In recent years, a considerable volume of academic literature and researches in the field of translation are being focused on the concept of gender in translation (e.g. von Flotow 2001, Simon 1996, and Chamberlain 1998). According to Chamberlain (1998: 96), “the issues relating to gender in the practice of translation are myriad, varying widely according to the type of text being translated, the language involved, cultural practices and countless other factors”. Von Flotow (2001) offers a comprehensive overview of research areas in which the issue of “gender and translation” could be investigated:

Whereas most of researches done regarding gender in translation have dealt specifically with the issue of the translators’ gender identity and its effect on their translations, the main focus of current article is on how gender itself is translated and produced. Following paragraphs will try to clarify what gender is, how gender manifest itself in grammatical and social systems of language, and what problems translators encounter when translating or producing gender-related materials.

Grammatical Gender

Most linguists consider gender as a grouping of nouns into classes of masculine, feminine, and sometimes neuter such that the choice of a noun of a given class syntactically has an effect on the form of some other word or element of the sentence or discourse (such as articles, adjectives, and pronouns). According to Pauwels (2003: 557), languages with a “grammatical gender” system categorize nouns into gender classes on the basis of morphological or phonological features. However, while many believe that a grammatical gender system does not have connection with ‘extralinguistic category of sex’, Corbett (1991), the author of Cambridge textbook of Gender, acknowledges that grammatical gender system is not merely a morphological system, but it has also a semantic basis which becomes obvious, particularly, in gender assignment to human (agent) nouns, where most nouns referring to women are feminine, and those referring to men are masculine (p. 557).

From a historical point of view, Romaine (1999) explains how gender got into grammar. She states, “Linguists have traced the origins of grammatical gender in the Indo-European languages (which include the present-day European languages) to a system of noun classification based on similarities of sound”. The use of the terms ‘feminine’ and ‘masculine’, Romaine (1999) maintains, goes back to the 15^th century when Protagoras divided the two noun classes of Greek in groups tagged by them. She asserts that “the grammatical term is derived from the Latin genus, which meant race or kind and had nothing to do with sex” (p. 67). In the 19th century, she maintains, German grammarian Jakob Grimm spoke of the concept of grammatical gender as the metaphorical extension of ‘natural’ order of sex onto each and every object. Things named by masculine nouns are, in Grimm's opinion, earlier, larger, firmer, more inflexible, quicker, active, movable, and creative; those that were feminine were later, smaller, softer, quieter, suffering/passive, and receptive. Romaine (1999) believes that that Grimm's analysis shows a radical belief in male superiority.

In Romaine’s (1999) belief, the modern European languages probably inherited grammatical gender from a pattern of noun classification arising in ancient Indo-European, which originally grouped nouns according to phonological or sound-based principles which then developed into a grammatical system of syntactic concord or agreement. She claims, “Over time, however, these noun classes acquired a certain amount of semantic motivation by association with certain prominent nouns belonging to them. Thus, classes with a large number of nouns referring to female animates became associated with the female sex, whereas those containing a large number of nouns referring to male animates were associated with the male sex” (p. 69).

Van Berkum (1996) believes that grammatical gender assignment in different languages could be on the basis of one of the following characteristics of the noun: 1) semantics of the referent (e.g. Dyirbal); 2) phonology of the noun (e.g. French); 3) morphology of the noun (e.g. Russian); or 4) a combination of the above mentioned factors (e.g. German) (p.27).

Translation Problems Due to Grammatical Gender

Grammatical gender may cause translators some difficulties when they translate from source languages in which gender is differently grammaticalized compared with the target language. These difficulties may be particularly intensified when grammatical gender coincides with the sex of the referent; for example when the source language shows no gender distinction in the first-person pronoun but grammatical gender agreement patterns which may produce the effect of gendered self-reference through gender concord, and the target language shows not only no gender distinction in the first person pronoun, but also no grammatical gender agreement (McConnell-Ginet, 2003: 89).

Nissen (2002: 27), for instance, presents an example in which source language (Spanish) shows grammatical gender syntactically in a way unavailable to target language (English), so that, difficulties arise for the translator as to how to convey the information about the sex of the person in question. He explains that in the first line of the following poem the first person reference ‘hago’ (‘I do’), in theory, could refer to both a male or female person, but in the second line this ambiguity is resolved, because the predicate construction reveals the sex of the referent:

Nissen (2002) argues that in such a case, where target language (English) does not mark gender in predicate construction, then, the translator should resort to other means to convey the necessary information about the sex of the referent, so has done the translator in the following translation of above poem:

He notices that whereas the Spanish original focuses on 'I (type: woman) + cultured/well-traveled', the English translation focuses on 'I + woman (type: cultured/well-traveled)'. He argues:

According to Nissen (2002), similar problems may occur in many other cases, in fact, everywhere where the source language, by means of agreement structures, operates differently from the target language, which is in connection with noun-modifications, pronoun uses, pronominal references, and so forth.

Likewise, Romaine (1999) presents another example for difficulties that the grammatical gender may cause translators. She states that in Spanish and many other European languages it is not possible to say something such as “you are tired” without indicating the sex of the person spoken to and the relationship the speaker has to the addressee. She explains that to say ‘estas cansada’ means not simply ‘you are tired’, but that the addressee is female (compare masculine ‘cansado’) and the speaker knows her well enough to address her in the intimate second person singular form (compare the polite form ‘esta’). The different male and female endings ‘-al, -o’ are gender displays or indexes (p. 21).

According to Romaine (1999), comparing English and Spanish in this regard, we can say that Spanish speakers are obliged to make such distinctions of status and gender, taking into consideration the fact that they speak Spanish. These distinctions have been ‘grammaticalized,’ or made obligatory, in Spanish, whereas they have not in English.

Romaine (1999) claims that there is evidence for the existence of ideological factors which enter into gender assignment in systems that are supposedly purely formal and arbitrary as well as in systems where gender is supposedly determined by sex. She adds that the gender systems of both types of languages support a world view that is inherently gendered at the same time as they allow ideological construction of what is female as Other (p. 66). Consequently, as translators translate gender-related materials, they inexorably must face with the ideological load these materials carry with themselves as well as the problem of how to handle them.

Semantic Gender: Natural vs. Social Gender

Where grammatical gender is a category with syntactic consequences throughout the grammar, English is said to show “semantic gender”, i.e. the nouns English speakers refer to as she are assumed to possess a biologically [or socially] feminine semantic property in the real world (Romaine, 1999: 73).

The distinction between social and biological gender (sex) as two different, but, however, interdependent, semantic levels is one of the most crucial factors in the discussion of gender. These two semantic levels of gender are often inaccurately conflated with each other. Where (social) gender usually refers to a socially constructed system of classification that, regardless of external genitalia, attributes qualities of masculinity and femininity to people, sex (natural/biological gender) refers to physical and biological characteristics of a person based on their anatomy (external genitalia, chromosomes, and internal reproductive system). Shapiro (1981) describes the differences between social and biological gender in the following terms:

According to McElhinny (2003), the distinction between sex and gender is the antithesis of those socio-biological views that attribute differences and inequalities between women and men to sex or biology as a natural determinant of behaviors and roles. She believes that in such socio-biological views “there is no gender, for there are no cultural determinants of human life. All is ‘sex’” (p. 23).

Nevertheless, McElhinny (2003) asserts that those who make distinction between sex and gender do not necessarily deny the existence of some biological differences between men and women, but they sharply criticize the stereotypes attributed to these differences. The tacit idea behind the distinction between sex and gender, she explains, is that gender as a socially constructed entity can be more easily transformed than sex which is biological.

Yet, McElhinny (2003) admits that sex/gender models like Shapiro’s are problematic, both in their conception of gender and in their assumptions about sex, because to say that gender refers to the social, cultural, psychological constructs that are imposed upon these biological differences implies that there are two genders, based upon two sexes. Similarly, Litosseliti & Sunderland (2002) believe that “a simple distinction between ‘biological sex’ and ‘social’ or ‘socialized’ gender is now recognized as inadequate, if agency and diversity are to be properly acknowledged, and if, crucially, language is seen as shaping or constructing gender, not simply as a characteristic of it” (p. 5). Accordingly, they prefer Wodak’s (in some way post-structuralist) definition of gender as a multiple, fluctuating variable shaped in part by language. Wodak (1997) characterizes gender as the understanding of “how what it means to be a woman or to be a man changes from one generation to the next […] between different racialized, ethnic, and religious groups, as well as for members of different social classes” (cited in Litosseliti & Sunderland, 2002: 6).

Social Gender and Gender Stereotypes

The assignment of social gender is chiefly on the basis of a stereotypical classification. Cameron (1988) defines stereotyping as an act which involves a reductive tendency: to “stereotype someone is to interpret their behavior, personality and so on in terms of a set of common-sense attributions which are applied to whole groups (e.g. ‘Italians are excitable’; ‘Black people are good at sport’)” (cited in Talbot, 2003: 468).

According to Romaine (1999), “gender stereo-types are sets of beliefs about the attributes of men or women, such as that men are stronger and more aggressive, women are passive, talk more than men, and so on” (p. 4). Talbot (2003) claims that on the basis of a stereotypical gender assignment, “naturalized norms and expectations about verbal behavior are imposed upon people” whom are “perceived through a ‘lens’ of gender [bi]polarization” (p. 468). Gender stereotypes often “refer to prescriptions or unstated expectations of behavior, rather than specifically to representational practices” (Talbot, 2003: 472), and are often associated with other salient variables such as race, class, culture, age, context, and so forth.

Talbot (2003) admits that gender stereotypes and thereof social gender assignments are closely linked with and support gender ideologies. Societies commonly have norms regarding gender roles; i.e. how males and females should behave, expecting people to have personality characteristics and/or act in a certain way based on their biological sex. Talbot (2003) claims that if we consider gender stereotypes as ideological prescriptions for behavior, then actual individuals have to respond to the stereotypical roles expected of them both in constructing and communicating gender (p. 472). As a result, as Livia (2003) explains, in the process of translation, if the social expectations of gender in the target culture are very different from those of the source culture, translators who work both as interpreters of the original text and, often, as guides to the culture which produced the text have to deal with this anomaly; and if the languages encode gender in very different ways, they need to devise a system to encompass the differences. “In their dual role as linguistic interpreters and cultural guides,” Livia (2003) believes, “translators must decide what to naturalize, what to explain, and what to exoticize” (p. 154).

The Effects of Societal, Chronological and Contextual Factors

According to Cameron (2003), ideologies of language and gender are specific to their time and place: “they vary across cultures and historical periods, and they are inflected by representations of other social characteristics” (p. 452). Under influence of these socio-historical characteristics, (social) gender is now viewed as a fluctuating variable over time which could be placed within or between societies and cultures.

Societal and cultural factors play an important role in understanding the fluid and dynamic nature of social gender. As Romaine (1999) maintains, different cultures vary in their expectations about what it means to be a man or woman; therefore, they may have different systems of stereotypical classification for gender. She refers to the handbooks traditionally written by both men and women in western societies in which expectations of what it means to be a man or woman (in those societies, in a specific period of time) are expressed, and argues that these expectations may vary across different societies and cultures: “When used by different persons in different […] cultures,” Romaine (1999) explains, “the same linguistic features can, often mean very different things” (p. 5).

As mentioned earlier, the assignment of social gender is highly dependent on societal factors which are, however, subjected to change over time. Therefore, another important characteristic of social gender, as Nissen (2002) aptly notices, can be “its dependency on time” (p. 31). Referring to the occupational title of secretary, Nissen (2002) shows how the gender role associated with this title has been reversed over time as societal changes occurred: “It may surprise people today to learn that only one century ago this occupation was predominantly executed by men. In the 19th century, then, the social gender of secretary was 'male', i.e. the opposite of what it is today” (p. 31). He also refers to Lyons (1977), giving another example: “at the turn of this century [= 1900, UKN] in Britain the expression 'lady typist' was quite commonly employed in contexts (e.g. in advertisements) in which 'typist' would now be used” (cited in Nissen, 2002: 31). Nissen (2002) construes the quotation implying that the word ‘typist’ had a masculine connotation at the beginning of the 20^th century, “because ‘typist’ without any sex-specific modification referred to man alone [1] ” (p. 32). He holds that, over time, as a result of changes in social status, the social gender of the word ‘typist’ has changed, so that the necessity of marking the word with attributes of ‘lady’ or ‘female’ in order to employ female applicants is now obviated.

The last important feature of social gender discussed here is its ‘dependency on context’. The meanings of words, including allegedly gender-marked (sexist) words, are not fixed and vary from one context to another. According to Romaine (1999), “although language is central to our constructions of the meaning of gender, much of language is ambiguous and depends on context for its interpretation, a factor far more important than gender” (p. 5). She claims that gender differences in language are rarely, if ever, context-independent. Romaine (1999) holds that “the same words can take on different meanings and significance depending on who uses them in a particular context” (p. 5). She presents the sentence "How about meeting for a drink later, honey?" (My emphasis) as an example to show how context can change the meaning of words:

Translation Problems Due to Social Gender

As mentioned earlier, the assignment of social gender is based on a stereotypical basis which makes it dependent on socio-historical and contextual factors. As these factors may change from one place, society, culture, context, or time period to another, translators frequently encounter the complicated problem as to how to translate gender which has so huge potential of variability. Nissen’s (2002) examples [2] indicate how translators tackle the problem of gender translation, and how the decisions they make imply “ideological consideration” as well. He refers to a scene in Daphne du Maurier's novel ‘Rebecca’, as an example, in which chief characters, Maxim and his wife, have invited some relatives to their house in the England countryside. After dinner, Maxim’s brother-in-law expresses his admiration for the meal by saying:

According to Nissen (2002), there is no reference to the cook and his/her gender throughout the novel, so a translator who wishes to render the above sentence into a language which shows grammatical gender in a way that the gender of the cook must necessarily be determined, will face difficulties as to how to decide about the gender of the ‘cook’. Nissen (2002) demonstrates decisions made by different translators who translated the sentence into five different languages which show grammatical gender:

Nissen (2002) argues that the example indicates that three translators have assumed the social gender associated with ‘cook’ to be generally feminine, while the remaining two have assigned ‘generally male’ gender to it. He believes that the translators have made their decisions on the basis of their knowledge as to of what gender a “cook is more likely to be in a noble English manor,” or “their ideological expectations” as to of what gender a ‘cook’ is more likely to be “in their own community” (p. 32).

In another example, Nissen (2002) demonstrates how translators’ expectation of social gender varies in different translated versions of a single source text. His example is taken from Bernard Shaw’s Back to Methuselah:

Nissen (2002) reports translations as follows:

The example, as Nissen (2002) asserts, indicates discrepancy in translators’ expectation of social gender of a ‘secretary’ who shows a ‘flattering behavior’ [3] to his/her male boss: three of them imagined the flatterer to be a male and two decided the secretary was a female. He concludes, “As no clues are given in the text as to the sex of the referent, the translators have to make their choice in accordance with the knowledge they possess of the source community” [4] (p. 33).

Pronominal Gender and the Related Translation Problems

In languages that are said to have a pronominal gender system, “gender is marked solely on personal pronouns” (Corbett, 1991: 12). English has a pronominal gender system based on semantic criteria that is reflected only in personal possessive and reflexive third-person pronouns. The use of he, she and it is determined by simple principles: “male humans are masculine (he), female humans are feminine (she) and anything else is neuter (it)” (Ibid).

Translating pronouns between languages that encode gender differently in their pronoun systems has been always problematic: whereas some languages, like Persian, do not encode gender distinctions in their pronoun system at all; some others, like Shilha [5] , extend gender distinctions to almost all of their pronouns.

According to Livia (2003), “when translating from a language in which there are many linguistic gender markers into a language which has fewer, either gender information is lost, or it is overstated, overtly asserted where in the original it is more subtly presupposed” (p. 157). The problems Al-Qinai (2000) addresses in his example, on the contrary, are arisen in a quite reverse direction: translating from a source language manifesting less detailed gender distinctions in its pronominal system, compared to the target language.

Al-Qinai (2000) draws our attention to the problems that the translators may encounter in translating pronominal gender from English to Arabic (which shows a more detailed pronominal gender system) in the following advertisement:

He notices that the pronoun 'you' and 'your' are indiscriminately used in English regardless of number and gender; for example, in "For a lot less than you think", there is no marker to specify number (singular/ dual/ plural) or even gender. He comes to the conclusion that the translator has, therefore, to make a choice from one of the following options (Arabic 2nd person pronouns), when translating 'you' into Arabic:

Al-Qinai argues that the translator has to make a decision between the masculine and feminine pronouns and the gender agreements entailed thereof. But as the sex of the referent in the ST is not known, the unmarked masculine rather than the feminine form is used. In other words, the use of the ‘dominant’ masculine form does not rule out the possibility of feminine reference. Therefore, the TT translator opted for a masculine pronoun throughout the text (p. 515). However, Al-Qinai (2000) warns:

Since the pronominal shift in gender cannot be resolved in Arabic by a neutral pronoun, Al-Qinai (2000) suggests one of the following strategies to translators to adopt:

“With the failure of the above strategies,” he believes, “the loss of gender neutrality as a result of using the masculine second person singular as a dominant pronoun becomes inevitable” (p. 515).

As briefly depicted above, languages may differ greatly in the way they encode the category of gender in their lexical and grammatical systems. They may also differ in the expectations of their relevant cultures concerning what is meant by gender. We all know that every translation inevitably entails making a number of choices; moreover, there is a strategy behind every choice, and a reason behind every strategy. Little is known about possible choices, possible strategies, or possible reasons involving in the process of translation in situations in which there are linguistic or cultural gender discrepancies between the two languages involved; therefore, the study of “gender translation” seems to be an interesting area of research in Translation studies in the future.

References

Al-Qinai, J. (2000). Translation Quality Assessment: Strategies, Parameters and Procedures. META, XLV, 3, 497-519.

Chamberlain, L. (1998). Gender Metaphorics in Translation. In M. backer (Ed.) Encyclopedia of Ttranslation Studies (pp. 93-96). London: Routledge.

Corbett, G. (1991). Gender. Cambridge: Cambridge University Press.

Litosseliti, L. & Sunderland, J. (2002). Gender Identity and Discourse Analysis: Theoretical and Empirical Considerations. In L. Litosseliti & J. Sunderland (Ed.) Gender Identity and Discourse Analysis (pp. 1- ). Philadelphia: PA John Benjamins Publishing co.

Livia, A. (2003). “One Man in Two is a Woman”: Linguistic Approaches to Gender in Literary Texts. In J. Holmes & M. Meyerhoff (Ed.) The Handbook of Language and Gender. (pp. 142-158). Oxford: Blackwell Publishing.

McConnell-Ginet, S. (2003). “What’s in a Name?” Social Labeling and Gender Practices. In J. Holmes & M. Meyerhoff (Ed.) The Handbook of Language and Gender. (pp. 67-97). Oxford: Blackwell Publishing.

McElhinny, B. (2003). Theorizing Gender in Sociolinguistics and Linguistic Anthropology. In J. Holmes & M. Meyerhoff (Ed.) The Handbook of Language and Gender. (pp. 21-43). Oxford: Blackwell Publishing

Nissen, U. K. (2002). Aspects of Translating Gender. Linguistik Online, 11, 25-37. Retrieved September 2 from: http://www.linguistik-online.de/11_02/nissen.html

Pauwels, A. (2003). Linguistic Sexism and Feminist Linguistic Activism. In J. Holmes & M. Meyerhoff (Ed.) The Handbook of Language and Gender. (pp. 550-70). Oxford: Blackwell Publishing.

Romaine, S. (1999). Communicating Gender. London: Lawrence Erlbaum Associates, Publishers.

Simon, S. (1996). Gender and Translation. London: Routledge.

Talbot, M. (2003). Gender Stereotypes: Reproduction and challenge. In J. Holmes & M. Meyerhoff (Ed.) The Handbook of Language and Gender. (pp. 468-86). Oxford: Blackwell Publishing.

van Berkum, J. (1996). The Psycholinguistics of Grammatical Gender: Studies in Language Comprehension and Production. Doctoral Dissertation, Max Planck Institute for Psycholinguistics. Nijmegen, Netherlands: Nijmegen University Press.

Von Flotow, L. (2001). Gender in Translation: The Issues Go on. Retrieved September 2 from: http://www.orees.concordia.ca/numero2/essai/VonFlotow.html