Number 74 January, 1997
Dept. of Asian Languages and Cultures
University of Michigan
Western linguists have been studying various aspects of sexism in the English language for several decades. Following on the heels of influential works by Robin Lakoff (1975), and Miller & Swift (1977), and others, a host of researchers began to identify and catalog types of sexism manifested in features of language such as vocabulary, grammar, discourse, and even intonation. (For a good historical overview of sexism in English, see Baron, 1986. Other recent research includes Stanley, 1977a, 1977b; Spender. 1980; Graddol & Swann, 1989; Eakins & Eakins, 1978; Francine & Anshen, 1983; Hill, 1986; Sorrels, 1983, and others.) Recently scholars have turned their attention to sexism in Mandarin Chinese as well, finding strikingly similar sexist phenomena in linguistic aspects such as Chinese characters, vocabulary, idioms, and proverbs. (See Shih, 1984; Wang, 1986; Wu, 1991; Liu, 1991 ; Jung-Palandri, 1991; Chen, 1995: Moser & Chen, 1995, just to name a few.) The most notable treatment of sexism in Chinese is that of Farris (1988), whose work has greatly influenced this article.
Like other cultures, China has a long history of sexist social conventions, and the Chinese language is pervaded with evidence of these. Research in this area has usually sought to identify and catalog aspects of Chinese that embody these sexist cultural traditions, such as sexist idioms, demeaning words for wife, derogatory terms of address for women. or the large number of characters containing the female radical (女) with negative connotations. Such elements tend to be rather easily identifiable and have been some of the earliest aspects to be targeted for linguistic reform. (The Chinese Communist Party, for example, in their attempts to elevate the status of women and eradicate vestiges of feudalism, has from time to time officially discouraged use of pejorative terms of address for women and wives.) Notable contributions have already been made in such research, but there are certain kinds of sexism in the Chinese language that are more subtly embedded in the grammar in such a way that they often escape conscious attention. This article attempts to shed light on some of these phenomena, since it is often in these hidden patterns of linguistic usage that sexist assumptions and notions are most powerfully present.
Overt vs. covert linguistic categories
In order to highlight and study these implicit sexist usages, it will be useful to borrow Whorf’s (1956) distinction between overt and covert linguistic categories. Overt categories (Whorf also uses the term phenotypes) are categories explicitly marked in the language: covert categories (also called cryptotypes) are meanings not coded explicitly in grammar or lexicon, but rather emergent only in syntax and various usage patterns in the language. Farris (1988) was perhaps the first to raise the topic of covert sexist categories in Chinese, noting ways in which the language implicitly relegates women to a kind of “negative semantic space.” This notion is a constant theme throughout various aspects of sexist language in both Chinese and English, and we will deal with it in some detail below.
Examples of overt categories in English are plural forms (which are explicitly marked with either -s, -es, or a vowel change). and past tense verbs, which are marked in various ways according to verb type. An example of a covert category is the class of intransitive verbs (such as go, gleam, sleep, arrive, appear, etc.). These verbs are on the surface indistinguishable from transitive verbs like cook, kill, find, or see, until it is noted that they are not interchangeable in sentence structures like “It was V’ed” (“It was cooked”, “It was killed, “It was found”, “It was seen”, etc., where the intransitive class fails: *“It was goed”, *”It was gleamed”, *”It was slept”, *”It was appeared” etc.). In other words, the difference is not evident from surface level aspects or grammatical markings of the verbs themselves; it is only through contextual patterns of usage that the two categories emerge (the transitive category taking objects, the intransitive category not). Another example of a cryptotype is the set of transitive verbs that in. some way involve covering, enclosing, or surface attachment, to which the prefix -un may be attached to denote its opposite. Thus we have a class of items like uncover, uncoil, undress, unfasten, unlock, untangle, untie, unfold, but not unthink, unmake, unjump, unmelt, unthrow, etc. As we shall see below, there are many linguistic categories governing male and female terms and meanings which show sexist patterns that are not necessarily psychologically evident to speakers, but nevertheless have a powerful reinforcing effect, in that they fit so seamlessly into the prevailing worldview that they pass unnoticed as “common sense”. Linguistically, such structures come to be considered as merely “correct” or “standard”. A particular word or idiom perceived as sexist can be rather easily isolated and gradually weeded out of the language, but sexism more subtly encoded in patterns of usage is not as easily noticed, much less consciously avoided. Not all of the subjects examined below constitute covert categories per se, but all are examples of a kind of sexist bias embedded in the very warp and woof of the language, and which lie at the boundaries of sociolinguistics, psycholinguistics, and cognitive psychology.
Gender asymmetry of dyads
Lakoff & Johnson (1980) and other linguists have noted that oppositional and collocational dyads in English (and perhaps most languages) seem to adhere to a pattern in which the first component of the pair can be characterized as semantically “plus” and the second component “minus”. The pattern can take various forms: positive-negative, primary-secondary, prior-subsequent, cause-effect, and so on. Some examples are: good and bad, active and passive, more or less, up and down, joys and sorrows, light and dark, front and back, large and small, tall and short, love and hate, right and wrong, Heaven and Hell; and so on. This pattern is found in Chinese as well: hǎo huài 好坏 (“good-bad”), shì fēi 是非 (“right-wrong”), shì fǒu 是否 (“is-not; whether or not”), duì cuò 对错(“correct-incorrect”), shàng xià 上下 (“up-down”), qián hòu 前后 (“front-back”), dàxiǎo 大小 ( “big-small; size”), duō shǎo 多少 (“ many-few”), lǎo shào 老少 (“old-young”), gāo dī 高低 (“high-low”), cháng duǎn 长短 (“long-short”), kuān zhǎi 宽窄 (“wide-narrow”), yuǎn jìn 远近 (“far-near; distance”), ài hèn 爱恨 (“love-hate”), róng rǔ 荣辱 (“glory-shame”), xǐ āi 喜哀(“ happiness-sorrow”), etc. This ordering pattern of collocational and oppositional pairs in fact is observed in most languages, which suggests it is not merely a matter of convention, but very likely reflects basic aspects of human cognition. Note that the first member of such pairs very often functions as a generic, or primary category term. For example, the word tall in the dyad tall-short functions as the generic term in assessments of height. That is, when one says “How tall is Jane?”, there is no assumption that Jane is tall; the question is merely a neutral one concerning her height. But if one says “How short is Jane?”, the assumption is that Jane is short. And so with questions such as “How old is Jane?”, “How fast is her car?”, “How heavy is her suitcase?’ etc., where in each case the primary member of the pair is used to signal neutral quantification.
Such usage patterns are exactly isomorphic to the Chinese case. Lǎo Zhāng yǒu duō gāo ? 老张有多高？(“How tall is Lao Zhang?”) is merely an inquiry as to Lao Zhang’s height, whereas Lǎo Zhāng yǒu duō ǎi? 老张有多矮 ? (“How short is Lao Zhang?”) assumes Lao Zhang to be short. And so with other quantitative generics: duō yuǎn? 多远? (“how far?”), duō cháng? 多长? (“how long?”), etc. Compounds expressing abstract gradable qualities use the primary component of the dyad for their expression: chángdú 长度 (“length”), kuāndú 宽度 (“width”), gāodú 高度 (“height”), and so on. Another example of this positive-negative ordering pattern, one that is so obvious and pervasive as to perhaps be not worth mentioning, is one of the standard question patterns in Chinese, V 不 V?, as seen in tag questions such as Hǎo bù hǎo? 好不好？ (“Good, not-good?”), Duì bù duì? 对不对 ? (“Correct, not-correct?’), Shì bù shì? 是不是 ? (“Is, not-is?”), and so on. It is seems natural for human categorical logic to place the positive first, and the negated, modified, or special case second?
Given the longstanding sexist traditions in Western and Chinese culture, it is not surprising that such asymmetries are also evident in dyads involving gender pairs. Some examples in English are: man and woman, he and she, boys and girls, brothers and sisters, husband and wife, sons and daughters, male and female, guys and dolls, Mr. and Mrs., etc. Names of couples or pairs with male and female members also tend to follow this order: Jack and Jill, Romeo and Juliet, Hansel and Gretel, Anthony and Cleopatra, Samson and Delilah, Sonny and Cher, Roy Rogers and Dale Evans, Ike and Tina Turner, Burns and Allen, and so on. Through the nearly invariant male-female ordering, it can be seen that the male role overall is seen as primary, as normative, as in some sense defining the adjunct female role.
Examples also abound in Chinese, where the male-female ordering is, if anything, more invariant. The following are just a few examples of male-female dyads in Chinese: nánnǚ 男女 (“man-woman”), fūqī 夫妻 (“husband-wife”), fūfù 夫妇 (“husband-wife”), fù mǔ 父母 (“father-mother”) bàba māma 爸爸妈妈 (“father-mother”), xiōngdì jiěmèi 兄弟姐妹 (“brothers-sisters”), érnǜ 儿女 or zǐnǜ 子女 (both meaning “sons-daughters”), qiánkūn 乾坤 ((“male-female”, “heaven-earth”), lōngfèng 龙凤 (“dragon-phoenix” [symbolizing male and female, respectively]), etc. Of course, there are occasional exceptions to this convention in Chinese, such as yīn yáng 阴阳 (“female principle, male principle”) but such exceptions (which sometimes equally reflect a sexist basis, as well) are rare. Not surprisingly, famous couples in Chinese history and literature are also conventionally listed in male-female order. Examples are: Liang Shanbo 梁山伯 and Zhu Yingtai 祝英台, Jia Baoyu 贾宝玉 and Lin Daiyu 林黛玉, Xu Xian 许仙 and Bai Suzhen 白素贞, etc.
The important thing to note is that it is very difficult to make the case that this virtually invariant asymmetrical pattern is in any way a “natural” ordering (as might be the case with the other dyads above). There is simply no compelling logical reason why the female should occupy the secondary or negative semantic slot in the dyad; it is instead a deeply ingrained linguistic convention reflecting deep-seated underlying cultural assumptions. Given that women have been conceptually viewed by most societies as less important, less typical, or even “deviant”, it then becomes “natural” for native speakers to accord them second place in such dyads, on the cognitive model of pairs like up-down, good-bad, etc. This state of affairs goes unnoticed because it is in keeping with a kind of unspoken “common sense” that is absorbed and accepted (by both males and females) very early on in childhood during the process of language acquisition. Presumably, in a hypothetical non-sexist world such gendered dyads would merely be randomly in either order, or, if there were conventional orderings, they would exhibit no preference for male over female. English, like all languages, has a large set of collocational pairs with relatively fixed orders; we speak of knives and forks, dogs and cats (except when it rains), pens and pencils, meat and potatoes, tables and chairs, fruits and vegetables, and so on, but it is obviously difficult to maintain that these orderings reflect anything but arbitrary convention.
The male-female ordering convention extends to various stock phrases and idioms involving both genders, as evidenced in examples such as láng caì nǚ mào 郎才女貌 (“The man is talented, the woman beautiful”), fū chàng fù sui 夫唱妇随 (“The husband sings, the wife accompanies”) etc. Indeed there was even an entire genre of pre-modern fiction called cáizi jiāren xiǎoshuo 才子佳人小说, novels about “talented young men beautiful women” that again reinforces this gender ordering.
Thus the entire set of female members of such pairs constitutes a kind of covert category of dispreferred, de-emphasized, or marginalized items. As we shall see, this state of affairs manifests itself again and again in the usage patterns of Chinese grammar.
Asymmetric gender defaults of rén 人 (“person”)
A problem noted in almost all the literature on sexist language in English is the use of -man as a suffix in words such as businessman, congressman, chairman, mailman, etc., as well as the use of man as a generic for “humankind” or “person”, as in “Man is a tool-using animal”, “manned spacecraft”, “one-man show”, “manpower” etc. Some Western scholars familiar with Chinese have observed that the problem of such masculine defaults so common in the English lexicon historically do not arise in Chinese due to the use of the gender-neutral word rén 人 (“person”) in compounds. For example, a gender-marked English word like businessman becomes shāngrén 商人 (“business-person”); Englishman becomes Yīngguó rén 英国人 (“English person”), etc. And a sentence such as “Man is a tool-using animal” becomes Rén shì yòng gōngjù de yì zhòng dòngwu 人是用工具的一种动物 (“Humans are tool-using animals.”). The assumption is that rén 人, having no gender marking, is roughly equivalent in all cases to the English word “person” or “human”. But as we shall see, in many cases the word actually carries a strong implicit masculine marking that is only evident when observed in a range of contexts.
In sentences such as the following, rén 人 does indeed seem to have no specific gender association:
Zhè fángzi zhù sānge rén.
(“This room sleeps three persons [rén 人] .”)
Děng yìhuir zài dà. Gōngyòng diànhuà yǒu yìge rén zhèng zài yòng.
(“Make the call later. There’s a person [rén 人] using the pay phone right now.”)
In such contexts where the unnamed person or persons in question are not intended to become fully-fledged actors in any discourse, the word seems truly gender neutral. However, in a large number of linguistic contexts where the listener is required to create a mental “node” for a person who is to have an important role in a discourse, the use of rén 人 is somewhat different. An example of such contexts is the beginning of anecdotes, stories, or jokes. At the outset of a joke, for example, the narrator or writer typically introduces the main character or characters in a short expository sentence:
Yǒu yíge rén, hén yǒu qián…
(“There was a person [rén 人] who had a lot of money…”)
At first sight, the use of rén in here seems to be the same as in the above two sentences; in principle, the person could denote either a male or a female. But sentences do not exist in a vacuum. In the context of tendencies and regularities of Chinese discourse (subconsciously perceived), such sentences carry a strong male default, for the simple reason that when the character in question is female, her sex is most often specified at the outset of the discourse by some gender-specific descriptive term: “Yǒu yíge lǎo taìtai …” 有一个老太太 (“There was an old women…”), “Yǒu yíge gūniang …” 有一个姑娘 (“There was a young girl…”), “Yǒu yíge měinǚ … 有一个美女… (“There was a beautiful woman…”), etc. The covert male default of the word rén 人 is not evident until one actually examines how it functions in contrast to other overtly marked terms. The protagonist in a joke is assumed to be male unless otherwise noted. And when the customary gender assumption is violated, the effect can be slightly strange or disorienting. This can be made clearer by taking some actual beginning sentences of jokes and anecdotes and seeing what happens when gender references are switched. The following are typical opening sentences taken from Chinese joke books:
Yǒu yíge rén, lǎopo sǐ le.
(“There was a person [rén 人] whose wife died.”)
Yǒu yíge rén, húzi hěn cháng.
(“There was a person [rén 人] who had a long beard.”)
The rén 人 in these examples is clearly male. If the word rén 人 were truly gender neutral, one would expect that the word could equally apply to female characters in a similar context. However, note the effect of the word in similar sentences where the referent is female:
Yǒu yíge rén, zhàngfu sǐ le.
(“There was a person [rén 人] whose husband died.”)
Yǒu yíge rén, shēncái rén miáotiao.
(“There was a person [rén 人] who had a slender figure.” The word miáotiao is used exclusively to describe women.)
Most Chinese speakers notice a slight sense of strangeness about these examples. One almost feels as if a crucial bit of information was missing from the outset—namely, the sex of the new person introduced into the discourse. Since “female” is not specified, the assumption is that the person is male, and thus one feels slightly “tricked” when the second part of the sentence reveals the person to be female.
This sort of “substitution game” reveals even more striking asymmetries if one tries to substitute the word rén 人 for gender-specific terms. For example, note the following two sentences, both taken from Chinese joke books:
Yǒu yíge lǎotóur tiāntiān ràng tā lǎopo zuò niúròu miàn…
(“There was an old man who had his wife make him beef noodles every day…”)
Yǒu yíwèi xiǎojie zài haǐtān shài tàiyáng, chuānzhe sāndiǎnshì yóuyǒngyī…
(‘There was a young woman sunbathing at the beach, wearing a bikini…”)
What happens if we replace the words lǎotóur 老 头儿 (“old man”), and xiǎojie 小姐 (“young woman”) with the supposedly gender-neutral rén 人?:
Yǒu yíge rén tiāntiān ràng tā lǎopo zuò niúròu miàn…
(‘There was a person [rén 人] who had his wife make him beef noodles every day. “)
Yǒu yíge rén zài haǐtān shài tàiyáng, chuānzhe sāndiǎnshì yóuyǒngyī…
(“There was a person [rén 人] sunbathing at the beach, wearing a bikini…”)
Because of the strong male default for rén 人, the actual information lost by its substitution in sentence la. is merely the age of the (male) person in question, and so the immediate effect is benign; the listener experiences no gender confusion. The substitution of rén 人 in 2a., however, seems to have a slightly strange and disorienting effect; a subtle, subconscious assumption about the gender of the person is violated. In such contexts it would seem somewhat more natural to be more specific: yíwèi xiǎojie 一位小姐 (“a young lady”), yíge nǚde 有一女的 (“a female”), etc.
Thus the word rén 人 in actual use is not as gender-neutral as is usually supposed; psycholinguistically, it has strong implicit masculine connotations. As the usage patterns of the word become fixed in their cultural context, the implicit male imagery of rén 人 in everyday discourse gradually becomes the “normative” case. with nǚrén 女人, “female person”, becoming a specially marked form.
There are, of course, several compounds in which rén 人 connotes females, for example měirén 美人, jiārén 佳人, lìrén 丽人 (all three basically meaning “beautiful woman”), nèirén 内人 (“interior/inside person”, i.e., “wife”), méirén 媒人(“matchmaker, go-between”), but these are special cases that clearly denote traditional or stereotypical female attributes or roles. Most other compounds predictably behave according to the principle of the “male as default”: yàorén 要人(“VIP”), shèngrén 圣人 (“sage”), cáirén 才人 (“person of talent”), wěirén 伟人 (“great person”), fāmíngrén 发明人 (“inventor”), etc., all of which are equally unmarked, but are most often assumed to be male in actual daily use. Of course, this is because in a male-dominated culture, VIPs, sages, great personages, inventors, etc. will statistically more often be male, and thus it is easy for these unmarked terms to take on masculine connotations. We will have more to say about this below, but it is important to note here that often terms like shèngrén 圣人 (“sage”) and cáirén 才人 (“person of talent”) have special female-marked equivalents like shèngnǚ 圣女 (‘female sage”) and cáinǚ 才女 (‘female person of talent”), and the presence of these gendered lexical items serve to reinforce the masculine associations of the unmarked items, much in the way that the masculine assumptions of words like comedian and hero are reinforced by the existence of female-marked alternatives like comedienne and heroine. And, as usual with such terms, the male form in the plural can denote both sexes, whereas the female term is restricted to women; a list of “historical heroes” might include Joan of Arc, but a list of “historical heroines” could never include, say, Paul Revere.
Asymmetry in pronoun usage
The masculine default for the word rén 人 naturally bleeds into the third-person masculine singular in Chinese, tā 他, which also has strong male defaults, as in the following example:
Dāng yíge rén gǔqi yǒngqì jiù huì zǒu xiàng tā suǒ xīwàng qù de dìfang.
(“When a person [rén 人] musters [his] courage, [he] will head in the direction he [tā 他] wants to go.”)
The problem here is exactly isomorphic to the tradition in English of using the word he in generic contexts. Does the use of tā 他 here assume a male referent, or is the usage truly generic? As with English, the usual response is that the pronoun tā 他 in such contexts has simply taken on a generic meaning, and does not exclude women. But how valid is this justification? In recent years there have been psycholinguistic studies that cast doubt on such claims:
(1) The male generic is problematic when the referent is female. In one sense, the best evidence that the generic he/tā 他 is not truly gender neutral is that when the referent is clearly female, the male pronoun results in a ludicrous clash of imagery. Sentences like “Everyone in New York State is entitled to an abortion if he wants it’, (Hill & Mannheim 1992, p. 388), and “Menstrual pain accounts for an enormous loss of manpower hours,” (Martyna 1983, p. 31). Examples in Chinese are equally bizarre. No competent Chinese writer would ever produce a sentence like the following: Dāng yíge rén shēng haízi de shíhou, tā… 当一个人生孩子的时候 (“When a person gives birth to a child, he [他]”). The evidence is that, unless the context demands a female, the default image is of a male.
(2) The word she/tā 她 tends to be used in cases where the referent is likely to be female. A survey of psychology textbooks (American Psychological Association, 1975) found that hypothetical professors, physicians, and psychologists in college textbooks tended to be referred to with the pronoun he, while the pronoun she was more common when the antecedent references were hypothetical nurses, teachers, and librarians. While I know of no formal survey of such pronoun usages in Chinese, my own impression is that Chinese texts also tend to apply the female pronoun in contexts where a generic pronoun would seem to be required, though the occupation in question is statistically more likely to be female. （Yòuéryuán lǎoshī gōngzuò hěn xīnkǔ, érqiě tāmen de gōngzī hěn dī. 幼儿园老师工作很辛苦而且她们的工资很低。 “Chinese kindergarten teachers have a hard job, plus their [she (plural) 她们的] pay is low “.) Again, if the masculine pronoun truly has a generic function, why would writers feel the need to opt for the female pronoun for occupations traditionally held by women?
(3) Experiments show that readers tend to identify the referent for generic masculine pronouns as male. not neutral. MacKay (1980) actually conducted an experiment to test gender imagery associated with the generic use of he. Ten UCLA students (five male and five female) were given two paragraphs containing the generic he referring to neutral antecedents (person, writer, etc.). After reading the paragraphs, they answered three multiple choice questions, one of which indirectly assessed comprehension of the generic he and its antecedents. In reference to a textbook paragraph about “a beginning writer”, the question testing comprehension of the pronoun he was: “The beginning writer discussed in the paragraph is (a) male; (b) female; (c) either male or female.” Only 20% of the subjects consistently chose alternative (c); 80% of the subjects revealed masculine imagery on 63% of the trials, responding with (a) “male”, rather than (c) “either male or female”. Women made exactly as many errors as men, and in no case did the error involve choosing (b) “female”. More evidence for masculine associations of the generic he comes from Martyna (1978a, 1978b), who showed that women use he as a default pronoun less than men do, more often preferring alternatives such as they and he or she. While one must be cautious in inferring psychological facts about Chinese based on evidence from the English psycholinguistic world, one can be relatively certain that the default masculine tā 他 also semantically triggers the notion of “male person” rather than simply “person”.
Many Chinese writers have become aware of this problem, and have begun to apply the same solution available to writers of English, namely using “他或她” “他/她” (“he or she”, “he/she”) for each referent of unspecified sex. Of course, as many have pointed out, this solution raises the problem of the male-dominant order of dyads; the “natural” male-female ordering perpetuates the sexist assumption of the male as prior, while reversing the order (“she or he”, 她或他) is seen as distractingly unusual or even stridently political. The only other alternative, namely alternating the two orderings randomly in a book or article, is felt by many to be equally distracting and burdensome.
This problem does not arise in the Chinese spoken language, of course, since the word tā simply codes for third-person singular, the male-female-neuter/inanimate distinctions reflected only later on in the writing system. This seems to be a case where the handling of gender information in the lexicon can have psycholinguistic implications for linguistic relativity. From the point of view of English, it is interesting that often a person not present can be mentioned continually in a given discourse for quite some time without the listener necessarily assigning a gender to the mental representation for that person. Many years ago I was part of a translation team in Beijing that included a female physics professor who had left the team early on, and who several members had never met. Since we were revising texts she had translated, her name came up dozens of times over the course of several weeks. At one point far into the project, one of the members asked “Is this Professor Wang a man or a woman?’ My colleague was able to build up a quite complex mental image of the person that nevertheless did not include what for English speakers would be the most basic and (seemingly) crucial bit of information: her gender. This state of affairs is quite impossible in any language with gendered pronouns (and gender-explicit given names), but I have since encountered it countless times in the Chinese linguistic environment.
It is instructive to look at the origins of the character tā 她 to see how sexist defaults led to the current situation with the Chinese pronoun system. Prior to the May Fourth Movement, there was only one written form for the third-person singular, the allegedly gender neutral character tā 他. Later, due to the influence of foreign languages and the necessities of translation, prominent figures in the May Fourth Movement such as Lu Xun and Zhou Zuoren began to suggest creating a new character or characters to represent male and female third-person pronouns in the written language. According to Ling (1989), the first to advocate the use of 她 as the feminine written form while using the existing 他 as the masculine form was the poet and linguist Liu Bannong (刘半农 1891-1934). Interestingly, though Liu’s proposal elicited much debate as to whether the introduction of any new characters was truly necessary, no one at the time raised the obvious possibility of creating a symmetrical character for the masculine third-person singular as well, thus leaving 他 as a gender-inclusive third-person pronoun. In the 1920’s a few writers did propose, in the interests of gender equality, such a character with a “male” radical, written 男也. Poet Liu Dabai (刘大白 1880-1932) actually employed this character in his writing for a time during the early 1930s, but the character never achieved the widespread popularity of 她. Ling (1989) explains the reason for the failure of the character 男也 as follows:
Since the third-person pronoun in Chinese was originally genderless, there was an objective need to create feminine and neutral characters 她 and 它 for the third-person singular. Thus, there was no need to denote masculinity; and since there was no need for such a character, to create one naturally would have been superfluous.”
It is mildly incredible that a contemporary scholar researching the very origins of this asymmetrical “solution” to the Chinese pronoun problem would blindly perpetuate the very sexist assumptions that gave rise to the situation. It is jolting and disheartening to see the flawed logical argument that evidently was subconsciously operative at the time here spelled out so transparently and uncritically. That Ling can baldly assert that there was somehow an “objective need” for feminine and gender-neutral characters when he has just stated that the original character 他 was already gender-neutral is a testament to the intellectual stranglehold male chauvinist assumptions can exert on people. Note also that he seems not to have noticed that originally 他 was not, strictly speaking, gender-neutral, but rather gender-inclusive, covering either male or female, while 它 (“it”) was truly genderless.
The “inevitable” failure of the symmetric character 男也 meant that, once again women were linguistically relegated to a special class. If the non-sexist solution had been adopted, modern written Mandarin would have the following symmetrical, gender-equal pronoun system:
Instead, the pronoun system which resulted from the introduction of 她 is thus:
The character rén 人 serves a double function, either denoting humanity in general, or males in particular. Females are left to a special linguistic category all their own, outside of the general category of human—they are “out of the club”, so to speak.
Interestingly, even the second-person singular has fallen victim to this phenomenon as more and more people (mainly in Taiwan and Hong Kong) are beginning to use a relatively new character for the second-person singular nǐ 妳 (“you [female]”) with the female radical 女 which is in opposition to the formerly gender-neutral character nǐ 你 with the “person” radical (亻). Thus, the exact same gender inequality has been expanded to this case, as well. The character 你 now refers to either males or those of unspecified gender (or both males and females in the plural nǐmen 你们), whereas the character 妳 can only refer to females.
This introduction of a specific written form to differentiate male from female pronouns opens the door to certain kinds of anthropomorphic assigning of gender to inanimate or conceptual entities. Just as in English it is common to refer to ships, cars, or countries with gendered pronouns, since the introduction of gendered pronoun characters, Chinese writers now can apply gender to any entity, as in this example:
“China Spring is a publication started by students from the mainland. In the nine years since its publication, she [tā 她] has become …”
Of course, China, as other cultures, has always attributed gender to certain inanimate or abstract things (the Communist Party, for example, is characterized as a nurturing mother); this new development merely opens the door to more indiscriminate (and sometimes sexist) anthropomorphism.
Covert sexism involving the female radical 女
Countless scholars have pointed out that a great many characters with the female radical have pejorative, negative, or distasteful meanings (for example, the character jiān 奸/姦 —simplified/traditional) a pictograph with three women, meaning “adultery, wicked, evil, treacherous; illicit sexual relations”). But what is seldom noted is the more significant fact that, while characters with the female radical are quite numerous (the 1989 edition of Cihai 《辞海 》 includes a total of 275), there is no male radical at all! (In characters such as shēng 甥 (“nephew”), niǎo 嬲 (“tease”), and jiù 舅 (“uncle”), one could consider that the component nán 男 (“male”) carries a masculine semantic function, though in no case does it function as the radical.) But as we have pointed out, in some sense there is no need for a male radical, since “male” is the implicit default in so many characters and concepts; the seemingly gender-inclusive rén radical (亻) in most cases actually signals maleness just as surely as the female radical (女) marks femaleness. “Persons” are by and large men, and thus it is the female that is perceived as requiring a special marking.
It is difficult to assess the psycholinguistic salience of radicals for Chinese readers. Some studies suggest that characters, when used as radicals, tend to lose their semantic content. (For example, experimental subjects sometimes do not even notice that groups of stimulus characters share a common radical. See experiments surveyed in Hoosain, 1991, pp. 83-84.) It is almost certainly the case that the importance of radicals in remembering, writing, and understanding characters is highly variable and context-dependent. For instance, while the function of the water radical is no doubt fairly salient in characters like jiǔ 酒 (“wine”), yóu 油 (“oil”), and yú 渔(“fishing”), no doubt the semantics of the radical is virtually “dead” in characters like fǎ 法(“law”) or Hàn 汉 (“Han dynasty/nationality/language”).
This same degree of variable psychological salience is evident in characters with the female radical. Many such characters are quite obviously related to the social status of women or to female physiology, or stereotypical feminine traits: for example, qī 妻 (“wife”), fù 妇 (“married woman”), mǎ 妈 (“mother”), jiě 姐(“older sister”), jì 妓 (“prostitute”), jià 嫁 (“to many [of a woman]”), jì 嫉 (“jealous”), jiān 奸 (“wicked, evil, treacherous; illicit sexual relations”), wǎn 婉 (“beautiful, elegant”), jiāo 娇(“tender, pampered, etc.”), xián 娴 (“refined, elegant [of a woman]”), jìn 妗 (“aunt”), etc. There are other characters with the female radical whose female connotations are quite subtle or only dimly evoked: miào 妙 (“marvelous, wonderful”), xián 嫌 (“suspicious”), yú 娱 (“amusement, pleasure”), shàn 嬗(“change, evolve”), etc. And there are still other characters with the female radical that doubtless evoke no female imagery whatsoever in the minds of speakers: rú 如 (“if’), shǐ 始 (“begin”), yào 要 (“want”), Yáo 姚 (a surname), etc. The important point to note is that the feminine associations of various characters carrying the female radical need not be psycholinguistically active in the mind of the reader at each appearance of any given character for the reinforcement of sexist assumptions to be in place. Rather, it is the cumulative and pervasive presence of this factor integrated over a lifetime of conscious and subconscious language use that concerns us here.
Asymmetries in forms of address
While the system of terms of address in Mandarin is sociolinguistically quite complex and differs from that of English in a number of ways, it is interesting to what extent the sexist asymmetries almost exactly parallel those of English. The standard married/unmarried distinction marked for women in English has traditionally been the case in Chinese, as well (the term Miss being equivalent to xiǎojie 小姐 and the term Mrs. equivalent to taìtai 太太, with only one term (Mr., xiānsheng 先生) for both married and unmarried males. In addition, just as couples have traditionally been referred to formally using the husband’s name (“Mr. and Mrs. Jack Smith”), so Chinese newspapers refer to Fang Lizhi and his wife as Fāng Lìzhī fūfù 方励之夫妇
Wu (1991) notes that roles of respect and authority such as shīfu 师傅 master craftsman”) and guofù 国父 (“father of the country”) are virtually always accorded to males, and thus the seemingly symmetric terms shīmu 师母 and guomù 国母 are merely polite terms of address for the wives of such people; terms for independent and equal female fillers of the roles do not exist. By contrast, even though many people now use the gender-specific term lǎobǎnniáng 老板娘 (“boss mother”) to refer to a woman of the same economic position of power as a lǎobǎn 老板 (“boss”), the term is often conflated with the wife of a boss who is economically dependent upon her husband; there is no truly independent term for a female boss. One solution to this problem is to simply address the woman with the term that has traditionally denoted the male (lǎobǎn 老板), in hopes that the term will assume a gender-neutral quality (much in the way that the word wēng 翁, “old man”, in bǎiwànfùwēng 百万富翁 “millionaire”, has lost its literal meaning). However, as with the character tā 他 such usages still tend to assume the male as the “normative” or “generic” member, with women once again consigned to a “non-male” role; the few women who succeed in entering into these male occupational domains find that they are nevertheless unable to achieve a fully independent and normative semantic status in the linguistic domain. This often results in such terms, when applied to a woman, being prefixed with nǚ 女 or nǚde 女的 in everyday discourse in order to avoid the person in question being mistaken for a man. We will have more to say on this topic in the next section.
The term xiānsheng 先生, in addition to functioning as a standard term of address for males (“Sir, Mr.”), has also traditionally been used in academia as a polite form of address for a teacher who has achieved a great degree of respect and achievement (and such teachers were usually male, of course!). I have been surprised to find that women professors to this day are often respectfully addressed in this way in book inscriptions or letters. Xiānsheng 先生 as a term of address applied to a woman academician has a double meaning; not only has she achieved great success in her field, but she has also shown herself to be able to compete on equal footing with men. Such a term of address not only neutralizes the woman’s gender, but once again relegates her to a “special case” within the male-centered domain.
Wu et al (1995) point out that gēmenr 哥儿们儿 (“pal, buddy”, lit. “older brother” + plural suffix), a term connoting a chummy bond between males, has no female equivalent. As a result, many men use this gender-marked word to address or refer to close female friends, and many women also use the term among themselves to refer to or directly address other close female friends. Another option that has arisen is to use a parallel term patterned on the male term, jiěmenr 姐们儿 (lit. “older sister” + plural suffix). But in both cases, the female is somewhat marginalized; either she borrows a male term that etymologically had (and still retains) male connotations, or she creates a female equivalent that is nonetheless patterned after the male term. Either way, the term is subconsciously perceived as merely a copycat version of a male-defined category.
A full treatment of forms of address in Mandarin is beyond the scope of this paper. But a more systematic survey would no doubt reveal the same underlying pattern throughout; almost invariably, the female is linguistically treated as kind of diminutive version of the male standard, an imitation, or a Johnny-come-lately (or should I coin the term “Janey-come-lately”?).
Asymmetric gender marking in vocabulary items
As already mentioned, many scholars researching sexism in English have noted that women engaged in traditionally male occupations are often referred to with a gender-marked term involving the addition of a prefix “woman” or “female” to the standard occupation term. For example, when a truck driver happens to be a woman, she is referred to as a “woman truck driver” or a “female truck driver”. Farris (1989) notes that this state of affairs is seen in Chinese, as yell. Just as “doctor” becomes “woman doctor”, yīshēng 医生 (“doctor”), becomes nǚ yīshēng 女医生 (“female doctor”). But if the driver, doctor, author, police officer, scholar, president, etc., is a man, no such male gender marker is added. Using the same logic, we can infer the origin of the term nǚqiángrén 女强人 (“female strong person”). If the qiángrén 强人 “strong person” in question is male, there is no need to specify the gender—it goes without saying. The case of a female “strong person” is an exceptional case, a curious footnote in the male-dominated discourse. In addition, the literary term zhǔréngōng 主人公 (“protagonist”) includes both male and female literary characters; the term nǚzhǔréngōng 女主人公, “female protagonist”) is gratuitous and only serves to perpetuate the notion of women as special and second class.
Again, the Chinese word processor I am using to create this document has built in to its character entry dictionary several good examples of this asymmetry. One of the features of the software is that when the user enters a given character, in the character window will automatically appear a list of words and compounds for which the character entered is the first element. When I enter the character nán 男 (“man, male”), there automatically pops up a set of characters that can follow it in compounds: rén 人 as in the compound nánrén 男人 (“man”), zǐ 子 as in the compound nánzǐ 男子 (“man, male”), and so on. The list is quite small, only nine items. But when I enter the character nǚ 女, I get a very different and much longer list of possible compounds, including nǚgōng 女工 (“female worker“), nǚbīng 女兵 (“female soldier”), nǚshén 女神 (“goddess”), nǚqiángrén 女强人 (“female strong person”), nǚqīngnián 女青年(“female youth”), nǚhuáng 女皇(“female emperor”), nǚtóngzhì 女同志 (“female comrade”), and so on.
Of course, my word processor is not special in this way; there are a greater number of such female-marked terms in any Chinese dictionary. The interesting question is why such words with the nǚ 女 prefix are accorded the status of separate words. After all, the characters nán 男 and nǚ 女 could be affixed to any term for any occupation or role. If nǚgōng 女工 (“female worker”) is counted as a lexical item, why not nángōng 男 工 (“male worker”)? If it seemed worthwhile to include the special term nǚtóngzhì 女同志 (“female comrade”), why not the symmetrical term nántóngzhì 男同志 (“male comrade”)? And note that English dictionaries certainly do not include entries like “lady lawyer” or “female doctor”, which would be considered compound terms made up of two distinct words. Why should the Chinese lexicon include such items?
Part of the explanation may lie with the question of how tightly bound a given Chinese compound is. Because the principle of affixation functions differently in Chinese than in English, and because Chinese has no orthographic convention corresponding to word boundaries in alphabetic languages, word identity is perhaps psycholinguistically less clear in Chinese. Does the prefix nǚ 女 (“female”) in compounds like nǚshén 女神, nǚhuáng 女皇, and nǚ yīshēng 女医生 result in these being perceived psycholinguistically as analytical two-word phrases corresponding to the English “female god”, “female doctor”, and “female emperor”, or as more tightly bound chunks like “goddess”, “empress” or (to coin an equivalent term) “doctoress”? And do different compounds have different degrees of “wordness”, depending upon frequency? This problem is somewhat isomorphic to the word man in English, which seems to be bound with various strengths to other words in compounds such as chairman, postman, ice cream man, straight man, middleman, omsbudman, Superman, he-man, Peking Man, Sand Man, businessman, marksman, bogeyman, salesman, repairman, best man, madman, taxman, etc. It can be seen from these examples that the notion of a word is rather blurry in English as well, and our notion of which of these lexical items constitute a word and which are composed of two or more words is strongly influenced by our orthographical conventions as well as semantic factors.
At any rate, the perceived need to add a clarifying nǚ 女 prefix is often a reflection of the statistical distribution of the genders in various occupations in the real world, which often vary from culture to culture. For example, it is simply a statistical fact that taxicab drivers in both Chinese and American cultures are predominantly male, while nurses are predominantly female, and it would be strange if our mental assumptions did not reflect this fact. In Russia during the last few decades, there have been more female doctors than male doctors, and it would not be surprising to find that the average person in that country would be less likely to have strong male associations for this occupation. Of course, very few such words have 100% male or female default associations in any culture; rather, we tend to have certain default tendencies based on our experience of the world. Words in English such as “teacher”, “rock‑and‑roll guitarist”, “hairdresser”, “mayor”, “scientist”, “serial killer”, “nurse”, “welder”, etc. all have a subtle “halo” of gender associations, just as corresponding words in Chinese have. Though these words have no explicit gender marking, and thus on the surface are free of any sexist bias, the fact is that the world itself is not free of sexist bias, and there are noticeable gender asymmetries in all areas. Though cognitive defaults can sometimes lead one astray, they are merely the mind’s pragmatic way of paring down the daunting complexity of the world to a manageable level.
Thus, as with the word rén 人, many of these terms carry strong covert male associations. This exerts a subconscious effect on the interpretation of a given sentence. The mind tends to attempt to assign gender imagery to words that otherwise should be neutral. This sentence taken from a Chinese joke book (which I have slightly altered) illustrates the point:
Yī wèi xuézhě zài xīn hūn yàn ěr zhī jì, réngrán shǒu bù shì juàn dú shū zhàngfu fènfèn máiyuàn dào…
(“A just-married scholar on the first night of the honeymoon continued to read and study without letup. Her husband complained, saying…”)
Most people feel a slight jolt of surprise upon reading the word zhàngfu 丈夫 (“husband”); the default assumption that the scholar is male suddenly has to be mentally revised. (I have switched the sex of protagonists, of course.) In the absence of a gender marker, we assume the person in question to be male; otherwise the sentence would have begun by referring to the scholar as yī wèi nǚ xuézhě 一位女学者 “a female scholar”. Most of the time, in our daily language use, these expectations are not violated, and our defaults pass unnoticed and invisible. It is only when we encounter an atypical situation that we become aware that much of the information of the sentence was not in the sentence itself, but supplied by our default assumptions.
These assumptions are also context dependent; the mind takes into account statistical gender probabilities based upon the information given. For example, the word lǎoshī 老师 (“teacher”), is perhaps more likely to evoke female imagery when we are speaking of an elementary school teacher, but will be more likely to evoke male imagery when the context refers to college teachers, who are predominantly male.
Again, it should be stressed that this sort of probabilistic default is a natural and useful cognitive function. In dealing with virtually every situation one encounters and in understanding nearly every sentence one hears, it is necessary to call on a host of assumptions based upon our experiences in the real world. Since genders are not evenly distributed in all occupations, there will be inevitable gender asymmetries in our default associations. If nothing else, this fact underscores the fact that language reform with regard to sexism cannot be carried out independently of social, economic, and political reforms. As society evolves, the language will naturally evolve along with it.
Another linguistic phenomenon related to the gender asymmetry of roles in society is exemplified by words like jiātíng zhǔfù 家庭主妇 (“housewife”), jìnǚ 妓女 (“prostitute”), chǔnǚ 处女(“ virgin”), and the above-mentioned example guofù 国父 (“father of the country”). The gender asymmetry of these terms lies not in their gender associations per se, but in the fact that there exist no corresponding terms at all for persons of the opposite sex who might occupy these roles. For example, whether in China or the West, the spouse who traditionally remains at home and takes care of the housework has been the wife. Thus, both the word “housewife” in English and jiātíng zhǔfù 家庭主妇 in Chinese seem to have no corresponding terms “househusband” or jiātíng zhǔfū 家庭主夫 (“househusband”). It is the female sex that has historically given sexual favors for monetary return, and hence the term jìnǚ 妓女 (literally, “prostitute-female”). Though the emergence of the corresponding phenomenon involving males might seem to require terms like jìnán 妓男 (“prostitute-male”) or nánjì 男妓 (“male-prostitute”), such words have not become common. The case of chǔnǚ 处女 (“virgin”, literally “in-the-house female”) reveals society’s different expectations for males and females with regards to sexual morality. Society expects the woman to have no sexual experience before marriage, whereas there is not such a strong requirement for the male; thus the symmetrical term chǔnán 处男 (“in-the-house male”) never arose in the language. As one final example, teachers in nursery schools are almost exclusively women, and are addressed as āyí 阿姨 (“auntie”). Wu (1991) cites an amusing article in the Guangming Daily about the phenomenon of males taking on this occupation, reporting that they are sometimes referred to not with the corresponding form of address shūshu 叔叔(“uncle”), as one might expect, but as húzi āyí 胡子阿姨 (“ bearded aunties”) !
Asymmetries in jià 嫁 and qǔ 娶
As mentioned above, even when two symmetrical terms exist for both genders, there are often differences in the way the terms behave syntactically that reflect sexist attitudes. For example, qǔ 娶(“to marry [of a man]”) and jià 嫁 (“to marry [of a woman]”) would at first sight seem to be symmetrical verbs, but the pattern of usage reveals significant differences. Liu (1991) has systematically studied the syntactic behavior of these two words, and has shown how their use embodies traditional sexist traditions and notions concerning marriage in Chinese culture. In the context of usage, the verb 嫁 is associated with concepts like gěi 给 (“for, give”), diào 掉 (“drop, lose” [resultative verb expressing loss, discarding, or disappearance]), chūqu 出去 (resultative suffix expressing movement away, disappearance, or elimination), and so on, whereas the verb qǔ 娶 triggers a semantic field of connotations related to “acquisition”, “gain”, etc. For example:
Chūnhuā jià le gěi Xiǎo Lǐ.
(“Chunhua [female] married Xiao Li.”)
Chūnhuā yòu piàoliang yòu yǒu qián, wèishenme zǒng shì jià bu diào (chūqu) ?
(“Chunhua is beautiful and rich. Why hasn’t she been married off?”)
Xiǎo Zhāng bǎ Chūnhuā qǔ le.
(“Xiao Zhang married Chunhua.”)
Chūnhuā bèi Xiǎo Zhāng qǔ le wèi qī.
(“Chunhua was taken by Xiao Zhang as his wife.”)
Liu notes that the relationship implied in grammatical opposition of qǔ 娶 and jià 嫁 is similar to the opposition of “obtain vs. give” or to that of “agent/actor ([shī shì] 施事) vs. object/acted-upon ([shòu shì] 受事)”. Thus qǔ 娶 is semantically associated with or functionally equivalent to such concepts as qǔ 取 (“to take”), dé 得 (“to get, obtain”), huò 获 (“achieve, get”), mǎi 买 (“buy”), yíng 赢 (“win”), etc. The male subject of the verb qǔ 娶 is always the recipient of the action which is “for” (gěi, wèi 给, 为) him. Jià 嫁, on the other hand, is associated with the category of verbs that includes sòng 送 (“give to”), mài 卖 (“to sell”), shū 输 (“to lose”), etc. In other words, the female is always the object of the action—a passive, relatively powerless element, both linguistically and culturally. The covert categories revealed in these patterns are remarkable examples of a social inequality reflected directly in the grammar of the language itself.
All the above phenomena in both English and Chinese have in common the implicit relegating of women to secondary or special status. To borrow a term from linguist Julia Stanley (1977b), females in linguistic discourse occupy what might be called “negative semantic space”. The worldview promulgated, says Stanley, is one with males at the center of the universe, defined as the “Self’, while women are relegated to the status of “Other”. And this gender orientation is expressed in countless explicit and implicit ways, in all the cultural domains in which language plays a part.
To a child, of course, there is no dividing line between language and culture; children learn their native language and absorb the cultural worldview embedded in it before they are able to recognize and cogently analyze the defaults and biases presented to them. Miller & Swift (1977) write of the psychological alienation that girls must confront in learning English, with its historical load of masculine emphasis:
Even if the female child understands that yes, she too is part of “man”, she must still leap the hurdles of all those other terms that she knows from her experience refer to males only. When she is told that we are all brothers, that the brotherhood of man includes sisters, and that the faith of our fathers is also the faith of our mothers, does she really believe it? How does she internalize these concepts? (p. 33)
In adulthood, the gender assumptions embedded in the language become a natural and unconscious part of fluent language use. There is a tight fit between cultural assumptions and linguistic structures, and for this reason few people ever become aware of the implicit gender messages in the language—much less question or challenge them.
It is for this reason that the kinds of covert gender messages embodied in the Chinese language are particularly insidious. Lucy (1992) provides comments on the psycholinguistic significance of Whorf’s distinction between overt and covert categories:
Whorf’s principal claim was that speakers can readily reflect on lexical meanings but tend to be completely oblivious to the patterned grammatical meanings which ultimately govern a lexical item. And overt categories were more susceptible to the critical consciousness of speakers than covert categories with their cryptotypes. In short, some aspects of language are more susceptible to conscious awareness than are others. These remarks suggest that under certain circumstances speakers can and do reflect upon some aspects of their linguistic categories and that these reflections will be skewed by differential awareness of those categories…. This implies a differential salience for thought of some aspects of linguistic form and meaning relative to other aspects. This might be characterized as a differential intuitive awareness. (p. 37-38)
The implications for sexism are that further attention must be focused on the various ways in which gender prejudice is coded and expressed in the language. Hill & Mannheim (1992), in an analysis of the asymmetric pronoun system of English, have this to say about the way such linguistic systems function:
The category system creates a particular cultural hegemony, the unquestioned acceptance, by both men and women, of men as a normative, unmarked category of person…. The hegemonic structure is reproduced below the speaker’s threshold of awareness, unconsciously, but is challenged from above the threshold of awareness, consciously. (pp. 389-90.)
Surely real hope lies in the possibility of a conscious challenge to the generally accepted ways of speaking, thinking, and categorizing. Just as awareness of sexist conventions and social arrangements have led to a gradual elimination of their vestiges in the language, so explicit awareness of deeper levels of linguistic sexism can awaken speakers to previously hidden sexist patterns in the culture at large. It is my feeling that cross-cultural comparisons are especially helpful in this regard, since they tend to suggest productive domains less obvious within one linguistic system. I hope studies such as this one will prod researchers in various areas of linguistics into making further contributions in their particular areas. In addition, with the increasing cultural and scientific exchange between China and the West, there is no reason why the international community of scholars cannot begin to address such issues collectively. Given that both cultures share the same kinds of historical gender inequities toward women, it would seem appropriate that we work together in analyzing these problems and, hopefully, opening the door to solutions.
American Psychological Association (1975). “Guidelines for Nonsexist Use of Language: Task Force on Issues of Sexual Bias in Graduate Education.” American Psychologist, June, pp. 682-84.
Baron, Denis (1986). Grammar and Gender. New Haven: Yale University Press.
Chen Shunxin 陈顺馨 (1995). Zhongguo dangdai wenxue de xushi yu xingbie 《中国当代文学的叙事与性别 》(Narrative and Gender in Contemporary Chinese Literature). Beijing: Peking University Press.
Eakins, Barbara & R. Gene Eakins (1978). Sex Differences in Communication. Boston: Houghton Mifflin Company.
Farris, Catherine S. (1988). “Gender and Grammar in Chinese (With Implications for Language Universals)”. Modern China Vol. 14, No. 3, pp. 277-308.
Francine, Frank & Frank Anshen (1983). Language and the Sexes. Albany: State University of New York Press.
Graddol, David & Joan Swann (1989). Gender Voices. Cambridge MA: Basil Blackwell.
Hill, Alette Olin (1986). Mother Tongue, Father Time. Bloomington: Indiana University Press.
Hill, Jane H. and Bruce Mannheim (1992). “Language and World View”, Annual Review of Anthropology Vol. 21. pp. 381-406.
Hofstadter, Douglas (1985). “Changes in Default Words and Images, Engendered by Rising Consciousness.” In Hofstadter, Metamagical Themas: Questing for the Essence of Mind and Pattern. New York: Basic Books, pp. 136-158.
Hoosain, Rumjahn (1991). Psycholinguistic Implications for Linguistic Relativity: a Case-Study of Chinese. Hillsdale: Lawrence Erlbaum Associates.
Hoosain, Rumjahn (1992). Psychological reality of the word in Chinese. In H. C. Chen and Ovid Tzeng (eds.), Language Processing and Chinese. Amsterdam: North Holland.
Jung-Palandri, Angela (1991). “Gender and Sexism in Chinese Language and Literature”, Schriftestschrifl: Essays in Honor of John DeFrancis, a volume in Sino-Platonic Papers, 27, August, pp. 167-170.
Lakoff, George & Mark Johnson (1980). Metaphors We Live By. Chicago: University of Chicago Press.
Lakoff, Robin (1975). Language and Woman’s Place. New York: Harper & Row.
Ling Yuanzheng 凌远征( 1989). “Ta” zi de chuangzao lishi 《“她” 字的创造历史》 (“The History of the Character ‘ Ta’ 她”) Yuyan jiaoxue yǔ yánjiū《语言教学与研究 》 (Language Teaching and Research), No. 4, pp. 139-151.
Liu Danqing 刘丹青 (1991). “Qu” yu “jia “ de yufa duili he Han minzu dui hunyin de jiti wu yi shi “娶” 与 “嫁”“娶”与“嫁”的语法对立和汉民族对婚姻的集体无意识 (“Grammatical Contrasts Between Qu and Jia and the Collective Unconscious of Han Chinese Nationality with Regards to Marriage”), Wenhua de yuyan shijie 《 文化的语言世界 》 (Cultural Linguistic Worldview), Shanghai: Sanlian Shudian 三联书店.
Lucy, John A. (1992). Language diversity and thought: a reformulation of the linguistic relativity hypothesis. Cambridge: Cambridge University Press.
MacKay, Donald (1980). “Psychology, Prescriptive Grammar and the Pronoun Problem.” American Psychologist, 35, pp. 444-49.
Martyna, Wendy (1978a). “What Does ‘He’ Mean? Use of the Generic Masculine.” Journal of Communication, 28 (I), pp. 131-38.
Martyna, Wendy (1978b). “Using and Understanding the Generic Masculine: A Social Psychological Approach to Language and the Sexes.” Ph.D. Dissertation, Stanford University.
Miller, Casey & Kate Swift (1977). Words and Women. Garden City NY: Anchor Press/Doubleday .
Moser, David and Chen Shunxin （莫大伟，陈顺馨) (1995). Hanyu yinmi xingbiezhuyi dui renzhi de hanyi 汉语隐密性别主义对认知的含义 (“Cognitive Implications of Covert Sexism in Chinese”). Shanghai wenhua 《上海文化》 (Shanghai Culture), 5, pp. 25-36.
Phillips, Susan, Susan Steele & Christine Tanz (eds.) (1987). Language, Gender and Sex in Comparative Perspective. Cambridge: Cambridge University Press.
Schneider, Joseph & Sally Hacker (1973). “Sex Role Imagery and the Use of the Generic Man”. American Sociologist No. 8, No. 1, pp. 12-18.
Shih Yu-hwei (1984). Cong shehui yuyanxue guandian tantao Zhongwen nannü liang xing yuyan de chayi 从社会语言学观点探讨中文男女两性语言的差异 (“A sociolinguistic study of male-female differences in Chinese”). Jiaoxue yu yanjiu 《教学与研究 》(Teaching and Research), No., 6, pp. 207-229. College of Arts: National Taiwan Normal University.
Sorrels, Bobbye D. (1983). The Nonsexist Communicator. Englewood Cliffs: Prentice Hall, Inc.
Spender, Dale (1980). Man Made Language. London: Routledge & Kegan Paul.
Stanley, Julia (1977a). “Paradigmatic Woman: The Prostitute”, in Shores, D. L. and Hines, C. P. (eds.) Papers in Language Variation. University of Alabama Press.
Stanley, Julia (1977b). “Gender Marking in American English.” In A. P. Nilsen et al, Sexism and Language. Urbana: NCTE, pp. 44-76.
Thorne, Barrie, Cheris Kramarae & Nancy Henley (eds.) (1983). Language, Gender, and Society. Rowley MA: Newbury House Publishers, Inc.
Wang, Chuan (1986). “Sexism in Chinese”. Master’s thesis, Department of English, Beijing Institute of Foreign Languages.
Whorf, Benjamin (1956). Language, Thought and Reality: Selected Writings of Benjamin Lee Whorf. Cambridge: Technology Press of Massachusetts Institute of Technology.
Wu Tieping 伍铁平 (1991). Yuyan buneng bufen gaige ma? 语言不能部分改革吗 (“Can’t Language Be Partially Reformed?”) Xiandai waiyu 《现代外语 》 (Modern Foreign Languages) 1991, No. 3, pp. 1-8.
Wu Tieping, Ma Yi & David Moser (伍铁平,马毅，莫大伟) (1995). Guanyu biaozhi lilun he yuyan gaige deng wenti de zai tantao. 关于标志理论和语言改革等问题的再探讨 Further Thoughts Concerning Marked Forms and Linguistic Reform”) Xiandai waiyu 《现代外语 》 (Modern Foreign Languages), No. 1 , pp. 68‑71.
 There are occasional exceptions to this plus-minus ordering, of course, such as black and white, “The Agony and the Ecstasy”, etc., but these are often for reasons of euphony or special dramatic effect. Some dyads allow for more flexibility than others.
 There are, of course, occasional exceptions to the pattern in Chinese, such as hēibái 黑白(“black white”), which incidentally is paralleled in the English collocation pair black and white, but these are merely exceptions that prove the rule.
 The only exception I can think of is “Bonnie and Clyde”, in which the order may be for reasons of euphony, or perhaps due to the sheer novelty of a female bank robber.
 Farris (1989) points out that while “mother and father” is an acceptable compound in English, mǔ fù (“mother-father”) in Chinese is not only non-standard, it sounds utterly wrong to Chinese ears.
 Even the Chinese word-processing software I am using to produce this very article is designed in accordance with this convention. When I enter the pinyin string tā, the first three Chinese characters that appear in the selection window are 他, 它, and 她(“ he, it, she”), in that order. The order the characters is determined by frequency of actual usage, of course, and it is significant that the female pronoun is of even lower frequency than the inanimate pronoun tā 它 (‘it’)!
 This is true in English, as well. For example, the standard pair “Ladies and Gentlemen” reverses the male-female pattern, presumably for chivalrous reasons (“Ladies first!”).
 There are, incidentally, fixed compounds in which both gendered and neutral orderings can be seen side-by-side: nánnǚ lǎoshào 男女老少 (“male-female old-young”, i.e., “everybody”), xiōngdì jiěmèi 兄弟姐妹 older brother-younger brother, older sister-younger sister”), etc. Just as the old-young hierarchy is implicitly recognized, so the parallel male-female hierarchy is reinforced.
 The usual response to the claim that such usages are sexist has been that the word man has lost its literal semantic sense in these cases, and does not psychologically carry any masculine imagery. There is some research that tends to confirm that man carries a halo of male associations even in such generic cases. Two sociologists at Drake University, Joseph Schneider and Sally Hacker (1973). asked 300 college students to select from magazines and newspapers a variety of photographs to illustrate the different chapters of a sociology textbook. Half of the students were assigned chapter headings like “Social Man”, “Industrial Man”, and “Political Man”. The other half were given headings like “Society”, “People in Industry”, and “Political Behavior”. The not-too-surprising result was that those given the headings containing the word “man” overwhelmingly chose photographs of men engaged in the activities in question, whereas the students who had been given the sex-neutral headings chose more photographs which included women.
 Many people have noted that jokes in English that start out with male protagonists (“A man goes into a bar and orders a drink…”) do not establish expectations related to the man’s gender, whereas jokes beginning with a female character (“A woman goes into a bar and orders a drink…”) create the anticipation that the woman’s sex is going to play some crucial role in the story. Linguistically aware feminist friends of mine who try randomly to include women characters in jokes are often met with some puzzlement at the punchline. “But why a woman?” the listener asks, feeling (perhaps justifiably) that the joke-teller would not have “gone out of their way” to introduce a woman into the discourse for “no reason at all”. A woman in a joke is still perceived as a special case. It would be as if the joke began “A 650-lb. fat man walked into a bar and ordered a drink…”, after which the ensuing story made no further mention of the man’s obesity. The information would be perceived as being gratuitous.
 Farris (1988) also points out that rén 人 in idioms often has strong male defaults, as in the idiom Rén jié dì líng 人杰地灵 (“When the people [rén 人] are heroic, the country is auspicious”), and the proverb used to describe a promiscuous woman, Rén jǐn kě fū 人尽可夫 (“Any person [rén 人, i.e., “man” in this case] can be her husband”).
 Note that the female radical in the first character of this compound further reinforces the default female associations of the word.
 It should be noted that even in the case where the sex of the person is obvious or specifically noted in Chinese, the information about their gender is still not a property of the word itself, but rather incorporated as background knowledge. Thus one constantly hears Chinese people who speak English as a second language—even those who speak English quite fluently and who have been in the United States for years—make elementary mistakes in pronoun usage like “You should meet my wife, he … I mean she…”, whereas speakers of European languages seldom make this mistake. Chinese speakers are simply not accustomed to including information about the person’s gender in the pronoun itself.
 Original Chinese: “因为汉语第三人称代词原本无性的分别，由于客观的需要创造了阴性和中性的她字和它字，因此阳性也就再无必要表示了。既无必须，却又便要造新字，自然变成了赘疣。”
 This sort of logical lapse based on deeply ingrained assumptions is actually rather pervasive. I once ran across a college French textbook that stated matter-of-factly: “Unlike French, English-language nouns have no feminine form”; the assumption being that, in a language without grammatical gender, all the nouns are masculine by default!
 And I am speaking here of the subconscious gender imagery of modern Mandarin speakers with regards to the characters, not of the historical reasons for the use of the female component 女 in the character.
 Chinese is somewhat more egalitarian in this regard, however, since it is possible (but not as common) for the woman to be the defining member of the pair. Thus, while no American newspaper would refer to Elizabeth Taylor and her latest un-famous husband as “Mr. and Mrs. Elizabeth Taylor”, Chinese newspapers do occasionally refer to the famous Chinese actress Gong Li 巩例 and her husband as Gǒng Lì fūfù 巩例夫妇 “Mr. and Mrs. Gong Li”). But note the strange gender asymmetry that results from such usages; the order of the term of address, fùfū 夫妇 (“husband-wife”) still falls into the pattern of sexist dyads discussed above. There is no possibility of reversing the order when the woman is the primary member of the pair, e.g., Gǒng Lì fùfū 巩例妇夫 (“Mrs. and Mr. Gong Li” is rarely heard). There remains a residual sense that the wife is still an adjunct to the husband.
 Similarly, Hofstadter (1985) notes that an informal title like “the King of Comedy” applied to a male comedic actor implies that the performer is the best among all his peers, both male and female, but the title “the Queen of Comedy” applied to a woman implies that she is merely the best such female performer.
 This state of affairs is quite reminiscent of the way in which the word guys as a form of address, which originally was an informal word applied to any group of males with which the speaker wished to convey a sense of camaraderie, has been extended to mixed-gender groups or groups of women. The phrase “You guys” is now as likely to be uttered by a sorority member to her fellow members as it is to be said by a football coach to his team. Used in the singular as a common noun rather than a form of address, the word still denotes males only.
 If the occupation in question is one normally taken by women—such as nurse, secretary, etc.—a male taking on the role is also often referred to with a correspondingly marked term: “male nurse”, “male secretary”, etc. However, since the number of professional roles dominated by women has so far been rather small, this mirror-image type of gender marking is far rarer.
 Hoosain (1992) tested Chinese subjects for their knowledge of word boundaries by asking them to divide the string of characters in various texts into words, and found that even highly educated Chinese speakers have weak meta-linguistic awareness of the distinction between words vs. characters.
 Hofstadter (1985) notes that a possible indication of how tightly these “man” suffixes are bound to the word in question is that in tightly bound ones, the suffix tends to be pronounced with a schwa “ə” (châr´mən) rather than with the short vowel (châr´măn), and this may provide a clue as to how strongly the word evokes masculine imagery as well.
 Hofstadter (1985) has the most thought-provoking and entertaining treatment of this cognitive aspect of gender defaults that I know of.
 This is related to a term common in American newspapers, female-headed households. Virtually no one perceives this term as sexist, and yet the obvious underlying assumption is that any household with both a mother and father is by default headed by the male. There is no symmetrical term male-headed households to denote single fathers with children; or if such a term were used, it would simply be interpreted as denoting the standard nuclear family, in contrast to the case where the father was absent.
 We note in passing the obvious fact that both characters have the female radical nǚ 女—a curious asymmetry from the start.