Assessing translation quality

As with any field of study, there is a constant tension, in translation studies, between “grand” theory, that is, general theory, or just “theory” (highly abstract thinking about translation issues) and practice (resolving concrete translation problems). The question at the center of this tension is “To what extent are grand narratives of translation theory relevant?” The short answer to this question is: Grand narratives of translation (its general, highly abstract theories) are, of course, necessary, but their usefulness should not be overstated. Translation philosophy (an umbrella term that I use to designate the numerous general theories of translation) presents refined exercises for the mind and - as any philosophy - means “the love of wisdom,” in our case, concerning translation. However, the highly abstract nature of general translation theory (or theories) is not conducive to resolving the very specific problems translators have to face in real life.

A “Middle Range,” or “Operationalized,” Approach to Translating and Assessing Translations

Translation is a highly complex, cross-cultural, variance-prone activity, which is impossible to embrace with one neat theory. I advocate not one but a set of translation theories of the middle range. The concept of middle-range theories, as opposed to grand or general theories, was first developed by the American sociologist Robert Merton with reference to social thought1. I apply his general conception to problems of translation. In my analyses I am also guided by Isaiah Berlin’s idea that different aspects of beneficial phenomena or pursuits do not necessarily constitute a harmonious whole and may, in fact, come into conflict with one another2; in other words, pursuing one unquestionable good may well come into conflict with pursuing another, no less desirable good.

Robert K. Merton (1910-2003) was an outstanding American sociologist who is best known for developing several universally used concepts such as “unintended consequences,” “role model,” “self-fulfilling prophecy,” “reference group,” and others. Merton advocated “theories of the middle range,” which he defined as “theories intermediate to the minor working hypotheses evolved in abundance during the day-to-day routines of research, and the all-inclusive speculations comprising a master conceptual scheme from which it is hoped to derive a very large number of empirically observed uniformities.. ,”3.

What Merton said about social theory is, in my view, perfectly applicable to translation theory. In my scheme of thinking, a middle-range translation theory starts with some readily perceivable and accessible empirical phenomena, such as word senses that can be found in dictionaries and that can be contrasted with their contextual use in translations; different types of sentence structure that are described in grammar books and that can be contrasted with their actual theme -rheme structures in translated texts; and the pragmatic-stylistic parameters of text as described in the relevant literature that can be contrasted with their actual implementation in different translations.

Translations of works of verbal art are a very special kind of cross-cultural discourse carried out between pairs of specific languages (in our case - Russian and American English) through the medium of a translator who tries to make sense of the sociocultural “other” through the prism of his or her subjectivity. The translator has to re-create the “cross-cultural “other” in a context that is alien to and removed from the multifaceted “local other” in both time and space. This is a hugely complicated task that gets fulfilled with varying degrees of success, hence the variance in translations. In order to make translation discourse analysis practically significant, I operationalize its elements by examining the process of translation and its results from the perspectives of the semantics, syntax, and pragmatics of specific, existing texts.

At the level of the word, I single out nine concretely isolatable semantic parameters of word senses that need to be kept in mind while translating and analyzing the translation quality of other translations. These parameters are:

sense core, sense periphery, type of emotion, intensity of emotion, evaluative judgment, style, dialect, frequency of occurrence, and pragmatic reference involving different aspects of the cultural setting of the translation. A notional word in a specific text is uniquely contextualized. All non-predicative, contextually bound phrases or word combinations (that is, phrases that do not contain the main verb(s) - the predicate(s) - of a sentence or clause) are also contextually unique (ad hoc) word senses possessed of the denotative and connota-tive components indicated above. Thus, for example, the word combinations (phrases) “a latte-drinking, arugula-eating, wine-sipping, pacifistically-minded liberal” or “a knuckle-dragging, gun-toting, bible-thumping, Budweiser-swill-ing, America-the-best-minded conservative” are basically two self-contained “words” possessing complex structures of denotative and connotative elements linked by implied predication. [Possible translations of the two phrases: “либерал-пацифист - любитель кофе-латте, рукколы и вина” and “помесь вооруженного дебила, религиозного фанатика, любителя пива ‘Бадвайзер’ и шовиниста-консерватора.” These days the conflict between the two political stereotypes in the U.S. is particularly vitriolic. - Here and in the rest of the text, the author’s additional comments are given in brackets.]

The differences in the semantic composition of Russian and American lexical items and the ways these differences are negotiated in translation, are examined in detail in my book Translating Culture-1: Words⁴.

At the predication (sentence and clause) level, I distinguish a set of basic models or algorithms to reproduce different communicative structures of sentences and clauses in translating from Russian into English and vice versa. The central problem at the syntactic level of analysis is that the syntactic and semantic structures of sentences in Russian and in English often do not coincide. Syntax (the form of the sentence) and its communicative impact often collide. As a result, the word order in the translation has to be brought into line with the communicative structure (the theme-rheme relationship) of the original sentence viewed in the wider context of the paragraph. The nature of the problem can be illustrated by a brief example illustrating just one transformational model where the predicate of the Russian sentence serves as a rheme and is placed at the end of the sentence. Consider a simple sentence like the following: Какие-либо сомнения относительно благородства намерений руководителя американской делегации [subject - part of rheme] у российских участников переговоров [theme] отсутствуют [rheme]. In English translation this will be recast in a very different form - possibly like this: The Russian participants in the negotiations [subject - theme] don’t have any doubts [predicate - rheme] as to the noble motives of the leader of the American delegation [explication of rheme], [It should be noted in passing here that there has recently emerged a marked tendency, especially in news reporting in the Russian media, to imitate the word order (syntax structure) of English sentences, which reverses the positions of theme (topic) and rheme (message). For example, the wider context of the sentence “Вечер памяти жертв Холокоста пройдет в Москве” (January 24, 2017, indicates that the message (rheme) is in the fact that “Moscow will host the event,” and not in the fact that the event, of which the

reader is not expected to be aware, will take place in Moscow as opposed to any other place. In my view, this is a “generational” tendency that may mislead readers.]

In my book Translating Culture-2: Sentence and Paragraph Semantics⁵, I identify a set of typical Russian sentence structures that are not typical in English and provide a detailed examination of ways of negotiating those syntactic discrepancies in translating theme-and-rheme patterns within the framework of paragraphs.

My book “The Other" in Translation: A Case for Comparative Translation Studies⁶ focuses on the pragmatic aspects of “the other in translation”, which include (1) the sociocultural identity of the creator of the original, (2) the cross-cultural correlation of the two texts’ genres, (3) the cross-cultural correlation of the two texts’ functional styles, (4) the sociocultural and historical circumstances in which the author created the original text, including the perceived significance of the text with reference to the other texts created in the same period, (5) the distinguishing features of the authorial style comprising the uniqueness of the lexis, syntax, prosody (the rhythmic flow of the text), and the cultural specificity of the narrative, (6) the cultural and linguistic “remainder” and realia that get or do not get translated, (7) the unique socio-cultural imprint or “residue” of the personality of the translator in the translated text, and (8) the sociocultural identity and the expectations of the receivers of the translation, whether the translator targets a specific audience or not.

To make things manageable, I distinguish five clearly identifiable, if interpenetrating and overlapping, methods of or approaches to tackling “the other” in translating narrative fiction and films: neutralization, domestication, foreigniza-tion, contamination, and stylization. To a greater or lesser extent, all translators resort to these five strategies. The different pragmatics of translation resorted to by different translators can be illustrated by the following three different translations of the first sentence in J.D. Salinger’s The Catcher in the Rye;

ID. Salinger (1951)

The Catcher in the Rye

If you really want to hear about it, the first thing you’ll probably want to know is where I was born, and what my lousy childhood was like, and how my parents were occupied and all before they had me, and all that David Copperfield kind of crap, but I don’t feel like going into it, if you want to know the truth7.

Рита Райт-Ковалева (1960) Над пропастью во ржи

Если вам на самом деле хочется услышать эту историю, вы, наверно, прежде всего захотите узнать, где я родился, как провел свое дурацкое детство, что делали мои родители до моего рождения, словом, всю эту дэвид-копперфилдовскую муть. Но, по правде говоря, мне неохота в этом копаться8.

Максим Немцов (2008) Ловец на хлебном поле

Если по-честному охота слушать, для начала вам, наверно, подавай, где я родился и что за погань у меня творилась в детстве, чего предки делали и всяко-разно, пока не заимели меня, да прочую Дэвид-Копперфилдову херню, только не в жилу мне про все это трындеть, сказать вам правду9.

Яков Лотовский (2010)

Над пропастью во ржи

Если вы и в самом деле непрочь услышать обо всем об этом, вам сперва, наверно, захочется узнать из каких я мест, как прошло мое сопливое детство, род занятий моих родителей и прочую муру в духе Давида Копперфильда. Но, честно говоря, неохота в этом ковыряться10.

Translating Culture 1, Translating Culture 2, and “The Other’" in Translation concretize - or “operationalize” - specific aspects or elements of such general theoretical perspectives as the poetics of translation, polysystem theory, Skopos theory, etc. The books offer a complex of middle-range theories that can be empirically verified by applying them in the course of resolving similar translation problems in different stretches of language. The umbrella term that I use for my approach is Comparative Translation Discourse Analysis (CTDA). CTDA works excellently for teaching translation and, equally importantly, for assessing translation quality.

CTDA as a Qualitative Methodology

Quite undeservedly, qualitative methods of research and analysis have been given a bad reputation on the fallacious grounds that they are “not empirical enough,” i. e. they do not employ enough “measurements” or “metrics.” But this is an assumption that follows from a one-sided approach to research. I submit that the main reason for the widespread misconceptions about qualitative methods of research is that it is easier to set up an experiment in the natural sciences, which seek a syllogism-driven, single, replicable result, than to force the “crooked timber of humanity” - manifested, in our case, in the multiple possibilities of translating - into the Procrustean bed of the “hard” sciences. In the humanities, especially when analyzing the fluidity of sociolinguistic discourse (embedded in individual subjectivity), the rigid simplicity and materiality of hard metrics is often unrewarding and sometimes downright misleading. Of course, it feels much more comfortable to have definitive, hard answers than to accept the “unbearable lightness of being”. However, in translation studies, most standard quantitative/statistical techniques cannot be applied because of the nature of the object of study. When the set of items or data from which a statistically representative sample would normally be drawn is impossible to define and single out precisely, because it is too disparate, diffuse, small, or unique, the researcher resorts to case studies, focus groups and other instruments of research, involving a strong element of qualitative judgment. [In sociological research, the set of items or individuals from which a sample is drawn is called “population,” or “universe”.]

^a^r^s>

I advocate “comparative translation discourse analysis” in the form of case studies as a practically applicable and theoretically “substantiable” form of analysis in working out translation solutions and assessing the quality of existing translations. For the most part, I adhere to the principles of qualitative research as defined by Denzin and Lincoln in their Handbook of Qualitative Research" (1994) and Miles and Huberman in their Qualitative Data Analysis'² (1994). In view of the generally misunderstood and underrated status of the qualitative aspects of any kind of research, a somewhat extended summary of the recurrent elements in qualitative research and analysis is in order. I believe that the following principles are hard to dispense with in any kind of scholarly or scientific research, which in the final analysis is one and the same thing:

• “(1 ) Qualitative research is conducted through an intense and/or prolonged contact with a “field” of life situations. These situations are typically “banal” or normal ones, reflective of the everyday life of individuals, groups, societies, and organizations.

• (2)    The researcher’s role is to gain a “holistic” overview of the context under study: its logic, its arrangements, and its explicit and implicit rules.

• (3)    The researcher attempts to capture data on the perceptions of local actors “from the inside,” through a process of deep attentiveness, of empathetic understanding, and of suspending or “bracketing” preconceptions about the topics under discussion.

• (4)    <...> the researcher may isolate certain themes and expressions that can be reviewed with informants <.. .>.

• (5)    Amain task is to explicate the ways people in particular settings come to understand, account for, take action, and otherwise manage their day-to-day situations.

• (6)    Many interpretations of <.. .> material are possible, but some are more compelling for theoretical reasons or on grounds of internal consistency.

• (7)    Relatively little standardized instrumentation is used at the outset. The researcher is essentially the main “measurement device” in the study.

• (8)    Most analysis is done with words. The words can be assembled, sub-clustered, and broken into semiotic segments. They can be organized to permit the researcher to contrast, compare, analyze, and bestow patterns upon them”¹³.

In advocating and conducting Comparative Translation Discourse Analysis (CTDA), I proceed from the empirically incontrovertible fact that translations, whether written or oral, exist in a physical form and may serve as the object or subject of concrete research and analysis. But CTDA goes beyond textual analysis. It also factors in the sociocultural specificity of translators. Translators are not just highly skilled language manipulators but also cultural mediators. Their work creates and adjusts the cultures in which they operate. This influence is exerted through translated texts, which are linguistically codified experiences. Thus the primary objects of CTDA are original texts and their translations, including the perceived unique “linguo-cultural” imprint of the specific translator/s on the text/s of the translation.

Comparative Translation Discourse Analysis involves several interdependent operations: (1) establishing the perceived educational, sociocultural, entertainment, and aesthetic effects of the original text on its domestic audience;

• (2)    identifying the linguistic means with the help of which those effects are achieved in the original text; (3) determining the degree of approximation of the effects of the selected translation/s on their supposed audiences as compared to the effects of the original text on its domestic audiences; and (4) identifying the linguistic means that the specific translator/s used to achieve the established degree of approximation of the effects of the translation/s on the supposed cross-cultural audience in comparison with the perceived effects of the original of the translation in its source culture. All of these operations result in providing the translator or translation analyst with a tertium comparationis to be used as a basis for comparing original and translation. Put differently, the tertium comparationis of the translation analyst is constituted by his or her comparative perceptions of original and translation. These perceptions can be tested on native speakers coming from respective cultures and other translators working with the same pair of languages.

Translations of works of verbal art are a very special kind of cross-cultural discourse carried out between pairs of specific languages (in our case - Russian and American English) through the medium of a translator who tries to make sense of the sociocultural “other” through the prism of his or her subjectivity. The translator has to re-create the cross-cultural “other” in a context that is alien to and removed from the multifaceted local “other” in both time and space. This is a hugely complicated task that gets fulfilled with varying degrees of success, hence the variance in translations. In order to make translation discourse analysis practically significant, I operationalize its elements by examining the process of translation and its results from the perspectives of the semantics, syntax, and pragmatics of specific, existing texts.

Pseudo- and Genuine Translation Reviews

I use a middle-range comparative approach, outlined in the two previous sections, to counter a trend that manifests itself in ostensible reviews of translations of literary works, which are, in fact, reviews of literary works per se. Such “aesthetic reviews” usually make token references to the original text but fail to provide any substantive comparative analysis of original and translation. I call such one-sided aestheticism in reviews “pseudo-translation reviewing.” Lawrence Venuti14 calls it “belletrism”.

The absence of a serious comparative analysis of the artistic qualities of the original text written in one language and its representation as translation in another is harmful to the author/s of the translation because such an absence obfuscates the essential issues of translation and the fact that expert translation is a high art requiring years of deliberate practice. [I deal with the concepts “expert translation” and “deliberate practice” in Chapter 7 of What It Takes to be a Translator.^¹⁵

I will briefly consider just one example of this aesthetic trend in “pseudotranslation analysis.” It is a review of the most recent translation of Mikhail Lermontov’s classic A Hero of Our Time by Natasha Randall (2009)¹⁶. It was written by the famous English-American literary critic, essayist and novelist

James Wood and published in the prestigious U.S. magazine The New York Review of BooksT [I do not at all mean to belittle Wood’s accomplishments as a literary critic and cultural commentator. I am dealing here only with the way he comments about a translation of a literary masterpiece.] Not to put too fine a point on it, it is not so much a review of a translation as a review of Lermontov’s classic as if it had just appeared in the English language. It is as if the reviewer were using the occasion of the new translation in order to set forth some of his own views on this novel as well as introducing it to English-language readers for the first time. In a nutshell, Wood substitutes “aesthetic” literary criticism for comparative translation analysis. The lengthy article makes a cursory reference to Nabokov’s translation and contains no contrastive semantic, syntactic, and pragmatic analyses of either the original or the translation. What we get from James Wood by way of comments on the translation is just one brief passage - misleading and enigmatic to anyone familiar with Lermontov’s text and its several translations:

“Natasha Randall’s English, in her new translation, has exactly the right degree of loose velocity - this sounds like someone taking notes, patching it together as he goes along and unable to make up his mind. (Nabokov’s version, the best-known older translation, is a bit more demine than Randall’s, less savage.) So Pechorin, in this account, is both strongly male and slightly effeminate, bold and weak, fair and dark, finely dressed yet dusty from travel. On the one hand, the narrator is a confident 19,h-century analyst, conventionally reading the body as a moral map: a man who does not swing his arms is clearly secretive. On the other, he does not want us to set any store by such observations. He is also frank about his role as a maker who touches things up: he is obviously painting a romantic “portrait”1S.

All of that is well and good, but how does it relate to the actual original text in Russian? I personally have a number of specific questions about Wood’s comments, namely: What exactly is “loose velocity”? How does it manifest itself in Lermontov’s text? How would one measure it (given the current obsession with “metrics” in American academe)? How exactly is Randall’s translation less demure and more savage than that by Nabokov? I could go on with my questions, but my point must be clear: this is not a translation review, this is literary criticism, tenuously related to the Russian original - what Venuti calls “belletrism”.

In contrast to Wood, a radically different tack is taken by the American educator, translator, and translation critic Timothy Sergay. In his article “New but Hardly Improved”19, Sergay simultaneously engages four translations of A Hero of Our Time; those by Nabokov, Foote, Schwartz, and Randall. While admitting their general acceptability as translation variants, he points out some obvious (to a bilingual expert translator and translation critic) mistranslations or “gar-blings” (as Sergay prefers to put it) scattered throughout the four texts. Sergay concludes that “the cumulative effect of such ‘garblings’ for readers is almost certainly a misleading impression of mystery and incoherence. There appears to be greater ‘signal loss’ in recent ‘remasterings’ of Hero of Our Time than there was in the previous versions of this novel that the English-speaking world already had on its shelves. The errors are all understandable ones for Anglophone students of Russian, but they are not excusable in authoritative retranslations of a well-annotated classic”20.

To give the reader a taste of Sergay’s mode of translation analysis, here are two fragments of the table that he used in his paper to illustrate his arguments:

Onginal Russian	Nabokov, 1958	Foote, 1966, Rev. 2001	Schwartz, 2004	Randall, 2009
Будет и того, что болезнь указана, а как ее излечить -это уж бог знает!21	Suffice it that the disease has been pointed out; goodness knows how to cure it22.	Let it suffice that the malady has been diagnosed -heaven alone knows how to cine it!23	We will find that the disease has been diagnosed, but how one is to cure it -God only knows!24	[No translation]

The table above compares the translations of the last sentence in Lermontov’s introduction to the main body of his book. Sergay writes, “This final and crucial sentence in Lermontov’s introduction is entirely omitted in Natasha Randall’s translation. Her ‘Foreword’ simply ends on the preceding sentence, from which the words ‘i vashemu’ (‘and your misfortune as well’) were likewise omitted. This same sentence is garbled by Marian Schwartz, who did not recognize the construction ‘budet i logo, chto.. .’ as synonymous with ‘khvatit i logo, chto..? and construed it as ‘budetto, chto..?; ‘We will find that the disease has been diagnosed...”²⁵.

Onginal Russian	Nabokov, 1958	Foote, 1966, Rev. 2001	Schwartz, 2004	Randall, 2009
Хорошенькая княжна обернулась и одарила оратора долгим любопытным взором. Выражение этого взора было очень неопределенно, но не насмешливо, с чем я внутрейно от души его поздравил26.	t he pretty young princess turned her head and bestowed a long curious glance upon the orator. The expression of this glance was very indefinite, but it was not derisive, a fact on which I inwardly congratulated him with all my heart27.	t he pretty young princess turned and bestowed a long, curious look on the speechmaker. The feeling conveyed in her look was very hard to define, but it wasn’t scorn — on which I felt Grushnitsky was to be warmly congratulated28.	The pretty young princess turned around and bestowed upon the orator a long, curious gaze. The expression of this gaze was rather vague, but amused, for which I privately congratulated her in all sincerity29.	The pretty princess turned around and gifted the orator with a long and curious gaze. The expression of this gaze was very ambiguous but not mocking, for which I applauded her from my innermost soul30 (the italics in final columns are mine)

With regard to this second table above, Sergay’s analysis goes as follows:

“To my ear, the construction “to gift somebody with something” sounds like strictly legal language for the conveyance of funds and property. Here I appear to be a “conservative” in Paul Brian’s book Common Errors in English Usage: ‘Conservatives are annoyed by the use of “gift” as a verb. If the ad says “gift her with jewelry this Valentine’s Day,” she might prefer that you give it to her.’ [See also gift/html.] The Russian constniction ‘darit’/odarit’ kogo chem' as opposed to ‘darit’/ podarit ’komu chto' is archaic to begin with (Evgen’eva, Malyi akademicheskii slovar ’); Randall’s predecessors did well to handle it with the elevated ‘bestow upon? while Randall’s choice of ‘gifted the orator’ creates unwelcome associations with today’s legal and advertising jargon. Far worse, however, is the complete garbling of the sense of the final clause of this sentence. Why should Pechorin applaud or congratulate the princess for the momentary seriousness of her regard for his rival, Grushnitsky, whom he holds in polite contempt at this point? Mistaking ego, ‘him,’ for её, ‘her,’ is of course an elementary error. Again, this same error is committed by Marian Schwartz, who also misconstrued ‘no ne nasmeshlivo,’ ‘but not mocking,’ <...> and translated it as ‘but amused’ ”31.

I believe that it is such minute dissections - or “deconstructions” - of the text at different levels of analysis (semantic, syntactic, and pragmatic) that should form the foundation for an informed assessment as to how a translated text works with reference to its original. Wood’s review evinces what Venuti calls a “belletristic approach” to translation criticism, whereby the critic “assigns it [a translation] an aesthetic autonomy from the source text and judges it not according to a concept of equivalence, but according to the ‘standards’ by which he judges original compositions. <.. .> This approach [is] belletristic because it emphasizes aesthetic qualities of the translated text itself. It is also impressionistic in the sense that it is vague or ill-defined”32. In other words, there is no genuine comparative assessment of the quality of a translation. Sergay’s approach is an example of a “comparative translation variance analysis”33, which ought to form the basis of what I call “comparative translation discourse analysis,” this latter concept being a more comprehensive one, owing to its so-ciolinguistic and sociocultural dimensions.

Conclusion

Given the complexity and required erudition involved in a literary translator’s work, the expert translator is much more than “just a translator.” [I discuss the difference between the concepts of professionalism and expertise in translation in chapter 7 of What It Takes to be a Translator.\ From the perspective of the Comparative Translation Discourse Analysis (CTDA) that I advocate as an instrument for assessing translation quality, I would risk arguing that the expert translator is, in fact, a co-author of a work of fiction and an indispensable coparticipant in cross-cultural communication. I am, therefore, inclined to agree with the editor-in-chief of the Russian translators’ Mosty journal, Viktor Lan- chikov who, in one of his articles, has this to say about the qualities of a true high-art practicing literary translator:

The frustration and joy that the translator experiences are no weaker than the exact same feelings familiar to any creative person. Verbal inventiveness, stylistic flexibility, and a special acuity in distinguishing shades of meaning - all of these qualities are no less (and, perhaps, are even more) important in literary translation than in authorial creativity”34.

This article is based on chapters 4, 5, 6 and 10 of my book What It Takes to be a Translator: Theory and Practice (Saarbriicken, Germany: LAP LAMBERT Academic Publishing, 2014).