation, pauses, emphasis, etc.), which contribute to the efficient ST segmentation (Gile, 2009: 180).In addition, in the said phase, unlike the consecutive interpreting, the sight translator cannot concentrate on the pure understanding of the ST meaning but has to plan its translation to provide for the TT smooth delivery.
Other difficulties involve linguistic dissimilarities between the two languages; the distinction between written and oral communication; the crossover between the two (Ho, 2022), as well as poor development of the mechanisms of anticipation and the target language (TL) equivalents recall (Putranti, 2017).The sources of difficulties in SiT are generally believed to be related to the ST comprehension, corresponding TL terms, required processing capacity for coping with syntactic discrepancies between the ST and TT, as well as the interference between the two languages (Gile, 2009: 215).Abstracting from terminological problems, the remaining three sources seem to be interrelated.
The permanent ST accessibility to the interpreter may provoke the SL interference, specifically related to its syntactic characteristics.According to the research (MacDonald, 1997;Gile, 2009), the complicated ST syntactic structure may negatively affect the interpreters' anticipation abilities and their processing capacity.Syntactic similarity facilitates anticipation and replication of the ST syntactic structures in the TT (Gile, 2009).However, if the order of information presentation in the two languages is different, it may result in a higher short-term memory load, because the ST information has to be retained before it can be reformulated in the TL (Gile, 2009: 198).It may not matter if the languages involved are syntactically similar or if the ST structure is favourable for segmentation (Gile, 2009: 180).Otherwise, it may be a problem.According to Viezzi (1989), information-processing efforts are inversely proportional to the extent to which morphosyntactic transformations are necessary.
The ST syntactic complexity is assumed to be a factor affecting the SiT efficiency.In one experiment (Shreve et al., 2010), manipulation of the ST syntactic complexity resulted in the predicted decrease of the SiT efficiency, indicating that SiT is more sensitive to disruption than written translation.In another experiment (Chmiel & Mazur, 2013), it was found that that the ST sentence readability could be a better predictor of processing load than syntax.However, the readability factor might be related to syntax as well, as the sophisticated syntax may simultaneously affect the sentence readability.In Lee's (2012) opinion, the greater the syntactic difference between the SL and TL, the greater the challenge to the interpreter to coordinate reading and smooth TL production.

The Study
The study aims to establish the impact of the ST (Slovak) syntactic structure on the TT (English) quality and its delivery, as well as to identify the problems the interpreter students encounter in SiT, and the strategies they apply to solve them.Given the lack of sufficient data concerning the mechanisms on which SiT is based, the results of such investigation could be useful both for building a psycholinguistic model of SiT and for improving its teaching methodology.We also suppose that the proper application of SiT exercises throughout the training process, may enhance trainees' awareness of the segmentation process which is crucial for the adequate interpreting output.
The study was based on the following hypothesis.The congruent ST syntactic structure will encourage surface-oriented strategies (literal translation), while the incongruent one will trigger overloading of the interpreter's processing capacity.Because the order of information presentation in the SL and TL is different, the interpreters will have to reformulate the ST fragments in the TT.The ST information will have to be retained in the short-term memory before it can be reformulated in the TL.This transformation will consume most of the available effort and weaken the control over the TT meaning and the compliance with the TL norms.It will negatively affect the quality of translation deteriorating its accuracy and fluency.

Participants
The participants (P) included six MA student-interpreters of Matej Bel University (UMB) in Banská Bystrica (Slovakia), 22-23 years of age (one male and five females), and one 38-year-old female student with MA degree in English and Engineering (participant 5).Their A language is Slovak and their B language -English.All participants (except P5) also studied the second foreign language (German or French).By the participants' own subjective assessment, their command of language A ranged between the levels of C1 (29%) and C2 (71%), while that of language B -between B2 (29%) and C1 (71%).The number of interpreting seminars (one seminar equals 26 contact hours) the students had had before the experiment started, ranged between 1 (P7, 14%), 3 (P1 and P4, 29%), 4 (P5, 14%), 5 (P6, 14%) and 6 (P2 and P3, 29%).The participants' simultaneous and consecutive interpreting experience was mainly limited to university classes, with the exception of P2, whose consecutive interpreting experience was limited to work, and P5, who interpreted simultaneously and consecutively both at her classes and work.None of the subjects had had any experience in sight translation where English was involved, apart from P5 who had had a specific 2,5-year experience of sight-translating a book in one 30-40-minute session a week.Participants 2 and 3 had had some sight-translating experience in their second foreign languages.However, the share of sight translation in the overall time of their interpreting training did not exceed 9% in the case of participant 2 and was in the range of 11-30% as far as participant 3 is concerned.The participants' profile is presented in Table 1.

Table 1
The Participants' Profile.P -participant, M -male, F -female, semsemester, LA -language A (native language -Slovak), LB -language B (foreign language -English), NIS -number of interpreting seminars, SIsimultaneous interpreting experience, CI -consecutive interpreting experience, SiT -sight translation experience, EN -English, OFL -other foreign languages, Gen -general (non-specialized)  As it follows the table 1, the participants may be regarded as a rather uniform group with a few individual variations as regards their communicative competence, general interpreting and sight-translation experience.The said variations may impose some limitations on the experimental data interpretation.Nevertheless, inter-subject variability is thought to be inevitable even among participants with similar backgrounds (Lamberger-Felber, 2003;Lee, 2012).Thus, this group of subjects may be suitable for the aims of our experiment.

The Source Text
The word order in English and Slovak is partially congruent (direct word order), but due to the developed system of inflectional affixes, Slovak has a greater freedom as far the positions of words in a clause are concerned.For example, in the sentence Súčasná trhová hodnota ponúka zainteresovaným stranám cenu (The current market value offers the interested parties a price), the lexeme hodnota (value) is the semantic subject, and cenu (price) -a semantic object.Depending on the communicative intention, the Slovak speaker may front the object, while shifting the subject to the final position (cenu zainteresovaným stranám ponúka súčasná trhová hodnota) without a change in the propositional meaning, i.e. cenu (price) remains the object of the sentence and hodnota (value) -its subject.But a similar transposition in English (a price offers interested parties the current market value) would result in a complete distortion of the ST meaning.If the clause structure in the two languages is congruent, the surface-oriented strategy (literal translation) would not affect the TT propositional correspondence to the ST.However, in case of incongruence, this strategy might result in a complete failure.
To study the impact of the ST structure on the translation strategy and the TT quality, we chose a non-specialized 250-word text in Slovak.Then we changed its structure in such a way that, while remaining within the SL norms, it differed as much as possible from the structure of the normative version of its English translation.Based on this modified ST structure, we also created a predicted version of each fragment's translation using a surface-oriented (literal) strategy to compare it with each student's output.
If the structure of a specific student's TT was close to the predicted one, it was considered as evidence of the surface-oriented strategy (SurfOS), while the ST structure transformation was viewed as an indication of a sense-oriented strategy (SensOS).It was expected that the analysis of the results would allow to formulate preliminary conclusions about the impact of the ST structure on the translation strategy and the TT quality.
The modified text for translation and the expected variants of its literal translation into English are presented below.Has also had an impact on their overpopulation the abundance of food.
Especially as a carrier of rabies, to humans can be dangerous the fox.

17A
Pri ochorení touto nebezpečnou nákazou stráca plachosť, In falling ill with this dangerous disease, it loses the shyness 17B ktorá je pre ňu typická, that is typical of it 17C a k ľuďom prichádza.and to humans comes.
does not mean it rabies has

The Delivery
The participants were asked to sight-translate the ST.Their delivery was recorded, transcribed and analyzed.The preservation of the ST word order in the TT was considered to be the indicator of the surface-oriented strategy (SurfOS), while switching over to the direct word order (subject -predicate) -the sign of the sense-oriented strategy (SensOS).The correlation of the SurfOS and SensOS in rendering 35 ST fragments is discussed further.The examples retain the participants' original formulations irrespective of their conformity to the standard norms of English.

Results and Discussion
The overall findings are presented in Table 3.As it follows from Table 3, the average group speed of delivery is 50 words per minute, which agrees well with the data of Lambert (2004) who found it to be about 60 words per minute for beginners.The overall quality of delivery is quite low (except participant 5) -the average number of major errors (those that distort the ST meaning) is over 21, the share of omissions -16%, and the amount of unmotivated long (over 2 seconds) pauses -29,7.
The average share of the surface-oriented strategy in the group (21%) practically coincides with the share of congruent ST fragments (favourable for the SurfOS) -20% (fragments 1A, 2A, 2B, 6B, 7A-B, 10A, 13A).However, the participants did not always apply SurfOS only in the above-mentioned fragments.The average percentage of SurfOS in the fragments favourable for it, is 57%, ranging among individual participants from 86% (P1) to 29% (P7), the latter figure looking atypical as all other students' results are close to half of all cases (43% in participant 2) or beyond it (57% and 71%).The average share of attempts to use SurfOS in the fragments unfavourable for it, is quite low (11%) with little variation among the students -P6 practically did not use it (3%), while P7 seems to use it at random, having the lowest SurfOS share in the fragments favourable for it and the highest one -in the unfavourable fragments (17%).
The specifics of students' behaviour in the course of translation are described below.
Obvious Surface-oriented Strategy in Congruent Cases.If there are no obvious indications that the ST structure is unsuitable for transferring it to the TT, the participants try copying it, taking the necessary steps to adapt it to the TL norms in the process of translation.
In fragment 1A, over half of the participants' actual output (57%) coincided with the prediction: P1 -So closer to people, even away from their natural habitat; P3 -Close to the people and from their environment; P4 -From their natural habitat (omission of Closer to people); P6 -Close to people and from their nature environment.In fragment 8, the TT structure matched the expectation in 29% of cases (P4 -preventing this migration of animals is harder and harder; P7 -Halting these movements of animals towards people is hard to do).
Most of the students (72%) maintained the final position of the lexeme foxes while translating fragment 6B: P1 -but huge problems are now also created by foxes; P2 -but problem represent also foxes; P3 -and there are big problem as well as foxes; P5 -but another huge problem are also foxes; P6 -but… also another big problem are foxes.This tendency is also evident in rendering fragment 5, where 29% of participants strived to maintain the semantic subject (staršie jedinceolder individuals) at the end of the sentence: P2 -From they territory they were pulled out by older animal; P4 -From their natural habitat they've been removed by the older ones.
In some cases, the participants did not strictly follow the ST structure, but deleted or added words.For example, in fragment 4A, 43% of them, while following the ST structure, omitted the verb: For example, bears.In fragment 7B (SurfOS share -29%), P1 used addition (problems with overpopulated wild pigeons or bats is created, while P3 -both addition and omission (the problems of wild pigeons or rats or bats).In fragment 10A, four participants' output (57%) generally matched the prediction (P1-And, for example, even in Trenčín or Tornala in the last few years; P3 -And, for example, in Trenčín or Tornal in the couple of days later; P4 -Even in Trenčín and Tornala in the last few months; P7 -For example, in Trenčín and Tornala, in the recent times).However, the variation in the time reference among the participants, ranging from days (P3) to months (P4) to years (P1) and to an indefinite period (P7), demonstrates the presence of the SensOS strategy as well.
In fragment 13A, five participants (71%) kept to the ST structure, simplifying the final part of the original phrase: P2 -because they are not interesting for their fur; P3 -They are not interesting anymore for their fur; P4 -they are no longer interested in their fur; P6 -also they are not more interesting to be used for their fur.P5 -they are no longer interesting as fur animals Transformation Strategy in Congruent Cases.Even when the structure of the fragment in the SL and TL was relatively similar, some of the participants still resorted to its transformation.For example, in translating fragment 1A, some participants resorted to the sense-oriented strategy (SensOS share -43%), using the fronting mechanism to move the nominal elements from fragment 1B to the initial position of the grammatical subject.Two participants fronted Animals (P2 -Animals are putting out from their home; P5 -The animals are being pushed out of their natural habitat) and both omitted the phrase Bližšie k ľuďom (Closer to people), which might be related to the limits the transformation effort imposed on the sentence coordination effort (see further).
In rendering fragment 4A, 57% of the participants, while generally following the ST structure, inserted the subject and predicate within it (P1-As an example, we can take, for example, bears; P4 and P6 -for example, there are bears; P5 -As an example, we could talk about bears).
A similar situation is in the case of fragment 4B, where two participants (29%), while maintaining the ST structure (preserving the verb in the final position), added the subject, converting the fragment into a full clause (P4 -from the Lučenec and Šahy suburbs and they get lost; S5 -in the area of Lučenec or Šahy and they basically got lost).
In translating fragment 6B, one participant (14%) transformed the ST structure, fronting foxes to the initial position (P4 -foxes are huge problem).And P5, while copying the structure of fragment 13A (they are no longer interesting as fur animals) added the fragment that distorted the ST content (The number of foxes has decreasedin fact, it increased).
Obvious Sense-oriented Strategy in Non-congruent Cases.When the unsuitability of the ST structure for its transfer to the TT is obvious (for example, when the grammatical subject in the ST is positioned at the end of the clause and the expectation rate of the predicted SurfOS version is low, because of the TL norms violation), most of the participants transformed the ST structure.
In fragment 1B, all participants used transformations, applying the passive construction (P1 -animals in general are pushed away from forests; P3 -animals are forced to leave the forest; P6 -animals are pulled back), or fronting the grammatical subject (P4 -Logging is preventing animals from their natural habitat; P7 -Logging and excessive population growth makes people…).
Likewise, in fragment 2C, all students fronted the grammatical subject: P1 -cities and villages are struggling…; P2 -people fight with…; P3 -many people fight…; P4 -there are people fighting with… P5 -people are having problems with; P6 -we are battling with; P7there are a lot of wild animals.
Similarly, in fragment 3B, the grammatical subject was shifted to the initial position (S1 -animals come closer… to people; S3 -many animals come closer to people; S4 -they (animals) try to get closer; S6 -they (animals) are coming to our towns) or there was introduced a new one (S2 -it (food) is so accessible in town; S5 -food, which is closer to the human settlements; S7 -the food is bringing them closer).
In fragment 4B, 57% of the participants transformed the ST fragment into a relative clause (P1-who somehow got into the surroundings of Lučenec and Šiah; P2 -which came to Lučenec and Šahy; P3 -which were spotted near Lučenec and Šiahy; P6 -which keep coming to the near Lučenec and Šiah).
In fragment 5, two participants (29%) undertook a substantial structure transformation: P3 -there are older kinds of these animals, which forced them to go back; P7-Older animals also make these other animals go away from their natural environment.
In fragment 10B, four students introduced the plural subject (P1people were dealing with overpopulation of wild animals; P4 -they were dealing with this cases of overpopulation wild animals; P5 -they also have got problem with overpopulation of wild animals; P7 -they have to deal with invasion of wild animals), while three more of them used there is/are structure (P2 -there is a problem of overpopulation of wild animals; P3 -there were many problems… and there were too many wild animals; P6 -There are also cases of too many species… in Trenčín and Tornala).
In fragment 11A, two participants (29%) moved the grammatical subject to the initial position and the adverbial of place -to the final one: P6 -citizens of Trenčín come across foxes in their park on daily basis; P7 -The inhabitants of Trenčín meet a lot of foxes in the town park.
Combined Surface-oriented and Sense-oriented Strategies in Non-congruent Cases.Even when the signs of unsuitability of the ST structure for its transfer to the TT were obvious and the participants transformed the ST structure, at least some of them tried to preserve the sequence of words in the ST by adapting the TT sentence structure accordingly.That shows that even in transforming the ST structure, the latter might still affect the TT fragment.
In fragment 1B, though all participants applied transformations, which is a proof of SensOS, some of them preserved the semantic subject at the end of the sentence, like in the ST (indication of SurfOS) (P1 -due to logging and overpopulation; P2 -by logging and also overpopulation; P3 -because there is logging; P5 -by logging in the forests; P6 -because of deforestation in forests and growing population).
The same is seen in fragment 2C, where four participants (57%), having transformed the ST structure, preserved the collocation towns and villages (the grammatical subject in the ST) at the end of the clause in the TT (like in the ST) having converted it into a modifier of place: P2 -people fight with animals in towns and villages; P3 -many people fight with many animals emerging in towns and…; P5 -people are having problems with wild animals and that being in the towns and villages; P6 -we are battling with wild animals that keep coming from the forests to towns or even villages.
Similarly, in fragment 4B, two more participants (29%), while generally keeping to the ST structure (preserving the verb in the final position), added the subject, converting the fragment into a full clause (P4 -from the Lučenec and Šahy suburbs, and they get lost; P5 -in the area of Lučenec or Šahy, and they basically got lost).
In fragment 5, three participants (42%), generally kept to the ST structure but fronted the subject-predicate group (P1 -And they were pushed away from their territory by older individuals; P5 -They have been pushed out from their territory by the older wild animals; P6 -They were pushed to the town by their elder relatives).
In fragment 7B, most participants (57%) transformed the ST structure into there is/there are clause, which required only fronting the said phrase, otherwise maintaining the original structure (P2 -There are also problems with overpopulation of pigeons and… (omission, probably failure to recall the equivalent under the stress); P4 -there are problems with overpopulation of boars (wrong recall, presumably because of the stress) or with pigeons, or bats; P5 -there are problems with overpopulated pigeons and bats; P6 -there are problems with too many pigeons or bats).
In fragment 7D, three participants preserved the ST structure, but added the grammatical subject in front of the clause (P1 -and that is dealt with by cities; P2 -this problem is solved by municipality; P5these problems are being dealt with local councils. In fragment 11A, all participants transformed the ST structure.However, five students (71%) preserved the adverbial of place at the beginning of the fragment, while transforming the remaining part of the clause: P1-And, for example, in their parks, there's like many foxes; P2 -and in the local forest, they could meet foxes; P3 -in the park of Trenčín, there were spotted wild foxes; P4 -In their park, foxes are meeting with people living in Trenčín; P5 -In the local forest park in Trenčín, you can also see foxes.
The same tendency can be seen in fragment 11B, where practically all participants (P2 omitted this fragment), while having transformed the ST structure, preserved (with sporadic additions) the adverbial of place at the beginning of the fragment: P1-and not only in parks but also in streets and in city centers, where there was this problem with one wild fox; P3 -as well as on the playground or city centre where there was an accident caused by fox; P4 -but also among the block of flats or in the city center where there is traffic and one fox was making a lot of problems; P5 -and also near the flats or in the city centre.There you could have seen a fox; P6 -also in suburbs and centre (unfinished).
In fragment 12, a number of participants, while applying transformations, tried to preserve the structure of some ST phrases: P1 -And the number of wild animals is supposed to be dealt with by hunters, and they want to do so by organizing some hunts;

P4 -They want to reduce the number of foxes by organization of common fox hunting; P6 -They want to reduce their population by organizing hunts on them
Effect of Transformation Effort on the Preservation of the ST Content and the Overall Utterance Control.The transformation effort seems to impose additional limits on the overall utterance control effort, which results in unmotivated pauses, backtracking, omissions and distortion of the ST information in the TT.
In fragments 1A and 1B, P7 undertook a total transformation of the ST structure, having fronted the subject (logging in forests and overpopulation) to the initial position and having moved the modifiers (closer to people and out of their natural habitat) to the final position.However, this process consumed so much of her efforts, as evidenced by the numerous unmotivated pauses and false starts in her TT, that the overall control over the correspondence of the TT content to the ST text was weakened.As a result, this TT fragment became meaningless: Logging and excessive … popu… population growth … makes people… turn away from their … natural environment and closer towards people.
In sentence 2 (fragments 2A, 2B, and 2C), the cognitive overload (proved by unmotivated pauses in the TT production) resulted in misinterpretation of fragments 2A (Still … there are many towns and cities… and villages… built in the last years…) and 2B (and many people fight… with …), as well as an unfinished fragment 2C (still with many animals emerging in towns and…) by participant 3.
Similarly, in fragment 2C, the transformation might have consumed too much of the P7's available effort, which resulted in the failure to render the meaning of the ST (there are a lot of wild animals living there).
By the same token, while rendering fragment 4B, P7 strictly followed the ST structure, but failed to finish the fragment (in the near outskirts of Lučenec and Šahy...), presumably, because of the cognitive overload.
Since transformation consumes a certain amount of available efforts and their supply is limited, the remainder may not be enough to effectively process the entire amount of the ST information and convey it in the TT.As a result, some of this information is omitted and not represented in the TT.This can be clearly seen in the translation of fragment 6A, where most of the participants (86%) applied various kinds of transformations combined with omissions.The latter was related to circumstances of place (among humans), frequency (more and more often), manner (wandering), type (hunting animals) or names of specific animals (roe deer, wild boar).P5 omitted circumstances of place and manner (Very often you can see the game, such as deers and boars); P2 -those of place, frequency and manner (but also we can find there… other animals as deer or boars); P4 -frequency, manner and type (Many other animals are... coming among the people, like… deers, or ... does, boars and other animals); P6 -place, manner and type (There are also… more examples of… animals that are coming to towns, for example, deers or boars); P7 -frequency, manner, type and the name of one animal (A lot of other animals such as deer and … and other wild animals come closer to people); P3 -place, frequency, manner, and type (There are … other wild animals… wild animals such as does or boars).
The extended sentence 7, with subordinate clauses and the inverted grammatical subject and predicate in the final position, required a substantial cognitive effort to deal with.In translating fragment 7C, under the cognitive overload, P2 omitted almost the entire ST content (because they are on the buildings); P3 essentially distorted the ST content (because they are emerging in people's houses through small holes which can be found in buildings); P6 did not finish her relative clause (which search for holes where they can live and …), probably due to the recall failure; P7 omitted the bulk of the ST information: (birds and bats) search for enclosures.The latter two participants (P6 and P7) failed to finish this sentence altogether.
The problems with the amount of effort required to simultaneously understand the ST, transform its structure, and control the compliance of the TT with the norms of the TL are aggravated when the source text is semantically and structurally complex.For example, sentence 9 (fragments 9A, 9B and 9C) contains pronouns (svojou, ich, tomu, ktorým, ich), whose contextual understanding requires their correlation with their referents.Given the need to transform the ST sentence structure in the translation process, such correlation can complicate both the understanding of the ST content and the control over the compliance of the TT with the TL norms.
In translating fragment 9A of this sentence, the participants used transformations with varying success: P1 -And… most of times, even people push them to do so because of their actions; P5 -Often people are the cause of these activity… of this migration of the animals… because of their activities in the forests.P7 attempted a very extensive retelling of the ST fragment with many hesitation pauses and additions to explain the implicit information: The ways in which people … affect their environments… the natural environments of mentioned animals, such as… forests and other places, … make... makes the animals … move away from their homes.P4's version (People are making animals to come into their natural habitat) looks ambiguous because of the vague meaning of the pronoun their (people's or animals'?),while that of P6 (because … we are not only one… that… endanger animals in… forests) is completely unacceptable having no relation to the ST content.
Problems with interpreting fragment 9A worsened in fragments 9B and 9C.The former was omitted by all participants (except P4), and the latter -by P1, P2, P3 and P4.P6 suggested a nonsensical version unrelated to ST (but also another animals they are terrestrial are problems), while P5 and P7 used transformations with a varying degree of detail (P5 -And that´s why the animals come to the settlements of the people; P7 -(animals) go towards people).
In fragment 11B, P7, having translated smoothly the congruent part of the fragment (but also in the middle of the town or on the outskirts and in suburbs) considerably slowed down while transforming the second part of it (…one specific… fox … made... made a lot of … problems there).
In fragment 12, the transformation consumed too much effort resulting in the omission of some of the ST information (in this case, the one related to organising joint hunts): P2 -when hunters want to decrease amount of these animals; P3 -hunters want to decrease the number of the foxes.
In sentence 13 (fragments 13A and 13B), in some cases, the transformation effort might have weakened the control over the TT meaning.Participant 3 swapped the cause-and-effect relationship between the two fragments of this sentence (P3: result -They are not interesting anymore for their fur, reason -because their number has rapidly increased, while in fact, the increase is the result).Similarly, P5, in the transposition of the two fragments, changed the result (increased) into the opposite one (decreased): P5 -The number of foxes has decreased because they are no longer interesting as fur animals.The remaining three participants (43%) failed to finish this sentence altogether (omission of fragment 13B): P2 -because they are not interesting for their fur; P4 -they are no longer interested in their fur; P6 -also they are not more interesting to be used for their fur.
In sentence 14 (fragments 14A and 14B), its ST structure was unfavorable for the SurfOS strategy.Thus, all participants had to transform it, but not all of them managed to transfer its full meaning to the TT.The only one who did manage it while preserving the sequence of the ST fragments, was P1: And hunters are not as interested in hunting foxes as they were in the past, because prices of fox fur has significantly decreased.P5 also managed to cope with the meaning transfer but reversed the sequence of the sentence fragments: The price of the fur of foxes has decreased, therefore the hunters are no longer interested in hunting them.P7 attempted similar permutation, fronting fragment 14B, but evidently had more problems in the process as evidenced by unmotivated pauses: The price for the… foxes'… fur is lower as well… The production of this fragment has probably consumed too much of her effort as she omitted fragment 14A altogether, adding (after a pause) the information that has already been mentioned earlier: This also affected… the larger number of foxes (see fragment 13B).An omission and addition have been also registered in the output of P6, only he omitted fragment 14B but added a similar information (and that's why they are not hunted anymore and that's why their population is still growing).P3 and P4 omitted one fragment each in their TTs: P3 managed to render only fragment 14B (So also the price of the fox fur has decreased), while P4 -just that of fragment 14A, and the numerous pauses in her reproduction prove the difficulties she experienced (So the... there are… no... fox… hu... hu… so there is no ho... fox hunting anymore).P2 attempted to reproduce fragment 14A (hunters also said that…), but presumably did not cope with the ST processing and gave up.
Similar processes are underway in fragment 15, where four participants transformed the ST structure using different approaches.P1 switched over to the direct word order by fronting the semantic subject (abundance of food) and substituting it with the pronoun this (And this also had huge influence on their overpopulation) -this substitution created serious comprehension problems as the abundance of food had not been mentioned before and it is impossible to infer its meaning from the context.P3 also fronted the semantic subject (abundance of food), but extended it into a separate clause (And there are enough food for them).Similarly, she extended the semantic subject (abundance of food) into another clause (So their number has increased).P5 effectively transformed the original clause into a passive construction (Their overpopulation was also caused by food sufficiency), while P7's attempt failed as in the process of her fronting effort, she lost the meaning she was trying to convey and finally gave it up (and as well as food that they can encounter…).P2, P4 and P6 omitted this fragment altogether.
The inability to simultaneously control the transformation of the structure and compliance with the TL norms can be also seen in fragment 16.Practically all participants (P2 omitted this fragment) transformed the ST structure.The most economical version was suggested by P7 (They can also make people contract rabies), who fronted the grammatical subject (fox) but substituted it with the pronoun they that would hardly be correctly interpreted by the addressee as foxes because they had not been mentioned before, and in the previous sentence, they stood for wild animals in general.P1 managed to transform the sentence in the way to maintain foxes at the end of the clause, like in the ST: And one of the main animals actually that is responsible for rabies is fox… are foxes.Though the ST was constantly in front of the students' eyes, other participants' versions looked more like retelling the ST they heard but did not manage to remember exactly -their TTs abound in pauses, repetitions, and unmotivated additions: P3 -and also there were illnesses … such as rabies.And… these illnesses can be… people can be infected, so the fox become really dangerous; P4 -If the ... if the fox is carrying around the rabies and it carries around to people, the fox can be really dangerous; P5 -The problem… we have with foxes is that they may… have… rabies… you can… they can also infect the people; P6 -Also, they are dangerous because they can… they can carry the rabies and … also, we, people, can get rabies from them and… sometimes it can be lethal.
This tendency has been proved in rendering the complex sentence 17 (fragments 17A, 17B and 17C).Though all participants attempted to apply the ST structure transformation, the effort consumed by it affected the quality of translation of most of them.Three participants made some unnecessary additions and repetitions, and omitted fragment 14B: P1 -And… when you actually contract this very dangerous… disease… or when the animal contracts this very dangerous disease, for example, fox… is not as scared, for example… (omission); and they come closer and closer to people.P3 -When the fox is infected, it's not shy anymore… (omission); it can approach people and it comes closer to people.Other omissions resulted in complicating comprehension (P7 -When a fox… has rabies…(omission), it is not typical in terms of its behaviour; and is going closer to people) or making the TT meaning inconsistent with the ST (P6 -and when they are… affected by these rabies… their fear… their fear is not a factor anymore and they are more brave (omission of fragments 14B and 14C); P2 -Can say that when there are animals with rabies it couldn't be…).
The problems with rendering long sentences with several clauses and specific syntactic characteristics seem to be connected with the participants' tendency to proceed in a linear fashion from left to right, assigning, in Moser-Mercer's (1995) words, a semantic and referential interpretation to each word as they are encountered in the ST.It results in numerous misinterpretations, backtracking and self-corrections.Attempts to overview several ST fragments at once to find out the author's ultimate intention, choose the suitable TT structure and carry out appropriate transformations, lead to the overload of the informationprocessing mechanism and short-term memory.The beginners need some experience (and, preferably, training) to acquire the skill of coordinating their efforts to balance them in the course of actual sight translation.It is not accidental that P5, the only participant with essential experience in SiT (see above), who has the least omissions and the fewest major errors in her TT (see table 3), was also the only one who overviewed the ST before actual translation.She made a noticeable but short (less than 2 seconds) pause before the beginning of many sentences to look the sentence through and choose the syntactic model.P7 also tried to apply this strategy but, because of the lack of experience, her pauses were very long (over 2 seconds each, the lowest speed of delivery in the group -42,7 words per minute -see table 3).In fact, her output looked like a consecutive (and not simultaneous) sight translation.Even so, it did not help much as the quality of her TT was quite low -11 omissions and 26 major errors.

Conclusions
Our study was aimed to establish the impact of the ST (Slovak) syntactic structure on the TT (English) quality and its delivery, as well as to identify the problems the interpreter students encounter in SiT, and the strategies they apply to solve them.
This first research that involves the Slovak-English language pair, has corroborated the assumption (Chmiel et al., 2020) made on the basis of other language pairs, that SiT makes high cognitive demands on interpreters, specifically related to the limits on the reallocation of cognitive effort (Gile, 2009).The data obtained show that the ST syntactic complexity is a factor influencing the SiT efficiency.Specifically, our experimental results have supported Gile's (2009) suggestion that the syntactic similarity in the two languages facilitates anticipation and replication of the ST syntactic structures in the TT.If there are no obvious indications that the ST structure is unsuitable for transferring it to the TT, the participants copy and adapt it to the TL norms in actual translation.However, the surface-oriented strategy, whose share among the participants as a group, roughly corresponds to the share of congruent fragments in the ST, does not dominate in the composition of the participants' translation competence.Even when the structure of the fragment in the SL and TL is relatively similar, some of the participants still resort to its transformation.When the unsuitability of the ST structure for its transfer to the TT is obvious, most of the participants use the sense-oriented strategy, which requires the transformation of the ST structure.However, even in this case, the ST structure may still affect the TT structure as some of the participants try to preserve the sequence of words in the ST by adapting the TT structure accordingly.
In line with the other authors' postulations (MacDonald, 1997;Gile, 2009), we have also found that the complicated ST syntactic structure has a negative impact upon the interpreters' anticipation mechanism and their processing capacity.In particular, as previously implied by Gile (2009: 198), it results in a higher short-term memory load, because of the need to retain the ST information before reformulating it in the TT.Our findings are also in good agreement with Viezzi's (1989) and Lee's (2012) suggestions that the interpreters' information-processing efforts are inversely proportional to the extent to which morphosyntactic transformations are necessary.
The transformation effort seems to impose additional limits on the overall utterance control effort, which results in unmotivated pauses, backtracking, omissions, distortion of the ST information in the TT, and the inability to simultaneously control the transformation of the structure, preservation of the ST meaning and compliance with the TL norms.The beginners need some training to acquire the ability of coordinating their efforts in actual sight translation: to look through several ST fragments at once, find out the author's ultimate intention, choose the suitable TT structure and carry out appropriate transformations.
Thus, the research results confirm our hypothesis based on Gile's (2009: 215) idea that the lack of the required processing capacity for coping with syntactic discrepancies between the ST and TT is one of the principle difficulties the interpreters face in SiT.
These conclusions, though preliminary, may be considered in preparing teaching materials for sight-interpreter training.
The prospect of further research is to study various aspects of sight translation on the material of other language pairs in different combinations (from language A to language B and vice versa) in order to compare them with the conclusions formulated in this article and to develop them further with the aim to construct a model of sight translation operation and its training.

ADHERENCE TO ETHICAL STANDARDS
Ethics declarations.Before the actual experiment, we obtained informed consent from its potential participants to take part in the research.In the course of our empirical research, we observed the applicable ethical principles, specifically the principles of voluntary participation, informed consent and confidentiality.
Funding.The research was funded by the EU Next Generation EU through the Recovery and Resilience Plan for Slovakia under the project No. 09I03-03-V01-00148.The authors do not have any financial or property interests concerning the materials presented in this article.
Data availability.The data underlying this article are available in the international repository Social Science Research Network (SSRN) at https://ssrn.com/abstract=4568241 (Chernovaty et al., 2023).

Table 2
The Source Text and the Predicted Literal Translation Version 2AČoraz častejšie v posledných rokoch, More and more often in recent years,2Bmedzi panelákmi a rodinnými domami, among apartment blocks and houses, 2C bojujú s výskytom divožijúcich zvierat mestá a dediny.have struggled with the presence of wild animals towns and villages.3A Aj pre potravu, ktorá jej chýba, Also, for the food they lack, 3B sa ťahá bližšie k ľuďom divá zver.move closer to humans wild animals.4A Ako príklad sa uvádzajú medvede, As an example, are mentioned bears 4B ktoré do okolia Lučenca a Šiah zablúdili.that into the vicinity of Lučenec and Šiah have strayed.5 Z teritória ich vytlačili staršie jedince.Out of the territory they were pushed by older individuals.6A Medzi ľudí stále častejšie zablúdi aj poľovná zver ako srnky či diviaky, Among humans, more and more often, are also wandering hunting animals, such as roe deer and wild boar, 6B ale veľkým problémom sú aj líšky.but a major problem are also foxes.9A Svojou činnosťou v lesoch ich k tomu neraz nútia sami ľudia, By their activities in forests, they are often forced to do so by humans themselves, 13A Aj preto, lebo sú už nezaujímavé ako kožušinové zviera, Also, because they are no longer interesting as a fur-bearing animal, 13B sa zvýšil počet líšok.has increased the number of foxes.14A Nevenujú sa poľovníci lovu tejto zveri tak ako v minulosti,.Do not hunt hunters this game as much as in the past, 14B pretože klesla cena líščej kožušiny.because has fallen the price of fox fur 15 Mal vplyv na ich premnoženie aj dostatok potravy.