Corps de l’article

1. Introduction

The ideological potential of language has for almost three decades occupied a prominent place in the field of linguistic research. Various models have been developed to account for the interrelationship between language and ideology, a substantial number of them set within the framework of influential Critical Discourse Analysis (CDA) (e.g., Fairclough 1989; Fowler 1991; Simpson 1993).

This phenomenon also attracted considerable attention in the Translation Studies community, giving rise to, among others, a number of CDA-oriented studies in the field of translation (e.g., Hatim and Mason 1997; Puurtinen 2000, 2003; Schäffner 2003; Kemppanen 2004; Sidiropoulou 2004).

With the aid of the concepts and methods applied in previous research on the relationship of linguistic form and ideology in translation, the present study aims at investigating the translational shifts of lexical and syntactic markers of ideology, employing mostly the analytical toolkit of CDA.

Before venturing on the analysis proper, we shall begin with presenting the main concepts underlying the research and outlining the socio-political context, i.e., Polish engagement in the Iraqi conflict.

2. Polish Involvement in the 2003 Iraq War

Since the main aim of the paper is to analyse Polish translations of the articles concerning the Iraqi conflict in 2003, it is vital to present Polish attitudes towards the war itself and towards Polish involvement in the Operation Iraqi Freedom.

According to the report published in 2004 by the Warsaw-based independent and non-governmental Center for International Relations, Polish involvement in the 2003 Iraq war has been the most controversial undertaking of the Polish foreign policy in the last 15 years (Wągrowska 2004).[1] These controversies are reflected in the wide spectrum of Polish attitudes towards this conflict. The official stance of the Polish government contrasted explicitly with that of the public opinion. Polish daily press, depending on the title, either supported or condemned the military invasion of Iraq.

On 17th March 2003, Polish President Aleksander Kwaśniewski announced that Poland would send troops to Iraq to assist in enforcing the UN Security Council Resolution 1441.[2] The decision of the Polish government to send troops to Iraq meant Poland became a side in the conflict. All governmental sources present a uniform pro-war attitude. Bukowski and Ścigaj (2007) point to a variety of persuasive techniques adopted in the governmental documents. First of all, the war in Iraq was conveniently placed within a broader category of the war on terror. Moreover, frequent references to the elusive everyone and all, as well as to equally vague international community reinforce the impression that there is a uniform international agreement that the intervention in Iraq is essential (Bukowski and Ścigaj 2007: 18). It is also worth noting that the lexeme war is hardly ever used in governmental sources, and the lexemes invasion and intervention are never mentioned. US and coalition invasion of Iraq is euphemistically referred to as allied action, disarmament, Operation Iraqi Freedom, participation in the operation, and taking part in the coalition. The analysis of six Polish governmental documents from the year 2003 conducted by Bukowski and Ścigaj (2007) reveals also that in its announcements and speeches, the Polish government mostly focused on the imaginary vision of the world if the intervention had been abandoned rather than on the war itself. There was no attempt to answer those criticising the decision of the government. All critical voices were ignored as if there had been none.

Bukowski and Ścigaj (2007) also analysed four Polish major dailies (Gazeta Wyborcza,[3]Rzeczpospolita,[4]Nasz Dziennik[5] and Trybuna[6]) in terms of their attitude towards the war itself and towards Polish involvement in it. Pisarek (2002) categorised these four newspapers in terms of their political profile on the basis of keywords (Pisarek uses the term słowa sztandarowe in Polish). According to him, Gazeta Wyborcza and Rzeczpospolita take a centrist-liberal stand, whereas Nasz Dziennik is believed to have a Catholic-national profile. The fourth daily – Trybuna displayed social-democratic leanings (it ceased publication in 2009).

The results obtained by Bukowski and Ścigaj (2007) show that Gazeta Wyborcza, considered one of the most influential and opinion-forming newspapers in Poland, displayed a strongly pro-American attitude. This was reflected in the use of marked lexemes and the fact that the war was described in terms of a political or media event without any references to moral, ethical or social issues. It is worth noting that although Gazeta Wyborcza appears to share the pro-war stance with the Polish government, the newspaper did not refer in any way to the official announcements of the Polish President, Prime Minister or any of the ministers. These were never discussed or quoted in the paper (Bukowski and Ścigaj 2007: 24-34). The attitude of another Polish daily – Rzeczpospolita has been interpreted as ambiguous by Bukowski and Ścigaj (2007: 34-44). On the one hand, in quantitative terms, the majority of opinions published in the paper tended to be in favour of the invasion. On the other hand, unlike in Gazeta Wyborcza, there were also some critical voices, although these were mostly the opinions of the journalists not employed at the paper. Similar ambiguity can be observed in Trybuna. This newspaper’s stance cannot be perceived in terms of a simple binary opposition of “for” or “against.” Trybuna officially supported the actions of the Polish government, publishing articles emphasising the legitimacy of the military intervention. Marek Barański, Trybuna’s editor-in-chief, stated in the editorial that the invasion is to a large extent justified. There were, however, also some commentaries describing anti-war protests and pointing to potential negative consequences of the war (Bukowski and Ścigaj 2007: 44-52). The last analysed daily, Nasz Dziennik, strongly opposed military intervention in Iraq. Its radical anti-war attitude is reflected at the lexical level – the invasion is described as unfair, hostile, vile and cowardly. Instead of the frequent media lexemes like operation, intervention and mission, the war in Iraq is instead referred to as aggression or assault (Bukowski and Ścigaj 2007: 52-59).

According to the reports of the Polish Public Opinion Research Center (Centrum Badania Opinii Społecznej, CBOS)[7] published between January 2003 and May 2004 (CBOS January 2003, November 2003, April 2004; Grudniewicz 2003a, 2003b, 2004),[8] Polish public opinion was strongly against the invasion in Iraq and especially the involvement of the Polish troops. The January issue of CBOS’s bulletin Polish Public Opinion reports that 51% of the respondents declared diminishing support for the intervention in Iraq. In a survey conducted in March 2003 the number increased significantly – 61% of the respondents stated there should be no military intervention in Iraq. More than half (66%) of the Polish population opposed participation by its troops in Iraq. The author (Grudniewicz 2003a) summarises the report as clearly indicating lack of support for the military invasion of Iraq. Similar results were obtained in mid-March by the Polish public opinion service TNS OBOP[9] – 69% disapproved of the Polish government’s decision to send troops to Iraq (Wągrowska 2004). The poll conducted in June 2003 confirmed that many Poles feared the presence of the Polish soldiers in Iraq would result in terrorist attacks on Poland (Grudniewicz 2003b). In October 2003 the level of support was the lowest since the first Polish military contingent left for Iraq. As many as two-thirds of Poles did not approve of Poland’s participation in the operation (CBOS November 2003). Another survey by CBOS from April 2004 showed that the number of the respondents who believed that Poland would become a target of terrorist attacks due to the presence of the Polish troops in Iraq increased almost twofold compared with December 2003. The number of Poles who associated the mission in Iraq with possible terrorist attacks had grown by 16 percentage points to 87% (CBOS April 2004). The report published in May 2004 notes that as many as 74% of the Polish population declared lack of support for the mission in Iraq. This increase was probably due to the death of two Polish soldiers and two journalists in Iraq, as well as frequent media reports about Human Rights abuses of the Iraqi POWs by the American army (Grudniewicz 2004). The fact that most of the Polish population was strongly against the invasion of Iraq and the involvement of the Polish troops in the mission is also emphasised in the report published in December 2004 by the non-governmental Center for International Relations (Wągrowska 2004).

3. The Notion of Ideology

Ideology is a concept which is notoriously difficult to define. Expounded and analysed in terms of a political doctrine or a philosophical stance, frequently used in contexts in which it is imbued with negative connotations denoting overt manipulation and deception, the term ideology is also assigned a more neutral meaning. This understanding of the concept of ideology is reflected in Simpson’s (1993) definition of this notion, whereby ideology derives from “the taken-for-granted assumptions, beliefs and value systems which are shared collectively by social groups” (Simpson 1993: 5). Mason’s (1994) understanding of the term also goes beyond the frequently used sense of political doctrine and instead focuses on a more neutral “set of beliefs and values which inform an individual’s or institution’s view of the world and assist their interpretation of events, facts, etc.” (Mason 1994: 25).

Such views on the term ideology, adopted for the purpose of CDA-oriented research and much favoured by other translation researchers, form the basis of Puurtinen’s (2003) understanding of this notion. In her research into linguistic realisations of ideology, Puurtinen uses ideology as the umbrella term for three phenomena. In her view, “ideology refers […] to the ways in which linguistic choices made by the writer or translator of a text, first, create a particular perspective on the events portrayed, second, may reflect the writer’s opinion and attitudes, and third, may be used to influence readers’ opinions” (Puurtinen 2003: 53). This definition has been adopted in the present paper to form the basis for the analysis of linguistic markers of ideology in press translation.[10]

4. Linguistic Markers of Ideology

The discourse features claimed by numerous researchers to serve as vehicles of ideological meaning are: grammatical metaphors, transitivity, modality, lexical choice, and cohesion.

Grammatical metaphor, the concept propounded by Halliday (1985/1994) within his framework of Systemic Functional Grammar, is perceived as an alternative lexicogrammatical realization of a semantic choice. The situation described in a sentence might be realized in two different ways on the level of syntax: by a semantically congruent construction, when semantic functions fulfil primary syntactic roles, and by a semantically non-congruent construction, i.e., grammatical metaphors, when semantic functions play secondary syntactic roles (Ravelli 1988). It is worth noting that metaphorical constructions and their congruent equivalents should never be perceived in terms of a mere variance in syntactic form, since different structuralisations might be referring to the same entities but do not communicate the same (Jędrzejko 1993). Variation in a lexicogrammatical realization is generally associated with differences in interpretation and different discourse and stylistic effects (Sušinskiene 2004).

According to Halliday (1985/1994), the most powerful resource for creating grammatical metaphor is nominalisation, i.e., substituting verbal constructions with nominal ones. By means of such transformation, processes and properties (congruently worded as verbs and adjectives respectively), are reworded metaphorically as nouns, which makes the resulting construction more abstract, impersonal and increases its lexical density. The manipulative potential of nominal constructions is also frequently underscored in other works (e.g., Fairclough 1989; Fowler 1991; Hodge and Kress 1993). Obviously, the reverse transformation, i.e., substituting nominal constructions with verbal ones (denominalisation), which is an instance of disambiguating grammatical metaphors, also carries a potential of ideological manipulation (see Gumul 2006b; Gumul and Łyda 2010). It makes process participants visible in the sentence and reduces the experiential distance, i.e., distance between text and the reality which it portrays (Puurtinen 2003).

Ideological import of a text might also be reflected in the transitivity patterns. The understanding of the concept of transitivity adopted in the studies on the linguistic aspects of ideology (e.g., Fowler 1991; Simpson 1993; Halmari 1999) is much wider than that employed in traditional grammars. It generally refers to how meaning is represented in the clause and shows how speakers encode in language their mental picture of reality and how they account for their experience of the world around them. In Simpson’s (1993) transitivity model there are four major categories of processes: material, verbalisation, mental and relational processes.

An important part of the transitivity system of language is passivisation. This syntactic transformation is claimed to have a strong effect on the way we perceive events and actions (e.g., Fairclough 1989; Fowler 1991; Halmari 1999; Awdiejew 2004). Passive sentences can be an effective neutralising (or mystifying) means of representing actions, processes and participants in such a way that causal relations and responsibility for actions become obscure (Puurtinen 2000).

Another discourse feature which serves as a vehicle of ideological meaning is modality. Following Fowler (1991), the notion of modality is understood here simply as comment or attitude, which is by definition ascribable to the source of the text, and explicit or implicit in the linguistic stance taken by the writer/translator (Fowler 1991). According to Simpson (1993), it is the epistemic modality which is possibly the most important regarding the analysis of point of view. Epistemic modality is concerned with the writer’s confidence or lack of confidence in the truth of a proposition expressed (Simpson 1993).

The writer’s or translator’s point of view might also be reflected in the lexical choice. It is often the strength of powerful connotations attached to seemingly equivalent lexical items which gives the source text and the target text different shades of meaning changing the ideology conveyed. A target text might also create an impression of different point of view if the semantic prosody is altered, which is “a consistent aura of meaning with which a form is imbued by its collocates” (Kemppanen 2004: 93).

Cohesion is another discoursal feature which is believed to have the power to modify the ideology of the text. Hatim and Mason (1997) underline the manipulative potential of lexical cohesion providing abundant exemplification of, for example, varying lexicalisation, which in Halliday and Hasan’s (1976) terms would mean substituting reiteration in the form of identical/partial repetition with reiteration in the form of paraphrase. Another type of cohesion referred to in the studies on ideology in language is explicitation of clausal relations, which takes the form of adding connectives (e.g., Sidiropoulou 2004).

5. Research Design

As all these features may potentially lead to distortions of the ideology projected by the source text, the aim of the present paper is thus to ascertain whether the lexical choice and the syntactic structures employed in the target texts engender changes in the ideological content of the source texts. The purpose of the analysis is not only to identify these types of syntactic and lexical modifications, but also to establish whether these operations are performed by translators in a systematic and consistent manner, resulting in the manipulation of the source-text ideology.

This study is partly set within the framework of Critical Discourse Analysis (CDA), and thus relies heavily on the models developed and employed by Halliday (1985/1994), Fairclough (1989), Fowler (1991), and Simpson (1993). The analysis is based on Hatim and Mason’s (1997) model of 3 degrees of translator’s mediation: minimal, partial and maximal mediation.

The corpus of the study consists of 22 feature articles which appeared in the Financial Times, the International Herald Tribune, the Los Angeles Times, The Atlantic, The Guardian, The Independent, The Spectator, The Times, The Washington Post and Time (see Appendix). The articles selected for the analysis deal with the Iraqi conflict in 2003 and its aftermath. The analysed target texts are the translations of these articles published in the Polish reprint magazine Forum[11] between January 2003 and May 2004, which altogether constitute a corpus of approximately 32,000 words. Forum published 22 articles concerning the Iraqi conflict and all of these articles were analysed in the present research project.

Forum is a Polish reprint magazine that publishes only translations of the world press articles. This weekly magazine was established in 1965 and has a circulation of 54,000. Their official statement regarding the paper’s editorial policy reads: The editorial staff reserves the right to change the titles and leads of translated articles and to shorten them as long as the changes do not distort the meaning of the original. The magazine’s own texts, commentaries, and editorial annotations are signed separately. According to the media report by Brief.pl,[12] the Last Issue Readership (COW) indicator is 0.4% (118,958) and Season Cycle Readership (CCS) is 0.8% (244,248). It is aimed at educated readers. Readership surveys indicate that Forum has a fairly high ratio of young readers (29% are below the age of 24), 29% are students and 39% are city dwellers. The magazine claims to have no ideological stance, which is also confirmed by Brief.pl media report.

Since the main function of the reprint magazine seems to be to acquaint the Polish readership with the opinions expressed by journalists in foreign press, the initial hypothesis holds that the vast majority of the articles might either mediate the source-text ideology unchanged or fall within the category of minimal mediation. We are referring here to the term coined by Hatim and Mason (1997) to refer to those target texts in which the scale of introduced translational shifts is relatively small. The other degrees of mediation in this classification are partial and maximal mediation, the latter describing a situation when translator’s mediation “issues from and constructs a different ideology” (Hatim and Mason 1997: 158) and when significant discoursal shifts occur between source- and target-language versions throughout the text. Thus, different degrees of translator’s mediation can be said to reflect “the extent to which translators intervene in the transfer process, feeding their own knowledge and beliefs into their processing of a text” (Hatim and Mason 1997: 147).

While discussing Hatim and Mason’s (1997) model of three degrees of translator’s mediation, it seems essential to mention the potential problems with its application in empirical quantitative research. The major shortcoming of the model is the lack of clear-cut quantifiable boundaries between individual categories. Thus, the author of the present paper made an attempt to adjust the model in order to facilitate quantitative analysis. An assumption was made that minimal mediation should involve no more than 3 shifts per each 1000 words of the target text, partial mediation – between 4 and 7 shifts, and maximal – over 7, on condition that the identified modifications form a consistent pattern in terms of the point of view conveyed. Obviously, such division is clearly an oversimplification, as it is not only the number of translational shifts that adds to the final effect of the target text, but also the force of the impact individual modifications have on the target audience. Nevertheless, it is hoped that establishing such borderlines might facilitate to a certain extent the quantitative analysis. It should be emphasized, however, that it is not the only criterion applied in the present research project.

Before presenting the results of the research, it seems essential to mention the major assumptions underlying analysis in the present study. First of all, it was borne in mind that no marker of ideology is inherently ideologically loaded. It is frequently underscored in works dealing with ideological aspects of language that “none of the linguistic forms is always an indicator of ideology in a text and should not automatically be interpreted as such” (Puurtinen 2000: 179). A particular linguistic form should never be automatically interpreted as an expression of a particular ideological meaning. Firstly, there might as well be a host of other reasons underlying translators’ surface choices, and secondly, these surface choices might not affect the ideological perspective of the text.

Therefore, although the analysis was performed on the level of texture, i.e., dealing with the micro-level manifestations of ideology, the judgement concerning the potential ideological significance of a given item was always made with reference to the text as a whole as well as the extralinguistic reality (i.e., the socio-political context).

6. Results

The results fail to support fully the initial hypothesis, as only 5 of the analysed articles mediate source text ideology unchanged and as many as 8 might be considered as falling into the category of partial mediation:

Table 1

Degrees of translators’ mediation

Degrees of translators’ mediation

-> Voir la liste des tableaux

In fact, the category labelled as “ideology unchanged” also subsumes those target texts in which potentially ideologically loaded constructions or lexical items could be found, but the pattern was not consistent, and therefore they could not be interpreted as projecting an altered ideological point of view. In 9 cases the translator’s involvement can be described in terms of minimal mediation, and none of the target texts appears to be promoting ideology that is fundamentally at variance with that of the source text.

The most common shifts (mentioned here in the order of frequency of occurrence) include alterations in: lexical choice, cohesion pattern, transitivity pattern, and the level of modality. Table 2 presents the results of the analysis – the total number of shifts occurring in each article and the dominant shifts:

Table 2

Most common shifts occurring in the analysed articles

Most common shifts occurring in the analysed articles

-> Voir la liste des tableaux

Presenting all the examples illustrating individual shifts is clearly beyond the scope of the present paper. Thus only a selection of the most representative instances of translational shifts of syntactic and lexical markers of ideology will be presented below. Each of the selected articles is represented by two or more examples in order to demonstrate the patterns of shifts within particular texts.

The first example illustrates changes in the lexical selection. The neutral item continued is rendered as prolonged (przedłużająca się) in the TT, giving it a distinctly negative connotation. It is further emphasised by the addition of a conclusive temporal marker of cohesion finally, which makes the acceptance seem more delayed in time and even less probable:

There are some more shifts encountered in the same article, which form a consistent pattern with those we have just discussed:

The translator opts for a more explicit and emphatic form to mark additive relation (after all) and substitutes nominal structure with a verbal one (he already spent), a clear case of disambiguating a grammatical metaphor. Such operation brings to the surface the agenthood, pointing directly to the actor of the process. Addition of already appears to provide further emphasis. All these alterations seem to be adding to the condemnation of American policy in Iraq. The point of view presented in the target text appears to be even more critical of it than that of the source text.

Some of the other features encountered in this particular article, consistent with other, previously discussed, shifts, include, for instance, the addition of intra-textual headings: Help is superfluous or Dead end to refer to US occupation of Iraq. Even though the expression dead end is employed in the body of the source text, its recurrent use in the target-language version and assigning it a status of a heading, naturally standing out from the text (Reah 1998), further emphasises the problematic character of American endeavours.

Another feature of this article that merits attention is the eight-node lexical cohesion chain of the word occupation. The reiteration pattern is not fully maintained in the target text, as there are significant shifts in four of the nodes (b, c, e and f):

In nodes (b) and (e), there is a shift of focus to another element of the script occupation – from the aggressors to the victims. Additionally, in node (c), there is clearly a metonymic relation between source-text item America’s occupation and its target-text equivalent Americans. Due to this shift, it is not the impersonal phenomenon of occupation that is to blame, but the actual people – the Americans. Shifts within this particular reiteration chain may also have a more global impact on the target-text reception, as reiteration or recurrence, to use de Beaugrande and Dressler’s (1981) term, is a textual feature “used to assert and reaffirm one’s viewpoint” (de Beaugrande and Dressler 1981: 55). Thus, any modification of a cohesive chain of this type is bound to result in changes of the point of view conveyed.

The following two extracts from another article illustrate, among others, a marked tendency of altering the level of modality:

In the above example, a definitive will is rendered as it is possible, which results in a drastic change in the level of certainty. When we analyse this particular example in the context of other shifts introduced to the target text, it becomes obvious that it might attest to the translator’s lack of confidence in the truth of the proposition expressed or alternatively, to his or her intention to create such an impression on the target audience. The impression of an undermined importance of Qatar is reinforced by other shifts: addition of the whole proposition it is not a mirage, implying the implausibility of the whole project, and omission of the details regarding the reasons why Qatar forged close links with America (to protect itself from predatory Arab neighbours). Substituting the latter information with the adjective pro-Western strengthens the impression of an altered point of view projected by the target text.

The case is further illustrated by another extract from the same article, in which the choice of the tentative form sometimes, implying occasional commitment of the state, to substitute a definitive and unconditional even if it means, marks the shifts in the level of modality. The highlighted lexical selection seems to communicate a message that is consistent with the other lexical and syntactic shifts that can be found in this text. Highly appreciative item remarkable is rendered as merely efficient – a form implying adequacy or acceptability rather than esteem:

The article in question abounds in other shifts in modality, forming a consistent pattern with those discussed above.

An extract from another article illustrates two more types of translational shifts:

In the first place we can see the consequence of a lexical choice: juxtaposing the noun responsibility with its collocate bears changes the semantic prosody of the utterance, loading it with a distinctly negative connotation and implying blame. This impression is strengthened by the omission of the phrase where to go in the target text.

The following example coming from the same source is an instance of alteration in the transitivity pattern – a shift from material to relational process:

This results in a clear change in the projected point of view: source-text letting problems slide out of control suggests negligence, giving an impression of an involuntary action in a way, whereas the target-text solution – being incapable to solve problems points to incompetence.

The following extract from another article provides further examples of similar translational shifts:

This particular fragment shows how the picture of the American/British and Iraqi attitudes towards the war is altered in the process of translation. Rendering a significant minority as a significant number clearly changes the expectations of the readers as to the actual numbers of people behind this general phrase. This change in proportions is strengthened by the omission of the conditional sentence that follows immediately. This rather emphatic sentence seems to suggest a certain degree of uncertainty and caution, possibly even a desire not to appear over-confident. That is at least how the author of the article attempts to present the American/British position. The rhetoric question, which takes the form of a matter-of-fact statement in the target text, appears to have the same function. Thus, as a result of such modifications, the reader of the target text receives a false message that implies more doubts on the part of the Iraqis and more certitude on the part of the Americans and the British than is suggested in the source text. It is also interesting to note that whereas in the source text, by means of a graphic convention, American and British involvement is not equally prominent (the Americans (and the British, for that matter)), in the target text both nations are put on an equal footing, which implies the same degree of involvement of both subjects (the Americans and the British).

Another extract from the same article illustrates shifts that appear to be consistent with those discussed above:

The shift that has some influence on the point of view conveyed in the text is the alteration of the cohesion pattern: emphatic even is substituted with additive also, making the proposition appear more natural and predictable. This impression is reinforced by a shift in modality – addition of otherwise (skądinąd), which expresses less confidence in the truth of this proposition. Moreover, there is another translational shift that forms a consistent pattern: whereas the source text reiterates the information (those who say we were right to go to war – the pro-war camp), the target text omits the second part of this cohesive tie, weakening the impact of the proposition.

7. Conclusions

Although the exemplification presented here is far from exhaustive, the results of the present study reveal that a substantial proportion of the analysed target texts communicate an altered point of view to the Polish readership. Despite the fact that none of the text projects a fundamentally different ideology, the pattern of the translational shifts is sufficiently clear to legitimize a conclusion of an altered viewpoint.

As illustrated by the examples extracted from the analysed texts, the articles displaying a consistent pattern of ideologically-loaded translational shifts tend to reflect the anti-war attitude of the Polish public opinion rather than the government’s pro-war stance. In many cases the analysed translations are more critical of the invasion of Iraq than the source texts. It is worth noting that none of the target-text presents military intervention in Iraq in a more favourable light.

Such changes might be seen as reflecting translators’ personal opinions and attitudes. Obviously, it would be tempting to assume that they are fully conscious choices of the target text producer. However, given the nature of the analysed material (press translations performed by anonymous translators) and the research design (purely product-oriented analysis), it is hard to determine whether a given shift is a result of a deliberate translator strategy. In fact, as underscored by various researchers, such modifications, even if reflecting a personal point of view, are not necessarily conscious. One cannot also rule out the possibility that some of these shifts might be attributable to lack of expertise on the part of a translator, although the shifts that were incidental were not taken into account and the criterion of a consistent pattern was of a paramount importance. Finally, it is worth mentioning that since a substantial proportion of discussed translational shifts fall either within the category of explicitation or implicitation (see Gumul 2006a, 2008, 2010), both of which are widely recognized as translation universals, some of these modifications might simply be seen as a by-product of language mediation.

Notwithstanding the exact reasons for such translational decisions and the resulting shifts, the analysed material proves that lexis and syntax are powerful tools in the hands of translators and that seemingly innocent and neutral operations involved in interlingual transfer might have serious ideological consequences.