コンテンツへスキップ

学会誌11号-研究ノート3(Kubota)

Japanese Gender Ideologies and Letters
—Female Writing and Its Historical Changes—

Nobuo Kubota

1. Introduction
Language and gender has become a major issue in socio-linguistics since Robin Lakoff (1975) posed the eleven features of female speech style (Usami, 2005). As one of the features of female speech styles, Lakoff notes that women tend to use polite expressions more than men, indicating that they are socially powerless. In fact, Japanese society was dominated by men under the Japanese gender ideology of Dansonjohi (predominance of men over women). Women were considered socially inferior, and due to this hierarchical structure, they were expected to behave accordingly. Thus, being polite and humble in their language was seen to characterize women’s behavior (Reynolds, 1998).
 
However, the above tendency does not seem to have lasted until today, and women’s language may have taken on a different aspect. Satake (1980) states that a presence of vernacular writing in casual letters among young Japanese women started to appear in the mid-1970’s. Following Satake’s study, Kataoka (1997) analyzed 65 casual letters written by young Japanese women (ages 15 to 30) to their close friends regarding affect, such as pictorial signs (hereafter, non-verbal symbols).1 Findings reveal that their letters were filled with various kinds of emotional forms in order to create intimacy and solidarity with the receivers.
 
It is arguable that this phenomenon was related to a new notion of gender equality expressed in the new constitution promulgated in 1946, following the end of World War II. In view of the fact that there was a different notion of gender ideology before and after 1946, it is worth observing how the writing of women is different before and after this period or, more specifically, how the degree to which they express intimacy and solidarity toward their letter’s receiver is different according to the receiver’s status and gender. In this study, I will analyze casual letters written by women to friends collected from three different periods: the years 1918 to 1947, 1948 to 1985, and 2005 to 2010. (Data for the years 2005 to 2010 comes from emails, but I term them ‘letters’ when I refer to general correspondences.) Based upon the frequencies of linguistic elements (honorifics, interactional particles, and non-verbal symbols), I will investigate whether different writing styles among women are seen within and between the three periods.
2. Gender Ideologies in Japanese Society
 Japan used to be a male-dominating society and female social inferiority has a long history. According to the moral textbook called Onna Daigaku2 (The Great Learning for Women) compiled during the Edo period (1603 – 1868), the norm for women’s behavior was defined as the following:
 
 A woman has no master. She must consider her husband her master and serve him with respect. Generally, the way of a woman lies in obedience. Towards her husband, she must be polite, subservient, and humble in her expression and language. She must neither be impatient or disobedient, nor proud or rude. This is the foremost obligation of a woman (as cited in Reynolds, 1985, p.13).
 
Speech styles are closely tied to a hierarchical relation of people. Lakoff (1975), which points out that the speech styles are affected by gender ideology, notes that female speech styles are seen to represent women’s lower social status. And, with respect to the female speech styles and the gender ideology in Japanese, Okamoto and Shibamoto-Smith (2008) indicate that “women should speak quietly (low amplitude) in a high-pitched voice and should use honorifics, certain SFPs [sentence-final particles] like wa and yo ne, indirect speech act, and the like, because these linguistic items/features express the gentleness, politeness, and refinement which constitute onnarashisa ‘womanliness’, or normative femininity” (p.89). Ide (1979) also points out that women’s tendency to use more polite forms and humble forms reveals that women are considered to be socially lower than men.
 
 Japanese society used to be dominated by men until the mid-1940’s, and female speech style inevitably reflected women’s inferior status. However, Articles 13 and 14 of the new constitution express gender equality as follows:
 
Article 13:  
All of the people shall be respected as individuals. Their right to life, liberty, and the pursuit of happiness shall, to the extent that it does not interfere with the public welfare, be the supreme consideration in legislation and in other governmental affairs.
 
Article 14:
 All of the people are equal under the law and there shall be no discrimination in political, economic or social relations because of race, creed, sex, social status or family origin.
 
Some linguists observe a decrease in linguistic differences between men and women, as awareness of gender equality rises after the new constitution was promulgated after World War II (Ide, 1979; Reynolds, 1998; Sugimoto, 1975). One example of this phenomenon is demonstrated by Okamoto and Sato (1992), and they found that in the younger generations, the less-feminine speech style is used. Regarding this tendency, Koyano (2000) notes that a decrease in women’s use of feminine expressions reveals that women have been gradually ranked with men in Japanese society.
3. Degree of Solidarity and Intimacy in Japanese
The vertical society of Japan requires the speaker to adjust his or her speech according to the relationships between the speaker and hearer and their social status. The use of certain elements of the Japanese language, such as honorifics and endearments devices including interactional particles and non-verbal symbols, is dependent on the social status and psychological distance between the speaker and the hearer (Uyeno, 1971; Harada, 1976; Ide, 1982; Kataoka, 1997; Okamoto, 1998).
 
It is generally agreed that Japanese honorifics are separated into three categories: sonkei-go (respectful form), kenzyō-go (humble form), and teinei-go (polite form). The first two forms are considered as a single category called ‘referent honorifics’, and the third form is also known as ‘addressee honorifics’ (Okamoto, 1998). Honorifics are generally regarded to have not only the property of showing politeness but also a certain psychological distance. For example, the speakers use referent honorifics to outgroup persons to keep a certain psychological distance.
 
Interactional particles/sentence final particles are also relevant to the psychological distance between speakers and hearers. Uyeno (1971) points out that “the closer and less formal the relationship between speaker and hearer, the more sentence final particles they will use” (p.63). Interactional particles have traditionally been divided by gender, and Okamoto and Sato (1992) categorize interactional particles into three forms: masculine, feminine, and neutral. As shown below, the adjective, Omoshiroi (interesting) is followed by each gender form, -daro, -wayone, and -yone, but the meaning of the three sentences is completely the same.
 
Examples: Sentence-final particles for seeking agreement

  Masculine Feminine Neutral Meaning
Omoshiroi -daro -wayone -yone ‘(It)’s interesting, right?’
(Okamoto & Sato, 1992)
   
Kataoka, in his study (1997), examined the use of “unconventional practices” of Japanese orthography in casual letters written by young Japanese women.3 He showed that their casual letters are saturated with “affect-laden shape, form, and function” in order for them to build solidarity and intimacy in their relationship with the receivers (p.103). Although the factors of social change are various, he notes that the transition of writing conventions and Japanese social changes seem to be consistent with each other. And, he defines this vicissitude of female writing as “a new mode of literacy” (1997, p.103). What Kataoka means by ‘social changes’ are the changes in Japanese subcultures (e.g., emergence of comic books) and the economic growth after the late 1960s.
4. Research Questions
Given the facts that I have introduced so far, it is of great interest to explore the following research questions:
 
1. Are there differences in female writing styles in letters regarding the given linguistic forms (honorifics, interactional particles, and non-verbal symbols) within and between the three periods?
2. If there are differences within and between the three periods, what are implications of these differences?
3. How has letter writing been influenced by gender ideologies in Japanese society?
5. Methodology
5.1  Data
 
For the purpose of exploring Japanese gender ideologies and letters, I examined casual letters of three different periods: (I) -- the years 1918 to 1947, (II) -- the years 1948 to 1985, and (III) -- the years 2005 to 2010. In total, 35 letters written by 35 different female authors were collected from each period. The letters of the first two periods, (I) and (II), were collected from several collections of letters that contain correspondences written by Japanese female novelists. However, for period (III) I used emails as it has become the accepted form of correspondence, even though letters are still preferred for certain situations, such as: seasonal greetings, business documents, or any other formal letters.4 Data used in this study were letters and emails sent to friends but the ages of the authors were not necessarily equal.5 Contents of the letters were basically reports of the latest news of the authors. There are many different types of letter writing, but I excluded ceremonial and seasonal letters which frequently use set phrases. The data of the letters and authors is summarized in Table 1 and 2, and Tables 3 through 5 explain the hierarchical relation based upon the age gap between senders and receivers (Older persons are highlighted with gray).
 
Table 1: Number of letters (and sentences) in each period.

  (I) 1918-1947 (II) 1948-1985 (III) 2005-2010
to Female 5 (149) 6 (156) 6 (154)
to Male 6 (50) 5 (55) 7 (124)

Table 2: Age range (and average) of authors in each period.

  (I) 1918-1947 (II) 1948-1985 (III) 2005-2010
to Female 25 – 49 (33) 25 – 57 (45) 19 – 40 (26)
to Male 26 – 41 (33) 37 – 66 (50) 24 – 32 (29)

Table 3: Presence or absence of hierarchical relation between senders and receivers in period (I).

Female to Female
 
Female to Male
Sender
Age gap
Receiver
 
Sender
Age gap
Receiver
S1 (25)
3yrs >
R1
 
S6 (30)
1yrs >
R6
S2 (28)
3yrs <
R2
 
S7 (31)
1yrs >
R7
S3 (49)
15yrs >
R3
 
S8 (35)
1yrs >
R8
S4 (31)
=
R4
 
S9 (41)
17yrs >
R9
S5 (31)
=
R5
 
S10 (34)
2yrs <
R10
 
 
 
 
S11 (26)
3yrs <
R11

Table 4: Presence or absence of hierarchical relation between senders and receivers in period (II).

Female to Female
 
Female to Male
Sender
Age gap
Receiver
 
Sender
Age gap
Receiver
S12 (51)
2yrs <
R12
 
S18 (51)
2yrs <
R18
S13 (25)
26yrs <
R13
 
S19 (47)
10yrs <
R19
S14 (55)
7yrs <
R14
 
S20 (66)
9yrs >
R20
S15 (57)
2yrs <
R15
 
S21 (49)
8yrs <
R21
S16 (31)
20yrs <
R16
 
S22 (37)
6yrs <
R22
S17 (51)
20yrs >
R17
 
     

Table 5: Presence or absence of hierarchical relation between senders and receivers in period (III).

Female to Female
 
Female to Male
Sender
Age gap
Receiver
 
Sender
Age gap
Receiver
S23 (26)
1yrs >
R23
 
S29 (30)
5yrs >
R29
S24 (25)
6yrs >
R24
 
S30 (24)
1yrs <
R30
S25 (19)
6yrs <
R25
 
S31 (29)
=
R31
S26 (23)
5yrs >
R26
 
S32 (28)
3yrs >
R32
S27 (21)
4yrs <
R27
 
S33 (27)
1yrs <
R33
S28 (40)
15yrs <
R28
 
S34 (32)
10yrs >
R34
     
 
S35 (32)
8yrs >
R35
 
5.2   Analysis
 
The letters were analyzed according to three factors: (1) referent honorific verb forms, (2) interactional particles of each gender form, and (3) non-verbal symbols (e.g., symbols such as face marks, exclamation marks, ellipses to indicate pauses, phonographic marks, parenthetic Chinese characters, etc). I examined referent honorifics produced anywhere within sentences. However, since honorifics are also used to refer to a third person such as the hearer’s parents, those referred to the third person and the sentences that include them were excluded from frequency counts. Interactional particles that attached phrase-, clause-, and sentence-final position were counted and identified the three gender forms respectively.6 Non-verbal symbols produced at any positions within sentences were counted for frequency as well, but not particularly categorized in any way. 
 
The frequency of use of referent honorifics, interactional particles, and non-verbal symbols were calculated by dividing the total number of the given linguistic forms by the total number of sentences within each period. Calculated frequencies were compared on the given linguistic forms within a particular period, and also compared in order to gauge the differences in female writing styles and their historical changes between the three periods.
6. Results
Although individual differences were more or less observed in each period, analysis revealed obvious differences in female writing style according to the receiver’s gender in the given linguistic elements. In this section, I will show the frequencies of referent honorifics, interactional particles, and non-verbal symbols in each period.
 
6.1  Referent Honorifics
 
There were gender differences in frequency of referent honorifics in the letters within the first two periods, but not within the most recent period. Figure 1 shows the frequency of referent honorifics in each period.
 
Figure 1   Frequency of referent honorifics per sentence in each period
 
Interestingly, the frequency of referent honorifics was different depending on the receiver’s gender in periods (I) and (II). In period (I), women’s use of referent honorifics in letters to the opposite gender was recorded about three times higher than that of letters to the same gender (0.24 for female to female, 0.78 for female to male). In period (II), similar to the previous period, it is clear that female authors tended to use referent honorifics more frequently in letters to the opposite gender (0.24 for female to female, 0.83 for female to male). The result of period (III) is significantly different compared to that of periods (I) and (II). Regardless of the sender and receiver’s gender, referent honorifics were rarely produced by female authors.
 
6.2   Interactional Particles
 
Analysis shows that interactional particles were not produced as frequently as honorifics by female authors during periods (I) and (II), but significantly produced in period (III). Figure 2 below shows the frequency of each interactional particle in each period.
 
Figure 2   Frequency of interactional particle type per sentence in female letters in each period
 
Similar to period (I), interactional particles were rarely produced by female authors in period (II). However, very interestingly, female authors produced masculine forms just as often as feminine forms even though the frequency was low (Masculine: 0.02, Feminine: 0.02). In contrast to the results of periods (I) and (II), interactional particles were much more frequently produced in the emails of period (III). Masculine forms were produced regardless of the receiver’s gender and interestingly, the frequency was higher than that of feminine forms in the female to male emails (Masculine: 0.21, Feminine: 0.07). The feminine forms were slightly less produced than the masculine forms in the emails addressed to the same gender compared with those addressed to the opposite gender (0.04 for female to female, 0.07 for female to male).
 
6.3   Non-Verbal Symbols
 
Non-verbal symbols show the most significant difference from the other given linguistic items. The following figure shows the frequency of non-verbal symbols in each period.
 
Figure 3  Frequency of non-verbal symbols per sentence in female letters/emails in each period
 
Regardless of the receiver’s gender, non-verbal symbols were rarely produced by female authors in periods (I) and (II), but significantly more frequently produced in period (III). Closely looking at gender writing of period (III), they tended to use non-verbal symbols more frequently with the same gender: 0.62 for female to female, 0.5 for female to male.
 
7. Discussion
Politeness theory proposed by Brown and Levinson (1978) is relevant to psychological distance and politeness. If we consider the results, generally, letters in the first two periods (1918 to 1947 and 1948 to 1985) were more ‘honorific-oriented’, and email messages in period (III) (2005 to 2010) were more ‘harmonious-oriented’ with interactional particles and non-verbal symbols to maintain solidarity and intimacy between the senders and the receivers. That is, there is a general shift from negative politeness to positive politeness nearly within the last 100 years. What, then, does this shift imply and how letter writing has been influenced by Japanese gender ideology?
 
From period (I) to (II), the frequency of referent honorifics per sentence in both the female to female letters and the female to male letters did not show a significant difference at all. If speaking politely or femininely is seen to index their lower social status as Lakoff states (1975), women’s frequent use of referent honorifics shown in periods (I) and (II) would support her argument. However, a closer look at Table 4 above reveals that it is more likely that the use of referent honorifics is caused by the fact that the sender is younger than the receiver. Therefore, it is not necessarily appropriate to suggest that the tendency observed in period (II) is linked to women’s lower social status. With respect to interactional particles, interestingly, masculine forms started to be used in period (II). The following are some examples of the masculine forms observed in the female to female letters in period (II).
 
Excerpts
(1) demo, mada nakanaka, ima no nihon dewa, “iiotoko” ga dekiru nowa mutsukashii naa.
(2) onna no ikiru hariai mitaina mono danaa.
 
It is generally considered that women’s use of the masculine forms is unconventional. However, if the notion of gender equality has gradually spread after the new constitution was established in 1946, it is possible to infer that as Koyano states (2000), women have started to adopt themselves to the social norms of gender equality (i.e. women have been gradually ranked with men) since this period. 
 
In period (III), women’s use of the masculine forms was continuously observed and its frequency was greater than that of period (II). Considering this tendency and the very low frequency of referent honorifics in the same period, it is possible to consider that their adoption of the social norms has advanced as time passes. Very interestingly, however, this study reveals that the frequency of the feminine forms in period (III) was greater than that of period (II). Furthermore, considering women’s frequent use of non-verbal symbols observed in this study as similar to the tendency reported by Kataoka (1997), this “new mode of literacy” (Kataoka, 1997, p.103) possibly represents a ‘new femininity’ among young Japanese women. These tendencies observed in modern gender-equal society contradict Lakoff’s argument that “the marginality and powerlessness of women is reflected in both the ways women are expected to speak, and the way in which women are spoken of” (1975, p.45).
8. Future Studies
Although women have adapted themselves to social norms to a certain extent as time passes, there is a possibility that their choice of self-expressions has not been bound by or based upon the vicissitude of Japanese gender ideologies. However, the amount and quality of data used in this study was not completely convincible to support my point of view. For future research, including increasing the amount of data, I would like to more carefully consider the generation of the senders and receivers in order to explore what the vicissitude of female letter writing infers in relation to the current gender ideology.
 
Notes:
  1. Kataoka also examined 27 class notes and memos in addition to the letters written by Japanese young women. (Back to text)
  2. Onna Daigaku is believed to be written by Ekiken Kaibara (1630 - 1714) and published in 1716, but it does not have solid evidence in terms of its authorship (Reynolds, 1985).   (Back to text)
  3. What Kataoka means by ‘unconventional practices’ are “uses of invented punctuation marks, pictorial signs, and intentionally transferred letters to cater to the writer’s need of self-representation” (1997, p.109).   (Back to text)
  4. According to the Japanese Ministry of Internal Affairs, the total number of letter delivery in 2006 has decreased 15% relative to that in 2001 (2008).  (Back to text)
  5. Determination of a relationship between senders and receivers is based upon explanations provided at the bottom of book pages in the collections of letters.   (Back to text)
  6. Okamoto and Sato (1992) specified ‘strongly masculine’ and ‘strongly feminine’ in addition to normal ‘masculine form’ and ‘feminine form’, but not for this study. In terms of unexplained gender forms in Okamoto and Sato’s study, I categorized my own decision.   (Back to text)
 
References
 
  • Brown, P., & Levinson, S. (1978). Politeness: Some universals in language usage, Cambridge: Cambridge University Press.
  • Harada, S. I. (1976). Honorifics. In M. Shibatani (Ed.), Syntax and Semantics 5: Japanese Generative Grammar (pp.499-561), New York: Academic Press.
  • Ide, S. (1979). Onna no kotoba otoko no kotoba. Tokyo: Nihon Kēzai Tsūshinsha.
  • Ide, S. (1982). Japanese sociolinguistics: Politeness and women’s language, Lingua, 57, 357-358.
  • Japanese Ministry of Internal Affair. (2008). Hōdōshiryō. Retrieved July 18, 2010, from http://www.soumu.go.jp/menu_news/s-news/2008/080702_5.html
  • Kataoka, K. (1997). Affect and letter-writing: Unconventional conventions in casual writing by young Japanese women, Language in Society, 26, 103-136.
  • Koyano, T. (2000). Hangyaku suru josēkotoba no jendā. In A. Fujiya, K. Ito (Eds.), Jendāgaku o manabu hito no tameni, (pp.146-159), Kyoto: Sekaishisōsha.
  • Lakoff, R. (1975). Language and woman’s place. Language and Society, 2, 45-80.
  • Okamoto, S. (1998). The Use and Non-Use of Honorifics in Sales Talk inKyoto and Osaka: Are they Rude or Friendly?, Japanese/Korean Linguistics, 7, 141-157.
  • Okamoto, S. & Sato, S. (1992). Less Feminine Speech among Young Japanese Females. In Hall et. (Eds), Locating Power: Proceedings of the 2nd Berkeley Women and Language Conference, 478-488.
  • Okamoto, S. & Shibamoto Smith, Janet S. (2008). Constructing linguistic femininity in contemporary Japan: Scholarly and Popular Representations, Gender and Language, 2 (1), 87-112.
  • Reynolds, K. A. (1985). Female Speakers of Japanese. Feminist Issues, 5, 13-46.
  • Reynolds, K. A. (1998). Female Speakers of Japanese in Transition. In J. Coates (Ed.), Language and Gender, (pp.299-308), Oxford: Blackwell.
  • Satake, H. (1980). Wakamono zasshi no kotoba: Shingenbun icchi-tai. Gengo Seikatsu, 343, 46-52.
  • Sugimoto, T. (1975). Onna no kotoba shi. Tokyo: Yūzankaku Shuppan.
  • Usami, M. (2005). Jendā to poraitonesu: josē wa dansē yori poraito nanoka.  Nihongo to jendaa,  5. Retrieved from https://www.gender.jp/journal/no5/1_usami.html
  • Uyeno, Tazuko. (1971). A study of Japanese modality: A performative analysis of sentence particles, Unpublished doctoral dissertation, University of Michigan, Ann Arbor.
(久保田宣生 くぼたのぶお ハワイ大学マノア校大学院東アジア言語文学部大学院)

 

目次に戻る

Copyright © 2001 The Society for Gender Studies in Japanese
All Rights Reserved.