A QUANTITATIVE ANALYSIS OF THE ROMANIAN TRANSLATIONS OF SHAKESPEARE’S BAWDY PUNS

This article proposes a quantitative analysis of the Romanian translations of 325 ribald Shakespearean puns, which originate in 20 plays and 71 renditions, with special focus on assessing the impact of translator-subjective and objective factors on the rendition process in the pre-communist, communist, and post-communist periods. The findings invalidate several widespread beliefs: Dragoș Protopopescu’s renditions, banned by the communist regime for their ‘modernizing’ approach to the Shakespearean text, bowdlerized more bawdy puns than ‘ESPLA’, which replaced it as the Party-approved Romanian edition of the dramatist’s plays; Adolphe Stern’s translations, harshly criticized in his period, fare better in terms of ribald pun rendition than Scarlat Ghica’s and Dimitrie Ghica’s, hailed as the most successful of their time; modern translations of Shakespeare display a heterogeneous distribution of target-text puns across the surveyed rendition strategies, despite enjoying similar availability of and access to pun translation studies.

studies such as Magdalena Krawiec's (2017) analysis of the frequency of pun rendition strategies in the Polish translations of Love's Labour's Lost, the quantitative approach to the playwright's puns and ribald language-play in particular is hardly a popular avenue of research among scholars of Shakespeare translation, with the relevance-theoretic approach proposed by Francisco Díaz-Pérez (2013) constituting the most widely used framework in Shakespearean translation studies.
Recently, Eriko Sato (2019) has successfully deployed a translanguaging-based model, yet in her case, like in Krawiec's and Díaz-Pérez's, the corpus is limited to a low number of puns and translations.
One important factor is responsible for the relative absence of quantitative approaches to this aspect of Shakespeare's language: the lack of a methodological framework developed to this end. This article proposes a quantitative analysis of the translation of 325 unanimously accepted ribald puns originating in 20 Shakespearean plays and 71 Romanian renditions, with special focus on assessing the impact of translator-subjective and objective factors on the translation of the dramatist's bawdy wordplay in the pre-communist, communist, and post-communist eras. In this respect, I will use a modified wordplay rendition methodology, which draws on Dirk Delabastita's (1993) competence model and Díaz-Pérez's cognitivepragmatic reworking (2015), and associates the various approaches to pun translation with specific rendition strategies.
It is only rarely and of fairly recent date that quantitative analyses of Shakespeare's bawdy wordplay have been attempted in Romania. In 2003, Romanian Shakespeare scholar and translator George Volceanov delivered a paper in Utrecht, revealing the findings of his comparative analysis of 306 ribald words and phrases singled out arbitrarily from 29 plays and two poems translated into Romanian during the communist regime. He counted 179 instances of meaning-formeaning translation, 93 cases of lost bawdy connotations, and 34 situations in which the translators appear to have surpassed the ingenuity of the source text (Volceanov,"Bowdlerizing" 120). The study per se has not been published of late, but its results and the critical materials deployed in the compilation of the corpus have been mentioned in several of Volceanov's articles (2006;. There, the scholar indicates that he selected the bawdy expressions based on Eric Partridge's Shakespeare's Bawdy (1947). Yet this glossary of Shakespearean ribaldry brings together non-punning ribald words and bawdy puns.
Since the previously mentioned articles do not provide a detailed account of the corpus, it is difficult to say whether Volceanov's inquiry includes language-play or is limited to studying bawdy words. A set of legitimate questions then arise: if his analysis includes puns, what does sense-for-sense imply in that case? If some words lose their ribald connotations, does this indicate that those instances of bawdy constitute plays on two meanings, ribald and non-ribald, of the same word? Indeed, the results of his statistical inquiry are provocative: according to Volceanov, 70% of the surveyed sample was rendered effectively into Romanian, yet to what extent is this the outcome of a methodology-based research or the empiric estimations of a Shakespeare translator? My thesis is that only a survey based on a quantitative studies-oriented framework can answer this question.

Considerations on the methodology
In consulting Frankie Rubinstein's 1984 Dictionary of Shakespeare's Sexual Puns and Their Significance, the latest of this kind to be printed to date, by far the lengthiest, and theoretically the most suitable to my purpose, "one has to ask oneself who is making the pun: the author (through the character), or the interpreter" (Wells 27

A quantitative analysis of bawdy wordplay translations in Hamlet
In the pursuit of providing a statistical perspective of bawdy wordplay translation in Shakespeare's works, I have opted to chart the phenomenon from the particular to the general. In so doing, the play that can, in theory, offer the most relevant insight in this respect is Hamlet, perhaps one of the most recognizable Shakespearean plays and undoubtedly his most frequently translated into Romanian. The corpus of puns identified in this play amount to 12 instances, whose dispersal across the six rendition strategies is displayed in Figure 1. Although separated by more than a quarter of a century, Stern's and Anestin's renditions do not appear to indicate major shifts in translational approach to the playwright's ribald puns, which is indicative of their belonging to the same translational mentality/system, which extends from the earliest Shakespearean translations to World War I. Indeed, Stern's rendition features a higher incidence of bawdy puns translated effectively both in terms of language-play and suggestiveness [see P > P (B)]. This may be due to subjective factors such as, perhaps, his lack of prudishness; take, for instance, Guildenstern's pun on 'privates', which plays on 'soldiers' and 'genitals' (Ham. 2.2.244). Anestin translates it as 'protejați' ('protégés') (88), while Stern opts for 'intimi' ('confidants') (90). As for Hamlet's ensuing comparison between Faith and a strumpet (Ham. 2.2.246), which the former renders euphemistically as 'femeie publică' ('public woman') and the latter bluntly as 'târfă' ('whore'), it is only in Stern's translation that the bawdy subtext of 'privates' is recreated through association with the adjectival meaning of 'intim'. It should, however, be noted that the two translations exhibit the same number of puns acknowledged by the two translators [see P > P (B) and P > N-P (B)]. At a macro level, a lack of ideological or ethical interference in the translation process is observed, as neither one of the puns submitted for analysis is expurgated from the target text.
Compared to the two previous renditions, Dragoș Protopopescu's is a special case in that no less than four puns of the total 12 instances of wordplay are expurgated from the target text [see P > ZERO]. His rendition is missing, among others, Act 3 scene 2's lines 246-255, a textual fragment containing no less than three specimens of bawdy language-play between Hamlet and Ophelia. Hamlet's line 121 -"That's a fair thought to lie between a maid's legs" -, whereby his pun on 'to lie' is effected, disappears from his translation, although Protopopescu's 'a se culca' ('to sleep') can potentially recreate it by itself (103). This omission, like in Anestin's case, may be attributed to the translator's relying on the First Quarto of the play, which was then commonly thought to be a memorial reconstruction of the text by a spectator of a Hamlet performance (Erne 199). Should this theory be found correct by further research, the question still remains as to the reason why the Q2 [Second Quarto] and F [Folio]'s Act 4 scene 7 appears two scenes earlier in his translation and why it does not feature said missing lines since they occur in the First Quarto, though, admittedly, not in their entirety. Therefore, as compared to Stern and Anestin, who, at the very least, rendered the wordplay non-punningly and nonbawdily, Protopopescu goes on to eliminate them and their textual fragments. This, I believe, is due to both subjective and objective factors: on the one hand, there was Nicolae Iorga's hunt for pornography in literature and on the other, Protopopescu's collaboration with royalty-owned publishing houses and his far-right political leanings against the onset of the communist regime, which led not only to his untimely demise and subsequent ban from public libraries, but also to a self-inflicted interventionist approach to the Shakespearean text.
There is reason to suspect that either self-censorship or some coercive action on the part of a regulatory body also affected other later renditions of the play.
Translators and essayists Vera Călin and Maria Banuș published the first translation of Hamlet following the imposition of the socialist regime. Despite the heavily supervised backdrop against which it was produced, their rendition appears not to have suffered as much as Protopopescu's. Only one of the 12 instances of wordplay submitted for analysis disappears from their translation [see P > ZERO]. It should, however, be noted that Călin and Banuș's rendition exhibits no intention to preserve the bawdy substratum in the cases where the language-play could not be reproduced.
In contrast, Ion Vinea appears to have gone to greater lengths to preserve not only the ribald undertones but also the language-play of the puns subjected to investigation. This achievement is all the more remarkable as his translation was published as part of the 1955-1963 Shakespeare edition, intended to take "a safe distance from Protopopescu's lively, modernizing versions," which had been banned only nine years prior within an extensive campaign to "establish 'the [party] line' . . . that all critics and performers of Shakespeare were supposed to toe, unless they were willing to risk censorship" (Nicolaescu 286). The circumstances under which Vinea's Hamlet appeared were, however, significantly more complicated. At that time, the poet was no longer allowed to publish his works due to his reluctance to join the Party and in the summer of 1959, when his rendition was issued, he would be arrested for selling some two hundred gold coins to his friend Petru Dumitriu seven years earlier (Cordoș 178  Gheorghiu 's earlier 1955-1963 project and, in some cases, puts forward re-renditions of popular plays. Produced one year before the official debut of the 'liberalization' (Terian 2013), their individually printed Hamlet "received almost no attention when released," in spite of its comprehensive critical apparatus and astute paratexts (Cinpoeș 97), which "displayed an openness towards foreign criticism" (Milică 34).
In 1982, when Levițchi's 'Univers' edition debuts, the translator goes on to consolidate this new paradigm of Shakespeare, whereby he is envisioned in humanistic terms as "a dramatist-poet-thinker" (Volceanov,"Bawdy" 218). 32 This is, as Milică notes, a clear indication that the "turn towards openness starting in the middle of the 1960s is more obvious in the 1970s and 1980s" (33). This 'prolonged liberalization' and subsequent increased access to more recent critical materials (34) appears to have also positively impacted Duțescu and Levițchi's approach to bawdy puns (Martin 2018

Shakespeare editions
In following the research trajectory set out in the previous subsection, I will now proceed to expanding the quantitative study to include other renditions that may help to substantiate or provide further nuance to the link between translation outcomes and their socio-political background. To eliminate any proportional disadvantage between the three Romanian Shakespeare projects, only those plays that were rendered thrice, once for each complete works collection, were submitted    The three other translators stand at similar percentage levels, namely 6.45% and 6.78% respectively.
With regard to the potentially expurgatory PUN > ZERO, Stern fares better than the Ghicas and Protopopescu. Specifically, Protopopescu resorts to seven omissions, while the Ghicas and Stern to five and two respectively. The fact that Stern, a Jewish-Romanian lawyer, recreated roughly as many ribald puns as Protopopescu, a Professor of English, did almost sixty years later, is indicative of the former's familiarity with Shakespeare's ribald puns and vocabulary, which then critics overlook due to his Jewish origins and lack of formal training, while the higher frequency of PUN > ZERO renditions in the latter's translations points once again to the existence of negative translator-subjective and objective influences in his case.

The rendition of Shakespeare's bawdy wordplay in the communist period
In the quantitative analysis of Hamlet, it has been shown how Levițchi and Duțescu's re-translation of Hamlet fares better than Vinea's in terms of successful bawdy pun translation, while the latter surpasses the co-translators' cumulatively, when the instances of wordplay translated via the PUN > NON-PUN BAWDY, and OTHER strategies are also taken into consideration. The close competition between the two renditions of this Shakespearean play has raised the question of whether the retranslations of other dramatic texts hold the key to ascertaining which of the two communist editions of the dramatist's works maintains a higher level of faithfulness to the Bard's bawdy puns. Since 'ESPLA' and 'Univers' overlap in many cases, a chart was generated based on those plays that were re-rendered for the latter project. What  The difference between the two is not especially pronounced in terms of full or partial reproduction of ribald wordplay, a phenomenon also witnessed in the case of their translation of bawdy Hamletian puns: 'ESPLA' comprises 23 instances of language-play effectively recreated in the target language, while 'Univers' contains 20 such specimens. As for the other translation methods, whereby the bawdy only is reproduced, the latter's re-translations amass 15 cases in which the PUN > NON-PUN BAWDY technique was deployed, whereas the 'ESPLA' renditions amount to 12 such instances.
As for the compensational translation strategies, 'Univers' appears to lean more toward suggesting the existence of a secondary ribald meaning, which is evident from its higher incidence puns rendered via OTHER strategies.  These findings indicate that a comparable level of access to and availability of studies devoted to the topic does not automatically lead to uniform translation outcomes in the case of translators operating at a similar point in time. In fact, it would not be too far-fetched to say that they confirm a precaution taken before carrying out this study: the more dictionaries and glossaries are released on the topic, the more difficult it is to distinguish between source-text puns of Shakespeare's making or the lexicographer's.

Conclusions
Some of the observations that follow from these graphs are only natural: there is an upward trend toward faithfulness to Shakespeare's language-play and/or bawdy that correlates with the availability of and access to studies dedicated to this instance of Shakespearean language use. Other findings invalidate commonly held beliefs among Shakespeare scholars: Protopopescu's edition of Shakespeare's plays, presumed to be 'modernizing' and therefore one of the most successful in retaining, at the very least, a positive bawdy balance, contains the highest number of cases of expurgation, while the 'ESPLA' collection, produced during the most vigilant period of the communist censorship program, exhibits a surprisingly large incidence of puns translated bawdily into Romanian. The former phenomenon is posited to be due to the hunt for pornography in literature led by Romanian academician and scholar Nicolae Iorga, while the latter could be attributed to the lack of censorship intervention (Volceanov 120