FROM SILENCE TO READING BETWEEN THE LINES: ON SELF-CENSORSHIP IN LITERARY TRANSLATION

Abstract

According to a free online version of Encyclopedia Britannica, censorship is "....the changing or the suppression or prohibition of speech or writing that is deemed subversive of the common good" (EB), which seems to be the classical definition found in almost all reference books. [...]silence, which is an integral part of the text as well as a language in its own right, has as many interpretations and roles in the translated text as any other parts of the source text; it thus urges associations that helps one read between the lines.2 After a time publishers became trapped between translators and the State, more precisely the authorities, and duly tried to neutralise the translators' 'intellectual game' and, most certainly, to escape any possible retaliations, therefore they employed readers, so called 'literary advisers', to read foreign books before commissioning any translators. Anything prohibited during the Kádár-era surfaced, and, with a few exceptions, aesthetic value became secondary. [...]public taste, as well as a number of political memoirs and the emergence of pulp-fiction, became central to generating demand on the market. 7. Dictionaries used http://www.merriam-webster.com/dictionary/censorship [accessed 2013, May 30 ÁGNES SOMLÓ Pázmány Péter Catholic University, Piliscsaba Ágnes Somló, a graduate of ELTE Budapest, has been teaching at Pázmány Péter Catholic University, Pilicsaba, Hungary since 1997 and is head of undergraduate and postgraduate translation programmes. Besides literary translation (nearly 50 published volumes), she is the author of two radio plays and several literary essays as well as articles dealing with translation studies.

Headnote

Abstract. There are countries where censorship is officially declared by a censorship law, but there are others where this law is missing (even officially denied as it was in Hungary), yet censorship is still present in a hidden form. To avoid any kind of trouble as a consequence of breaching a non-existent law, self-censorship is used as an effective form of prevention. The paper will try to demonstrate that self-censorship in literary translation works in various ways and on many different levels.

Keywords: censorship/self-censorship; rewriting/manipulation; external and internal constrains; intellectual freedom; political, sexual and religious self-censorship

1. Introduction

"If a translator allows ideology to color anything he or she translates, the profession suffers. And when translation is stifled either by repression or selfcensorship, entire nations are deprived of a glimpse into the mind of the Other" (Landers, 2001:86). This statement seems to present a dream, an ideal situation, a kind of 'neutral' socio-cultural environment devoid of time and place, which is non-existent, as all translators live in a particular society, and translate for certain target readers, who represent a group of people speaking the same language and sharing, by and large, the same cultural environment at a certain place and time. Thus, inevitably, there will be some kind of external and/or internal constraints referring, among other things, to "social taboos, political taboos, religious taboos" (Xi Chuang, qtd in Higgins 2012, par. 10) that may lead to the rewriting and general manipulation of a text. At the same time, translation itself is rewriting and the process of translation moulds the source text into a target text that should fulfil some requirements or accepted norms of the target reader; it follows then that the text will be manipulated or even self-censored by translator, editor or publisher. The sensitivity of different texts is one of the most crucial elements when dealing with translated literature:

The publication of a translation is a very special form of exposure. Like any other exposure or revelation of things hidden from view[...], the site, the timing and the purpose of the exposure as well as its extent and the company in which it takes place are crucial factors in determining what can and what cannot be exposed. (Holman 1997: 280)

Before elaborating on the possible forms of and reasons for self-censorship in literary translation, I would like to bring together some likely definitions, explanations and approaches to censorship itself, and see what impact they may have on culture and, consequently, on translators. According to a free online version of Encyclopedia Britannica, censorship is "....the changing or the suppression or prohibition of speech or writing that is deemed subversive of the common good" (EB), which seems to be the classical definition found in almost all reference books. Among its other forms, obscenity, for example, is considered to be a type of subversion of "the common good", decency in this case, which is an aspect of the long-established moral structure of society. More complex is the definition of Sue Curry Jansen (1991:221), who describes censorship as "all socially structured proscriptions or prescriptions which inhibit or prohibit dissemination of ideas, information, images and other messages through a society's channels of communication, whether these obstructions are secured by political, economic, religious, or other systems of authority". On the other hand, Bourdieu's approach (1998: 15-16) is slightly different, because he basically divides modem censorship into three categories: 1. political censorship, 2. self-censorship and 3. economic censorship, while Al-Quinai (2005: 496) states that there are traditionally four grounds for it: ''sedition, blasphemy, obscenity and libel which can be paralleled to a quartet of state, religion, decency and individual privacy", and that "[t]he potentiality of a given text to be sensitive depends on its content, context and target readership." Bourdieu's categories seem roughly to correspond to Jansen's "market censorship", a concept of which she examines the origins, development and limits and which "points to practices that routinely filter or restrict the production and distribution of selected ideas, perspectives, genres or cultural forms...[and] ... refers to the conditions of production and consumption that produce cultural hegemony" (Jansen 2010:13, 14).

According to another general categorization (merriam-webster.com/), we have two types of censorship: one that is "pre-emptive (preventing the publication or broadcast of undesirable information) [and the other is] punitive (punishing those who publish or broadcast offending material)". There are countries where censorship is officially declared by a censorship law, but in most cases such laws are absent, yet during periods when there is a crisis or political instability, a lingering opposition to a newly founded totalitarian regime or in wartime, in short, whenever the State (or power) is rather vulnerable or is afraid of becoming exposed, open censorship, or some forms of censorship of printed matter becomes usually more severe, irrespective of whether or not there exists a law. For example, although there was no censorship law in Hungary under the communist regime, and the existence of censorship was officially denied, it was still present in a hidden form; in order to avoid any kind of trouble resulting from breaching a non-existent law, a kind of self-censorship functioned as an effective form of defence.

To evade censorship under authoritarian political regimes, translators encode hidden oppositional messages that they know can be deciphered by their educated readers. In doing so, they bolster their own as well as their readers' cultural capital and their claims to cultural leadership, and further create a common group defined by its intellectual skills and oppositional mood. (Baker 2010: 97)

Following the 1956 revolution, which was an inevitable response to the harsh dictatorial regime of 1949-1956, the reinvented much softer regime gradually introduced a new and revised cultural policy; instead of maintaining a system that either suppressed or supported the works of particular Hungarian writers, it introduced the so called "three Ps" system: promoting, permitting or prohibiting the work of certain authors, which meant that, besides supported and suppressed writers, a new group of merely tolerated writers appeared. Consequently, it is quite clear that one does not actually have to be censored: fear of being censored in one way or another is sufficient to stimulate self-censorship1, which, in this case, might be called a kind of "conditioned" censorship.

One must therefore make a distinction between external and internal constraints in both censorship and self-censorship. On the whole, we might say that basically all forms of censorship, except self-censorship, seem to result from external pressures, i.e., pressure from an outside source. Yet, among the causes of self-censorship, we may also find external pressures, such as political (death threats, imprisonment etc.), religious (enforced by religious authorities such as the Inquisition) or economic (e.g. threatening one's livelihood) as well as internal pressures or considerations, based, among other things, on conscience, taste, education, social or religious background, general atmosphere etc. Part of the complex nature of the problem is well demonstrated in the following example:

...an eighteenth-century Portuguese translation of Robinson Crusoe omitted considerable material, notably the parts where Robinson explains to Friday what is wrong with the Catholic religion. Explanation is [...] in the biography of the [...] translator Henrique Leitäo, who had problematic relations with the Inquisition and thus engaged in self-censorship. By omitting the most contentious passages, the translator quite probably saved his skin. (Pym 2006:7)

This case is further analysed by Anthony Pym, who detects censorship at three levels: "the observed omissions in the translation, the biography of a self-censoring translator, and the European struggle between Protestantism and Catholicism" (7) that he compared to "concentric circles".

To describe the presence of internal as well as external pressures manifested in art (or translation), let us look at the possible causes of, for example, the silence of the artist. When a writer remains silent, it can be intentional (internal) or an imposed (external) state of muteness; thus, on the one hand, silence might be a kind of resistance or radical speech, and, on the other, a kind of punishment imposed on the artist by an outside force - we may perhaps compare the difference between the two to that between "exile" and "self-imposed exile". In her article on the functions of silence in modem Spanish literature, Janet Pérez (1984:115) describes some instances when Spanish writers used "literary silence as a reaction" to Franco's regime and ideology. To further expound this picture of silence, let us observe what Susan Sontag (1983:183) claims: "Silence is the artist's ultimate otherworldly gesture: by silence, he frees himself from the servile bondage to the world, which appears as patron, client, consumer, antagonist, arbiter, and distorter of his work." In other words, the artist actually frees himself from political or ideological censorship in the first case and from market censorship, as Jansen calls it, in the second, because in artistic communities "success in capitalist art markets is frequently equated with 'selling out', 'prostituting' one's talents."(2010:20) It was in fact due to the rise of modem capitalism at the turn of the 19th and 20th centuries that "control over the production and distribution of ideas and cultural artefacts was industrialized, and commodified and concentrated within private hands." (Jansen 2010: 19) And later Curry Jansen adds that "Most forms of market censorship are more elastic, but also more resilient, than religious or political censorship: that is, market censorship is a leaky system, but one that is easily repaired." (idem., 24)

2. Self-censorship and literary translation

When we turn our attention to self-censorship, which is an indirect form of censorship, and, in particular, to its working in literary translation, we may say that the translator who performs it has a variety of causes, reasons and aims. Generally speaking self-censorship may denote

...the creative decisions individual cultural workers make: decisions in areas where they have the autonomy to make choices to include some things and exclude others, or to use one approach or genre rather than another. Such decisions may reflect the individual's personal moral, rational, aesthetic or emotional preferences; they sometimes even express individual courage or temerity. (Jansen 2010:13)

According to Toury (1995:278), self-censorship can be defined as censorship "activated during the act of translation itself though, inasmuch as the translator has internalized the norms pertinent to the culture, and uses them as a constant monitoring device." The same is echoed in the conclusion of Al-Quinai's (2005:523) article stating that, on the one hand, "the translator cannot be exonerated from his role as a censor/mediator", while target text readers, on the other, "may impose their own expectations on the translator who modifies the TT to conform to established sociolinguistic constraints".

Also self-censorship can roughly be divided into three major fields: political, religious and sexual. Any of these and their subfields present in a text may be the cause of its becoming sensitive; the translator himself may be "already politicized [and] may take offence at texts that are unpalatable or politically unacceptable." (Al-Quinai 2005: 489) Or, on the other hand, as the translation of a literary text is mostly commissioned by a publisher, who in turn wishes to sell the final product, the translator might receive explicit instructions to pay attention to certain political or religious topics or simply to respect the target readers' sensitivities, and so, in spite of all our independence, courage or freedom, as soon as we write or translate to earn our living, self-censoring our words and work becomes imperative, because "[o]nce creators take their creations to market, they are subject to its disciplines". (Jansen 2010:13) In the case of a literary translation, there is always the possibility of either refusing to publish a certain piece or of deliberately interpreting the source text so as to create a different image in the minds of the target language reader.

2.1 Political (self)-censorship

Publishing was heavily subsidised in Hungary during the Kádár-regime, since the leaders of the State "were firmly convinced of the educational power of literature in the process of building Socialism; [thus] literature [...] was given an almost exaggerated significance" (Gombár, qtd in Czigányik 2011:225). Consequently, official cultural policy created a demand for translated literature as well as for educated and well-qualified translators. Traditionally, Hungarian writers often turned to translation for a variety of reasons, one being the silence imposed on them: "writers silenced by the regime, needed work to make a living. While translation offered the possibility of congenial employment, it was not a steady or full-time job, only an occasional contract, so politically 'unreliable' individuals could get away with taking it on from time to time." (Somló 2003-2004:320-321) At the same time not only Hungarian, but also foreign writers were 'silenced', i.e. their publication was not advised for ideological reasons, therefore at first only "[w]ell-known and 'harmless' eighteenthand nineteenth-century novels were retranslated" (idem, 321); later, from the late 1960s and '70s on, contemporary literature was also translated, but the choice for publication was ideologically directed by way of pre-emptive self-censorship.

The system of censorship throughout the Kádár era rested on two pillars. One was the firm conviction of the morally and intellectually constructive influence of literature (which is why pessimism or decadence was seen as a major argument against the publication of a book). The other pillar was the exclusion of political taboos, the most important of which were the following: criticism of the Soviet Union or the one-party system, anti-Marxism, and ironically, the existence of censorship. (Czigányik 2011:228)

As we have seen, suppressed ('prohibited') or merely tolerable ('permitted') writers were to turn to translation because it offered the possibility of occasional employment as well as a hidden way of expressing their thoughts and views by way of 'interpreting' the message of a foreign writer. For example, silence in the text, be it an omission or some consciously created hiatus, etc. along with various allusions - most probably understood mainly by intellectuals - ensured that, by reading between the lines, the reader would decipher the conveyed message; thus a form of intellectual freedom was maintained primarily by translated literature. As Al-Quinai (2005:495) says "we depend far more on the connotations of words [...] than on denotations. [...] It follows that no text is sensitive by itself, but the interpretation of its connotations makes it so." Thus silence, which is an integral part of the text as well as a language in its own right, has as many interpretations and roles in the translated text as any other parts of the source text; it thus urges associations that helps one read between the lines.2

After a time publishers became trapped between translators and the State, more precisely the authorities, and duly tried to neutralise the translators' 'intellectual game' and, most certainly, to escape any possible retaliations, therefore they employed readers, so called 'literary advisers', to read foreign books before commissioning any translators. The role of these readers was to read the book and tell the publisher if it was worth publishing in Hungary. The manner in which a reader's suggestion was written was the following: s/he summarised the content of the book; said something about the writer; wrote a critical assessment pointing out the merits and drawbacks of the piece (which also meant paying attention to any politically or otherwise sensitive passages), and finally suggested the possible range of readers and stated whether publication of the book was recommendable or not.3 Each book was read by two people, and, in the case of conflicting opinions, a third reader was asked to do it and settle the dilemma. By the beginning of the eighties, the importance of the reader's role must have declined, because, fresh out of university, I was entrusted by one of the most prestigious publishers to write such reports on books, and mostly make an aesthetic evaluation. I had to write primarily on African pieces, but sometimes I received Australian, American and even English novels or plays as well. This distinct change in the strictness of control could probably be explained by the course of events: there was a gradual softening of the regime that had earned Hungary its nickname - the most comfortable barrack in the Soviet camp. This was also observed by Czigányik in his essay:

From the mid-1970s until the end of the 1980s increasing financial problems forced all participants in publishing to take economic considerations more seriously, which meant that the demands of the market were also taken into consideration. [...] the potential success of a book was also calculated, and it was with a view to its commercial potential that the otherwise despised popular culture also appeared on the market. These factors together led to the easing of political control over the publishing industry (or to the diminishing effectiveness of political control) by the 80s. [...] The slackening of ideological control is reflected in the fact that in the late 70s and early 80s editors often tried to publish books that had been rejected in the 60s. (Czigányik 2011:227)

Despite the lessening pressure, well embedded and conditioned reactions such as self-censorship remained active at certain levels of management for a long time. Let me present a telling example from my own career. As a freelance translator I often worked for the Hungarian Radio, namely the Radio Theatre during the nineteen eighties. I was mostly commissioned to translate African plays that I was also allowed to select thanks to one of the dramaturges I had been working with for a long time. Following some successful productions, I ventured to suggest a radio play with a more universal topic, yet politically imbued - FullCycle, written by Gordon Tialobi. In it a general is tried for war crimes committed by soldiers under his command or so it seems. General Maga, awakened in the middle of the night by armed Man and Girl, is totally defenceless. He has no real choice: either he acknowledges his guilt or they kill him. Following a senseless or senselessly meaningful trial, which finally turns out to be a nightmare when Maga wakes in the morning, the play leaves open the question whether he was guilty or just the victim of a fabricated case, because he lies back for a minute and dozes off and the whole thing starts all over again. It is a "full-cycle", and so we do not really know whether it is reality or another nightmare. The dramaturge also liked the play, considered it worth producing, and submitted the Hungarian version to be reviewed by the Dramaturgic Council led by the head of Radio Theatre. But even translated texts were sometimes refused, as was the case with my translation. Although the dramaturge kept on submitting it, the manager of Radio Theatre kept on refusing it, with the excuse that he "did not like it so much"4. The emphasis is on 'so much', meaning he did not feel like taking the risk that some powerful people might also be able to read the message between the lines.

As I have already mentioned, the message of this Nigerian play "was rather universal: who is to decide or judge the individual's historical responsibility at turning points; and what is the nature of power in such cases?" (Somló 2010:137). The text described the working of "the senseless administration of faceless power over the individual, so typical of dictatorial, totalitarian regimes." (ibid.) However, the Hungarian audience was familiar with the problem discussed in the play as well as with the feelings it evoked, "as the so called 'conspiracy cases' of the 1950's - when innocent people were forced to plead guilty and when the state police could appear and take people in custody in the middle of the night - were still clearly remembered."(ibid.) Hence the meaning read between the lines was quite clear and this, I suppose, might have been the real cause of "silencing" the Hungarian version of the play. Thus, self-censorship (now on management level), yet again, took its toll.

Having seen an example of translation "silenced" due to self-censorship, I will offer yet another instance of political self-censorship in translation. This time it is a novel, Survive the Peace, written by another Nigerian author, Cyprian Ekwensi. The narrative is set during, or rather at the end of, the Biafran war, which in a way, similarly to many other 'minor' clashes of the period all over the world, served as a kind of test of strength between the two super-powers, the USA and the Soviet Union. The main character, a radio-journalist, is fleeing across the country trying to reach his family. On his way he meets various people and survives difficult situations, finally to die at the hands of armed robbers when the war is over and when he is full of hope. We see the real face of war, the suffering of people, and there is nothing heroic about it. There are two instances in the novel that were considered sensitive by the official ideology: at one point there is an aircraft attack on the people on the road and someone exclaims that those Russians should take their planes back home. The fact that both powers had sent arms was an open secret; we all knew how it worked, and so I decided to cut the word Russians from the text: that would not affect people's insight, but might blur the vision of the authorities and help to get the novel through. In the end, the novel was not published, because at some other point, someone refers to the Hungarian revolution of 1956, saying that in fact no one had really cared about that either, because the world and its politicians had had more urgent problems to deal with and simply let it be crushed. Now this was something I was simply unable to change or leave out, and thus the text managed to touch upon one of the cardinal taboos of the Kádár-regime: 1956.

2.2 Sexual self-censorship

Another field of self-censorship in translation covers sex-related themes and scenes, as well as linguistic elements such as swear-words or other fairly naturalistic descriptions of the human body or of sexual intercourse. In short, topics considered pornographic or obscene will all fall into this category. In these cases, self-censorship would make use of: "deletion or dilution of taboos or offensive material on religious, ideological or social grounds [and] compensation is conducted by way of using euphemistic expressions" (Al-Quinai 2005:511-512).

Following the change of the political regime, Hungarian publishers mostly lost all financial support and had to turn towards fulfilling the requirements of the market. As the Hungarian market is relatively small, aesthetic considerations became secondary. Among other things5, a number of pornographic as well as rather more directly sex-related books appeared on the market. As a result, some of the main forces guiding the work of translators became taste and decency, predominantly based on linguistic, literary and cultural traditions of the target culture.

For example, much of what Henry Miller says in his Tropics of Cancer6 has lost the power to shock the average American reader the way it did when the books were first published. Yet the same 'semitaboo' material is still considered extremely obscene in Spain or South America where translators are expected to sanitize the ST by exercising self-censorship and decorum. (Al-Quinai 2005:496)

In Hungarian we have yet another problem: the language of sexuality - sexual organs, sexual intercourse etc. It is quite difficult to translate a quasistandard, delicately erotic text, because we have at our disposal either the scientific names, mostly of Latin origin or, at the other end of the scale, a really rude and coarse vocabulary, so that finding some way of describing sexual intercourse in detail using standard Hungarian calls/called for much creativity. ''Translating is always a struggle to reach a compromise between one's ethics and society's multiple constraints-and nowhere can we see this more clearly than in the rewriting(s) of sex-related language" (Santaemilia 2006:246). It was especially difficult in the case of the historical romances that I was often commissioned to translate in the 1990s. Sexuality in printed matter used to be (and still is to a certain extent) an extremely sensitive area. As Santaemilia concludes:

Sexual language is a privileged area [...] a site where each culture places its moral or ethical limits, where we encounter its taboos and its ethical dilemmas. [...] we cannot prevent a certain degree of self-censorship, along the lines of an individual ethics and attitude towards religion, sex(uality), notions of (im)politeness or (in)decency, etc. (2008: 246)

During the translation of some of these historical romances, I found far too many meticulously described, 'hot' sex scenes that tested my linguistic ingenuity. After the first book had been edited, I saw that some of those scenes had been deleted by the editor, and when discussing the final version, we both agreed that Hungarian readers (mostly women in this case) were not prepared to face the same passages of sexual scenes so many (more than 6) times. Both the situations as well as the surroundings seemed to be repeated over and over again, so we considered at least half of them superfluous. By cutting these erotic scenes - which were intended to serve their own ends, without adding anything new to the story or to the characters' development - we managed to create a more dynamic narrative and a softer language that was more palatable to our readers at that time. When next commissioned to translate a historical romance, written by the same or another writer who used more or less the same style and language, I was asked to adapt the translation to meet the taste, expectations and requirements of the prospective readers of such books. "This strategy seeks to increase the book's appeal for the average reader[...] Under such circumstances, the translator assumes the image of a traitor who commits a deliberate act of rewriting, adapting and censoring an original which is deemed not fit for publication in the receiving culture in its ST form" (Al-Quinai 2005:516).

2.3 Religious self-censorship

The third main category of self-censorship I am going to consider is that involving religious topics. Religious connotations, moreover, religious texts are perhaps the most sensitive and "the translator's intervention to censor the text becomes mandatory. For instance, a derogatory reference to religious characters is strictly forbidden in an Islamic culture" (Al-Quinai 2005:502) and would cause great upheaval in deeply religious, Christian communities as well. Actually, here we have to face the special sensitivity of religious communities as well as the culturally laden intertextuality of such texts, which make the task of translators more difficult. Intertextuality is of course present in all texts, as "no text can ever be completely free of those texts that precede and surround it" (Bassnett 2007:82); yet religious writings, such as the Bible in our western culture, have an important effect on a great variety of other texts and interpretations.

Hungary is part of the Christian world and the majority of the population is Catholic, but Protestant Churches are also strongly represented by a large minority. In her essay, Aurelie Hagstrom says that "a dominant religion [might force] limitations on less prevalent ones" (2003:150). While the Catholic Church has passed through and gradually ended its period of severe censorship "most dramatically by abolishing the Index of Forbidden Books in 1966" (idem, 151) at the same time "the Church maintains its right to judge culture based on its own Christian values" (ibid.) It is especially true in the case of popular culture and a good example for the working of this kind of "self-censorship", or rather a kind of dialogue between different communities and their different sensitivities, might be "the announcement before a TV show that warns viewers that some language, violence, or material may be offensive to younger or sensitive viewers" (idem, 154).

A literary translator in the middle of Europe will inevitably come across quotations from the Bible or biblical allusions in a foreign text that he has to translate by using and relying on a target language version of the Bible. All its vernacular versions carry the common cultural memory and intertextuality of Christianity, as well as those of the target culture and of interpretations of the Bible in that culture.

Thus a biblical allusion in a foreign text is not precisely rendered by an allusion to the same passage from a Bible translation in the translating language. The allusion in the foreign text may well construct an intertextual relation to a version that, given the cultural importance of the Bible, has accumulated meanings, values, and functions which cannot be recreated merely by inserting an allusion to a Bible translation in a different language. (Venuti 2009: 161)

Venuti further elaborates on this by comparing the impact of Luther's version on German culture and of the King James Bible on the English language and culture and, besides pointing out similarities, he emphasises some significant differences. "Luther's version was instrumental in developing modem High German into the standard dialect and the literary norm, whereas the King James Bible is cast in early modem English, a phase of the language that grew obsolete by the nineteenth century" (2009:161); thus any quotations from or allusions to the King James Bible in an English text would carry different cultural connotations from the German allusions to the Luther version in its German translation.

I faced such problems when translating Elizabeth by David Starkey. Here I had to tackle not only the problem of the different Catholic and Protestant translations of the Bible, but also differences in the liturgy as well as in the wording and style of prayers - in short, I had to show a marked difference between the style and milieu of the Catholic reign of Queen (Bloody) Mary and the Protestant mle and Court of Queen Elizabeth I. It is the story of Elizabeth's early years, her 'Apprenticeship' and, as David Starkey is a renowned historian, the book is certainly based on thoroughly studied and analysed sources and methodically selected historical surveys and documents. Following her younger brother's shortlived Protestant mle, Queen Mary re-established Roman Catholicism that was changed after her death by Elizabeth. Concerning his sources on Elizabeth, Starkey (2001:327) says: "Other great influences, on both Elizabeth's thought and her language, were Cranmer's English Prayer Book7 and the English Bible. I have used both frequently."

Generally, when Hungarian literary translators have to use one of the versions of the Hungarian Bible translations of King James Bible, we are lucky to be able to turn to a beautiful Protestant translation, published in 1590. Its language is mostly understandable, yet slightly archaic, and thus it suits our purpose; it is perhaps the most famous in the series of Protestant Bible translations that were followed by Catholic translations, and by a good deal of polemics in the 17th century. The difference between the Karoli Version and the King James Version is not so pronounced as that between Luther's Bible translation and King James Bible. Certainly when Mary uses certain words or alludes to the Bible, I turned to a Catholic Translation of nearly the same period (1624). My real problem started when I had to make a distinction between the really subtle liturgical differences of the Anglican and the Protestant form used in Hungary, and when I had to find the right Hungarian wording corresponding to certain parts of the Book of Common Prayer quoted in the book. Unfortunately, none of the Hungarian prayer books of the reformed churches were of any use here, and I was not able to use any of the Roman Catholic prayers either, because what I had to reveal was the difference. Finally, I decided to create the language of these beautiful prayers in Hungarian by drawing some pivotal words from an early 17th century master of high-flown and powerful prose, Cardinal Pázmány. I used his correspondence and debates with protestant preachers in order to convey as much of the beauty of the language of the Book of Common Prayers as possible. At the same time, I managed to touch upon a very sensitive area, because both religious communities use their distinct wording in religious texts (this sometimes meaning only the change of a single letter) so I had to be very careful in my choice of words and try to navigate towards a more general and slightly archaizing usage. This kind of "self-censorship", rather than the manipulation of the target text, tries to respect the sensitivity of deeply religious groups as well as of the majority of the less pious by balancing on the verge of acceptability and, at the same time, by trying to convey the cultural and aesthetic message of the source text.

3. Conclusion

This paper has attempted to clarify some notions of censorship, especially self-censorship in general, paying special attention to the different types and methods of self-censorship in literary translation. Simple definitions have not proven satisfactory because the issue of (self)-censorship is manifold and fairly complex. We have seen that one of the main points of categorization is that censorship is mostly the result of some outside force, while self-censorship is rather self-imposed and indirect. At the same time, both may be initiated by external or internal pressures. We have seen various approaches to censorship and self-censorship, dealing with different aspects, and although I have briefly examined some of them, the classical triad of political, religious and sexual as basic categories seemed to serve as a logical starting point for the examination of self-censorship in literary translation. Yet again, I have made a distinction between two basic forms in all three categories; due to its underlying motivation, selfcensorship can be internal or external. External pressures are produced by various outside constraints threatening the translator's physical being, livelihood or freedom etc.; we even might say that, in this respect, self-censorship is a vital means of self-preservation. On the other hand, internal constraints originate in the translator's personality: his preferences, education, social surroundings, in short, any considerations within: "self-censorship is an individual ethical struggle between self and context. In all historical circumstances, translators tend to produce rewritings which are 'acceptable' from both social and personal perspectives" (Santaemilia 2008:221).

Severe censoring characterises totalitarian regimes or any authorities (be it the State, or the Church etc.) that are afraid of losing power, or feel vulnerable at some point in time. Bourdieu's economic censorship or Curry Jansen's market censorship are more difficult to capture, yet as soon as any artists - among them, in particular, literary translators8 - take their work to the market, they must accept the conditions dictated by it. (cf. Jansen 2010: 13) Thus translators have to take into consideration such external constrains as the expectations of readers, target norms, i.e. poetological and ideological norms of the TT culture, as well as inner conditions, such as the translator's own interpretation. Eventually all these are in connection with the special reading of translators at a given time and place. We always have to examine the above conditions in order to be able to answer questions like: what taboos, what religious or political considerations had the translator to accept in order to be able to publish his translation; and what were the directives that he would not or should not have accepted.

We have to make a marked difference between translators' self-censorship in democratic and in totalitarian or quasitotalitarian regimes. In the former, selfcensorship is a kind of freedom of interpretation as well as responsibility towards the reader of the target text as well as the author of the source text. It is the latter that requires ideological - mostly political - self-censorship, and it is this kind of self-censorship that Péter Esterhazy referred to in an interview for Der Standard: "Self-censorship is like melted snow. It will seep under your collar"9 (April 9, 2011); this kind of self-censorship may lead to a spiral of silence, for it is fear of isolation (social, even physical in certain cases) that triggers this phenomenon of silence. At the same time, we have seen that silence has a manifold meaning as well, because silence is also a means of communication. Silence is part of the text or the silencing of a text is part of the discourse as well.

I shall conclude by saying that self-censorship in literary translation may mean restriction or indicate freedom in interpretation, but it is never free of the impact of the time and the place in which the translator lives; therefore, although generalizations may be helpful, each case should be examined on its merits.

Footnote

1. Translation can be a dangerous job; some translators such as Étienne Dolet or William Tyndale even paid with their life for it in the 16th century; or a more recent case, the murder of the Japanese translator of Salman Rushdie's The Satanic Verses.

Footnote

2. Self-censorship can also be considered a form of preventive censorship to avoid a/ trouble, b/controversy, c/ offending the audience , d/ economic boycotts, e/ lawsuits (e.g., libel, invasion of privacy torts, etc.), V official censorship (self-imposed ratings to avoid government ratings) - see survey by J. Senat for his students at Oklahoma State University (http://joumalism.okstate.edu/ faculty/jsenat/ censorship/defming.htm [accessed 2013, May 30]).

3. See an explanation given by reception theory on the presence of stimulating blanks and gaps: "The discontinuities of the textual segments trigger synthesizing operations in the reader's mind because the blanks lead to collisions between the individual ideas formed, [...] These colliding ideas condition each other in the time-flow of reading" (Iser 2006: 66).

Footnote

4. Certainly, by knowing the rules of the game, publishers' readers were able to write a recommendation shaped to fulfil the requirements of the authorities, but at the same time make sure the book they wanted to "pass" would pass. Or at least they often tried to do so.

Footnote

5. Grateful thanks to Márton Mesterházi for sharing and revealing the truth behind the rejection of this play.

6. Anything prohibited during the Kádár-era surfaced, and, with a few exceptions, aesthetic value became secondary. Therefore public taste, as well as a number of political memoirs and the emergence of pulp-fiction, became central to generating demand on the market.

Footnote

7. A telling fact is that it was not published in Hungarian until 1991! â

Footnote

8. a.k.a. Book of Common Prayer.

Footnote

9. Literary translation is an ephemeral art; the "there and then" is conveyed by translators " here and now" consequently they want to have their work published and publishers are players in a capitalist market, governed by the regulations of that market.

10. "Selbstzensur ist wie geschmolzener Schnee, der in deinen Körper einsickert, da im Genick." 2011. http://derstandard.at/1301874201357/Interview-Ich-bin-ruhig-behaupte-ich-unruhig [accessed 2013, June 17].

References

References

Al-Quinai, J. 2005. 'Manipulation and Censorship in Translated Texts' [Online]. Available: http://www.aieti.eu/pubs/actas/II/AIETI_2_JQ_Manipulation.pdf [accessed 2013, May 30].

Baker, M. (ed.). 2010. Critical Readings in Translation Studies. London and New York: Routledge.

Bassnett, S. 2007. Translation Studies. London and New York: Routledge.

Bourdieu, P. 1998. On Television. New York: The New Press.

Czigányik Z. 2011. 'Readers' Responsibility: Literature and Censorship in the Kádár Era in Hungary' in B. Gárdos, A. Péter, N Pikli, M. Vince (eds.). Confrontations and Interactions Essays on Cultural Memory. Budapest: L'Harmattan Kiadó, pp. 223234.

*** Encyclopoedia Britannica Online. Available: http://www.britannica.com/EBchecked/topic/101977/censorship [accessed 2013, May, 30].

Hagstrom, A. 2003. 'The Catholic Church and Censorship in Literature, Books, Drama, and Film' in Analytic Teaching, no. 23 (20), pp. 147-156 [Online]. Available: http://www.viterbo.edu/analytic/vol%2023%20no.%202/the%20catholic%20church %20and%20censorship.pdf [accessed 2013, June 10].

Higgins, Ch. 2012. 'Social Media and Online Comments "causing writers to self-censor'" in The Guardian, Tuesday 21 August, 2012 [Online]. Available: http://www.guardian.co.uk/books/2012/aug/21/social-media-stifle-literaryendeavour#ixzz2UgLmRTmO [accessed 2013, May 30].

Holman, M. 1997. 'The Sanification of Tolstoy's Resurrection' in K. Simms (ed.) Translating Sensitive Texts: Linguistic Aspects. Amsterdam: Rodopi.

Iser, W. 2006. How to do Theory. Malden, Oxford, Victoria: Wiley-Blackwell.

Jansen, S.C. 1991. Censorship: The Knot that Binds Power and Knowledge. Oxford: Oxford University Press.

Jansen, S.C. 2010. 'Ambiguities and Imperatives of Market Censorship: The Brief History of a Critical Concept' in Westminster Papers in Communication and Culture. London: University of Westminster, no. 7(2), pp. 12-30 [Online]. Available:

References

http://www.westminster.ac.uk/_data/assets/pdf_file/0010/79966/003WPCC-Vol7No2-Sue_Curry_Jansen.pdf [accessed 2013, May 30].

Landers, C. E. 2001. Literary Translation: a Practical Guide. Clevedon: Multilingual Matters Ltd.

Pérez, J. 1984. 'Functions of the Rhetoric of Silence in Contemporary Spanish Literature' in South Central Review, vol. 1(1/2), Spring - Summer, pp. 109-130.

Pym, A. 2006. 'On the Social and the Cultural in Translation Studies' in A. Pym, M. Shlesinger, Z. Jettmarová (eds.). Sociocultural Aspects of Translating and Interpreting. Amsterdam/Philadelphia: John Benjamins, pp. 1-25.

Santaemilia, J. 2008. 'The Translation of Sex-Related Language: The Danger(s) of SelfCensorship(s)' in TTR : traduction, terminologie, rédaction, no. 21 (2), pp. 221-252. Available: http://id.erudit.org/iderudit/037497ar [accessed 2013, June 6].

Somló Á. 2003-2004. 'Searching for Meaning' in K. E. Dubs (ed.). What does it Mean? Pázmány Papers in English and American Studies, 3. Piliscsaba: PPCU, pp. 313324.

Somló Á. 2010. 'The Role of Literary Translators in the Mediation of Ideas and Literature across Cultures' in T. Naaijkens (ed.). Event or Incident. Bern: Peter Lang, pp. 1211-41.

Sontag, S. 1983. A Susan Sontag Reader. New York: Vintage Books Random House.

Starkey, D. 2001. Elizabeth: Apprenticeship. London: Vintage Books.

Toury, G. 1995. Descriptive Translation Studies and Beyond. Amsterdam/Philadelphia: John Benjamins.

Venuti, L. 2009. 'Translation, Intertextuality, Interpretation' in Romance Studies, no. 27 (3), July, pp. 157-173.

Dictionaries used

http://www.merriam-webster.com/dictionary/censorship [accessed 2013, May 30