Trans Internet-Zeitschrift für Kulturwissenschaften 16. Nr. Juni 2006

1.4. Reproduktionen und Innovationen in Sprache und Kommunikation verschiedener Sprachkulturen / Reproduction and Innovation in Language and Communication in different Language Cultures
Herausgeber | Editor | Éditeur: Rudolf Muhr (Universität Graz)

Dokumentation | Documentation | Documentation

BOKU or WATASHI: Variation in self-reference terms among Japanese children

Rika Ito (St. Olaf College, MN USA)



The paper examines the variation of self-reference terms (equivalent to English "I") by using the data from six children (2;11-7;4) of three families who live in and around Sapporo, Japan. There were 227 tokens of self-reference terms within over 3.5 hours of conversation. The quantitative results show that (1) children around 3 years old start using socially expected pronouns, and (2) non-traditional use of pronouns is observed in girls’ speech but not in boys’. Qualitative results from conversational data reveal that (1) WATASHI/BOKU alternation is a "performative strategy" (Okamoto & Shibamoto Smith 2004a) to create a context-appropriate social persona even for a girl in the first grade, and (2) adults’, especially mothers’ attitude towards non-traditional use of BOKU is extremely crucial for preadolescent girls, particularly during the onset of middle childhood (6-11 years old) in contrast to peer groups for teenage girls. Excerpts from a mother-child interaction suggest that an alternate interpretation of use of self-reference terms beyond dichotomized femininity vs. masculinity by adults enable young children to question and challenge the existing social norms.



The Japanese language has been traditionally characterized as having distinct male and female speech styles or two different languages, namely women’s and men’s language (Kindaichi 1957; Ide 1979; Jugaku 1979; Shibamoto 1985; Ide & McGloin 1990). Self-reference and address terms (i.e., first and second person pronouns, respectively), sentence final particles, polite expressions, pitch, and intonation have been identified to describe gendered speech (Ide & Yoshida 1999; Shibamoto 2003). Many studies - particularly the early ones - tend to make a direct connection between a certain linguistic feature and the sex of the speaker. However, studies since the early 1990s uncover inter-speaker variability within a gender group and explore motivation behind these "non-normative" uses (Okamoto & Shibamoto Smith 2004a). Many of these studies point out the danger in dichotomizing male vs. female speech. They argue that it oversimplifies the complex linguistic behaviour of actual usage by real people. Moreoever, such oversimplification only represents cultural constructs reflecting the hegemonic ideology of language and gender that works as linguistic norms to regulate everyone’s speech and to reproduce the norms (Okamoto 1997; M. Nakamura 2001; Okamoto & Shibamoto Smith 2004b). Thus, speakers - particularly women - are expected to follow this ideology, and those who do not conform to this are socially stigmatised (Nakamura 2001). Let us take self-reference terms (equivalent to English "I"), as an example. Table 1 summarizes the expected use in standard Japanese.

TABLE 1: Referring term "I" in Japanese with respect to age, sex, and style
(Based on Ide & Yoshida 1999:471)




















1st Name + chan (= diminutive suffix)






Table 1 indicates that the choice of self-reference terms in Japanese is determined not only by the sex of the speaker but also age and speech style. In a formal situation, WATASHI can be used by adult males and females, and young girls but not by young boys. However, "gender difference" is prominent in informal situations. In an informal situation, multiple forms are available - BOKU for male speakers regardless their age, WATASHI or ATASHI for adult females, and first name + chan (a diminutive suffix) for young girls. Moreover, it is only male speakers who can speak rough by denoting themselves as ORE but not females. Table 1 highlights why self-reference has been said to be "one of the centerpieces" of gendered speech (Shibamoto Smith 2003:209). However, this represents the expected norm, not reflecting actual usage as we see below.

Although the linguistic norms seem to be widely accepted, non-traditional use of self-reference by women is persistent throughout time and space. One of the oldest commentaries on women’s "misuse" (such as use of BOKU by female college students) goes back as early as the late 19 th century (Endo 2001). Critics were almost exclusively negative towards women’s non-traditional use of BOKU until the end of WWII. However, in the post-war period, there are some positive views (Nakamura 2001:211-212). It is the last three decades when researchers have sought for the motivation for the choice against the prescriptive norms. Jugaku ( 1979 :78-80 ) reports that junior high school girls in Tokyo use BOKU in order to compete with boys. About the same period, Ide (Ide 1990/1979) observes BOKU by pre-school girls when they imitate boys.(1) A decade later, Cherry (Cherry 1987) speculates that BOKU is used to create solidarity among young women.

In the more recent studies on this issue, Miyazaki (Miyazaki 2002; Miyazaki 2004) shows the change both in perception and production among middle school girls in Tokyo.



Gender ideology reflected in the literature:



Student’s interpretation





FIGURE figure 1: Assessment of self-reference terms and gender ideology by middle-school girls (Miyazaki 2002:362)

Figure 1 reveals an interesting mismatch between the gender ideology reflected in the previous literature (cf. Table 1) and the interpretation by middle-school girls in Tokyo. In contrast to a dichotomy between masculine and feminine terms in adult’s linguistic norms (or hegemonic linguistic ideology), middle-school girls perceive a continuum - ORE, the most masculine, ATASHI, the most feminine, and BOKU and UCHI,(2) more or less neutral. Miyazaki reports that some of these girls use BOKU quite frequently and sometimes use even ORE in their gakkyuu, their community of practice (Eckert & McConnell-Ginet 1992). Miyazaki’s study highlights that variation in self-reference terms indicates students’ beliefs and attitude towards dominant social norms and how they position themselves in their world in a given context by adopting, resisting, or simply ignoring the social norms.

In contrast to the recent development in the study on variation in self-reference terms summarized above, in the field of child language acquisition, previous studies have mainly focused on when children acquire expected gender-related distinctions (Okubo 1967; Ide 1977; Clancy 1985; K. Nakamura 2001). These scholars found stages in the acquisition of self-reference terms. First, both boys and girls use their first name or nickname to refer to themselves. Then, by age three, boys start using BOKU while girls’ acquisition of WATASHI is slightly later than boys. ORE becomes part of boys’ repertoire at age 3 and half to 4, when they start attending kindergarten (Clancy 1985:382-383). K. Nakamura (2001) finds peers to be critical agents for enforcing the social norms even for small children (age 3-6). In her data, "gender-inappropriate" expressions are corrected by other peers at a rate of 90% (K. Nakamura 2001: 40 footnote).

Although previous studies uncovered various aspects of use of self-reference terms by speakers of both sexes in different ages, there is at least one question that has not been addressed yet. That is, w hen do children start becoming sensitive to their own language use and identity? And more importantly, what and who influences such choice - peers or adults?

This paper reports variable usage of self-reference terms in children’s speech. Although the speech sample is relatively small, our quantitative results show that (1) children around 3 years old start using socially expected pronouns, and (2) non-traditional use of pronouns is observed in girls’ speech but not in boys’.(3) Qualitative results from conversational data reveal that (1) WATASHI/BOKU alternation is a "performative strategy" (Okamoto and Shibamoto Smith 2004: 6) to create a context-appropriate social persona even for a first grade girl, and (2) adults’, especially mothers’, attitude towards non-traditional use of BOKU is extremely crucial for preadolescent girls, particularly during the onset of middle childhood (6-11 yrs) in contrast to peer groups for teenage girls (Miyazaiki 2002; 2004) and smaller children (K. Nakamura 2001). Excerpts from a mother-child interaction suggest that an alternate interpretation of the use of self-reference terms beyond dichotomized femininity vs. masculinity by adults enable young children to question and challenge the existing social norms.


2. Data and Methods

The data come from audio-taped conversations of 3 middle-class families who live in and around Sapporo, Hokkaido, Japan. The dialect spoken in Hokkaido is quite similar to the Tokyo dialect, except for several lexical items and pitch accents.(4) Six family conversations at dinner table and play-time were collected. Family A has twins (a boy and a girl) and they were recorded at three different times (when they were 2;11, 3;5, and 3;7). Two girls of Family B were recorded twice with an interval of 9 months ( 4;0 & 6;7, and 4;9& 7;4). The two boys of Family C were recorded once (5;1 & 6;10). The conversations were recorded from August 2000 to January 2002. Each conversation (45-60 minutes long) was transcribed and every token of self-reference terms was extracted for analysis.(5)


3. Results

3.1 Quantitative Results

There were 227 tokens of self-reference terms within over 3.5 hours of conversation. Table 2 shows the overall distribution of self-reference terms used by the six children.

TABLE 2: Distribution of self-reference terms by children (N=227)



1st name




24 | 45%

13 | 25%

16 | 30%

0 | 0%

0 | 0%


3 | 2%

40 | 23%

59 | 34%

1 | 1%

71 | 41%







Although the frequency may seem to be very low to some readers, low frequency of self-reference terms stems from ellipsis (or the so-called zero pronoun) allowed in the Japanese grammar. In Japanese, the information considered to be recoverable from the context tends to be omitted and sounds more natural than the sentence without ellipsis (Hinds 1982). Example (1) illustrates this phenomenon.

(1) Conversation between a mother and her child(6)




aisukuriiimu tabe-ta?

(girl’s name)

already ice cream


"Sumi, have (you) already eaten ice cream?"


Un. Tabe-ta.

Yes eat-PAST

"Yes, (I) have eaten (ice cream)."


Watashi wa




I already


"Yes, I have, (but not Taro, or Hanako)."

In (1), the mother asks her child, Sumi, whether she has already eaten ice cream. In this utterance, the addresse ("you") is omitted. However, the contextual cues (such as eye contact, there are no other people in the room) provide whom the mother is addressing to. In S1, the actual response to the mother’s question, neither the subject nor the topic of the sentence, WATASHI nor the object, aisukuriimu, is specified. Although S2 is minimally different from S1 with explicit WATASHI, it does not carry the same meaning. With explicit WATASHI, S2 means "Yes, I have eaten in contrast to somebody else." Ide (1977:57) reports that it is quite normal for several children to play together without using either a first or second person reference term for half an hour. Moreover, she finds that ellipsis is the most frequent form of reference term. Thus, our results seem to be quite normal. Figure 2 presents the same information given in Table 2 more visually.(7)

Figure 2 shows that both boys and girls use multiple expressions to refer to themselves. Boys use the forms traditionally associated with male speakers (ORE & BOKU) with an overall preference for ORE (over 40% of the time) while girls use the forms traditionally associated with female speakers (WATASHI) with the same frequency as boys’ ORE. However, ORE and BOKU are still part of the repertoire for girls although the frequency is much lower (2% and 23%, respectively).

When speakers’ sex and age are taken into account, an interesting pattern emergies. Figure 3 shows inter-speaker variability among the three boys.

In Figure 3, the first three sets of bars represent the same boy, Yuta, at three different times (2;11, 3;5, and 3;7). From this figure, we can make four observations. First, consistent with reports by Clancy (1985), BOKU, the mildly masculine form, is found in Yuta’s speech at 2 years and 11 months, and it counts for more than half of his tokens(8). Second, 1 st name is used only by the younger boy. Older boys do not use their first names at all to refer to themselves. Third, as discussed in Clancy (1985) and K. Nakamura (2001), use of ORE increases as boys get older. The youngest boy, Yuta, starts using ORE, the more masculine form, at 3 years 7 months, and it counts for almost one third of his tokens. The 5 year old boy, Nobu, uses BOKU half of the time. The oldest boy, Takao, the 6 year old, uses ORE predominantly - close to 90% of his tokens. Speech samples from these boys are given in (2)-(6).

(2) Yuta-1 (2;11)

a. To his mom


BOKU shara (=sara) iranai.

"I don’t need a plate."

b. To his twin sister


YUTA unten suru n da.

"I-(1 st name) am going to drive."

(3) Yuta-2 (3;5)

a. Playing house


BOKU mo itte kuru.

I’ll go, too."

b. To his twin sister


YUTA tsugi?

Am I-(1 st name) next?"

(4) Yuta-3 (3;7)

a. To his mother


ORE dekiru mon!

" I can do (it)."

b. To his twin sister


Kore YUTA jiichan ni katte moratta n da yo.

"This is something that I-(1 st name) got from Grandpa."

(5) Nobu (5;1)

a. To his older brother


ORE mo.

"Me, too."

b. To his older sister


BOKU no niichan no.

"(That’s) my brother’s."

(6) Takao (6;10)

a. Talking to himself


ORE moo taberarenai.

"I can’t eat anymore."

b. To his mom


Okaasan, BOKU moo kuruma ni nottete mo ii?

"Mom, is it OK for me to get in the car?"

Example (2a) is particularly interesting since even though Yuta has not developed the motor skill to pronounce the first consonant in the word sara ("plate") correctly, his use of BOKU demonstrates that he has already developed sensitivity to the social norm.(9) The examples in (3) to (6) illustrate the effect of age, interlocutors, and speech context by each speaker. Although Takao, the oldest boy, uses ORE most of the time, he uses BOKU to talk to his mother as in (6b).

Now let us examine girls’ speech. Figure 4 shows inter-speaker variability among the girls. For simplicity, three tokens of ORE were combined with BOKU, and one token of ATASHI was combined with WATASHI. The first three sets of bars depict the same girl, Sumi, at different times (2;11, 3;5, and 3;7).The rest of the bars represents the two sisters at two different times - the first two for Nozo (4;0 & 4;9) and the latter two for Haru (6;7 & 7;3).

We find that girls’ use of self-reference terms is quite different from boys’. First, acquisition of feminine pronoun WATASHI appears to be later than that of boy’s BOKU. At 2 years 11 months, Sumi uses only her first name (cf. her twin brother’s use of BOKU 55% of the time at the same period). However, this result is consistent with previous findings: girls’ acquisition of WATASHI is later than boy’s acquisition of BOKU (Ide 1978-9).(10) However, 6 months later, at 3;5, Sumi uses WATASHI quite frequently as shown in (7). At this period, her use of WATASHI counts for 59% of the time.(11)

Speech samples of girls are given in (7)-(11).

(7) Sumi-2 (3;5)

a. To her mother


SHUMIKA (=Sumika) kore chukai tai (=tsukai tai).

"I-(1 st name) want to use this."

b. To her twin brother


WATASHI mo kaki tai.

"I also want to write."

c. Playing house


WATASHI mata okaimono ni itte kuru.

"I’ll go shopping again."

(8) Sumi-3 (3;7) Playing house-2


SUMIKA mo, WATASHI mo. Neru yo!

"I (=1 st name), too, I too. I’ll sleep!"

(9) Use of first name by older girls

a. Nozo-2 (4;9)


NOZO no daisuki na maguro da!

"That’s tuna, my-(1 st name) favorite!"

b. Haru-2 (7;4)


HARU no onaka no oto desu.

"That’s the sound of my-(1 st name) stomach."

(10) Nozo-1 (4;0)

a. N:

BOKU wa?

"How about me?

b. N:

WATASHI sakura gumi ni naritaai.

"I want to be in SAKURA class."

(11) Haru-2 (7;4)

a. H:

BOKU ne sansuu iya da.

"I don’t like math."

b. H:

Demo hon zuki no hito wa WATASHI kana.

"The person who likes books is me."

Second, we find that first name is observed even among older girls (5% in Nozo-2, and 18% in Haru-2’s speech) as in (9) although older boys never use their first name in their speech. Third, although use of WATASHI seems to be increasing as girls get older, its pattern is not as clear as we saw in the increased use of ORE in boy’s speech. In fact, the fluctuation between BOKU and WATASHI is quite high among older girls as in (10)-(11).

We can summarise the quantitative results as the following:

  1. Both boys and girls are acquiring "socially-expected" gendered expressions (except girl’s use of BOKU)

  2. Acquisition of BOKU by boys starts earlier than that of WATASHI by girls

  3. First name as self-reference is predominant in younger children and girls

  4. Boys tend to shift from BOKU to ORE as they get older

  5. Girls’ fluctuation between WATASHI and BOKU is quite high

Among those, girls’ frequent use of BOKU is quite striking but not surprising. The trend observed here supports Miyazaki’s finding. Considering the positive commentaries made by intellectuals in post-WWII period, Miyazaki’s study, and frequent use of BOKU by girls in this study, we can speculate that use of BOKU among girls is becoming more acceptable among the younger generation. But is it really true? The following section provides qualitative evidence for the use of BOKU expressed in one of the family conversations in order to understand the phenomenon better.

3.2 Qualitative analysis

In contrast to the quantitative results that suggest possible spread of BOKU among young girls, conversational data reveal that use of BOKU is not accepted at all by young children. Contrary to our expectation, the BOKU users are perceived as deviant by peers. The following conversational data suggest that it is hard to be a BOKU user, particularly at the school settings. However, with their mother’s guidance and support, girls can be persistent in using BOKU in spite of social pressure that they face in their community. The following speech samples are from a conversation between Haru at 6;7 and her mother.(12)

(12) Haru-1 (6;7) & her mother


Moo soro soro BOKU tte iwanai to dame da.
Kyoo genkai made WATASHI tte itteta.

"I have to use BOKU soon. I was using WATASHI today. I’ve reached my limit."


Fuun. WATASHI tte itteta no?

"I see. Were you using WATASHI?"


Un. Genkai made.

"Yes, up until my limit."





Þ H:

Datte sukoshi wa sa, WATASHI tte sa iwanai to sa minna

ni kiraware chau kara.

"Because if I don’t use "WATASHI" a bit, I’ll be disliked by everyone."

Here, Haru tells her mother that she needs to say BOKU soon because she was using WATASHI at school and she has reached "the limit." She was performing at her school against her preference since otherwise, she will be disliked by "everyone." Her speech reveals that the choice of self-reference terms is a very conscious decision even for a first grader.(13) Moreover, contrary to Miyazaki’s study of middle school girls in Tokyo, use of BOKU is clearly labled as abnormal and devient at least by Haru’s friends, first graders, at a public elementary school in Sapporo. Haru’s mother, on the other hand, seems to distance herself from linguistic ideology. Instead of preaching to her child, she encourages Haru to talk more about the issue ("I see." and "why?").

(13) (Continuation)

=> M:

Sonna koto de kirawareru no?

"Do people dislike (others) because of such a thing?"


Nai to omou.

"I don‘t think so."

=> M:

Un yamete yo.

"I hope not."


Demo sa kirawareru hito mo iru kara sa.

"But there are some people who are disliked (because of that.)"

=> M:



=> H:

Datte sore ja kirawareru yo tte iwareta kara ne chanto

WATASHI tte iou to omottte renshuu shiterun da mon.

"Because (I) was told that (I)’ll be disliked if I use that (=BOKU). So I am practicing using WATASHI properly."

Haru tells her mother that "There are some girls who are disliked because of their use of BOKU." Moreover, non-traditional use of BOKU is overtly criticized by her friends. As we see in Haru’s last turn, she knows that WATASHI is the socially expected way to refer to herself as a girl, thus trying to put some effort to use WATASHI to "fit in" so that she will not be rejected in her community. This clearly illustrates that her choice between WATASHI and BOKU cannot be a simple matter of her personal choice but it requires social acceptance and approval by her peers. In contrast, Haru’s mother, continuously keeps her position to be quite liberal on this matter as expressed in each of her utterances ("Do people dislike (others) because of such a thing?," "I hope not," and "Why?").

After this exchange, the topic of the conversation shifts to food. However, a couple of minutes later, Haru initiates the continuation of the previous topic. Obviously, for her, the previous conversation has not finished yet.

(14) A couple of minutes later


Ano saa, otokonoko no naka de yappari sa ORE tte iu hito ga sa ooi ne.

"Among boys, there are many who say ORE, right?"


Soo kai? Moo minna BOKU tte iwanai?

"Is that so? Don’t they use BOKU anymore?"

=> H:

ORE tte iu hito bakkari.

"They all say ORE."

=> M:

A soo. Jaa BOKU wa onnanoko ga moraeba?

"Then how about girls taking over BOKU?"


Soo da ne. Onnanoko ga moraeba ii n da.

"Yeah, good idea. Girls should take over BOKU."

In this part of the conversation, we learn that boys use ORE rather than BOKU. According to Haru, the boys in her class, "all say ORE." H er mother’s suggestion, "Then how about girls taking over BOKU?" obviously offers a solution for Haru’ conflict.

Although Haru seemed to have been satisfied with her mother’s suggestion, she was not. The topic on self-reference terms comes back more than 10 minutes later in their family conversation.

(15) More than 10 minutes later


Nanka ne WATASHI tte iu no hazukashii.

"I feel shy about using WATASHI."


Jaa BOKU de ii de sho?

"Then why don’t you use BOKU?"

=> H:

Datte iya na no sa. WATASHI tte iwanai to sa otokonoko to machigaerareru koto aru mon.

"But I don’t like it. I am sometimes mistaken for a boy."


=> M:

Dakara sonna jibun no yobikata nante sa, Jibun de yobitai yooni yobeba ii n desho.

"You can refer to yourself however you want to."


Un. BOKU ne

"Yes, BOKU."

Although Haru seemed to be happy with her mother’s suggestion on girls’ taking over BOKU, her utterance in (15) illustrates Haru’s dilemma on the choice between WATASHI and BOKU. She is still thinking about using WATASHI but feels uncomfortable. In addition, although her preference seems to be BOKU, she does not want to be mistaken for a boy due to her use of BOKU. Her mother encourages to " refer to herself however she wants to."

The following is a continuation of (15).

(16) (Continuation)

=> M:

"Hito ni nanka iwareta kara toka hito ni kirawareru kara yameru toka ja nakute WATASHI ga ii to omou n dattara sorede ii n ja nai?"

If you think WATASHI is good, that’s fine. But don’t use it just because someone might dislike you."


Jibun(14) ga ii to omoeba?

"If I think it’s good?"

=> M:

Soo. Jibun ga ii to omoeba, ii to omou yo. Mama wa.

"Right. If you think it’s good, that’s fine, I think."


Kore de ii.


Here again, Haru’s mother keeps encouraging the daughter to decide not based on what others might think but based on what feels right to her. At the end, Haru says "OK." But her statement is not clear as to what she meant. Will she stick to BOKU, which she feels comfortable with, despite possible conflicts with her classmates and others, or switch to WATASHI to conform to the social norm? This segment is toward the end of their conversation and there are only a few tokens of self-reference terms by Haru in the rest of the conversation.(15) However, about the same frequency of BOKU in the later period, (9 months later, Haru-2, at 7;4) indicates that she did not give up BOKU.(16)

During a follow-up interview, Haru’s mother told me that it is BOKU, not WATASHI, that Haru acquired first as self-reference term. According to her mother, Haru started using BOKU prior to her kindergarten (age 4). Among her playmates at that time, both boys and girls used BOKU but the mother did not correct her speech. Once Haru started kindergarten, criticism came from her classmates. Such criticism became even worse in the elementary school. This conversation took place 3 months after she started the elementary school. Thus, she must have struggled for a long time until she finally revealed her true feelings to her mother. In contrast to her peers, her mother takes an exceptionally liberal position towards her use of BOKU. Instead of imposing social norms and gender hegemony on to her young daughter, she encourages to challenge the norm.(17) She repeated more than a few times that it really does not matter how people refer to themselves. The author asked her whether she was a BOKU user when she was young. Her answer was negative. She has never used it. However, she had a "cool" athletic classmate who used BOKU, who was admired by many other girls in her middle school. Although the mother was not a BOKU user, she grew up with BOKU in the 70s-80s, the point in which non-traditional self-reference terms got attention by media and researchers as Jugaku (1979) and Cherry (1987) point out. It can be speculated that the combination of exposure to girls’ non-traditional BOKU and positive image toward users in the past, allows the mother to be critical to linguistic ideology.

As we have seen, support and social pressure that BOKU users face must be different depending on the age group. In contrast to teenagers whose peer groups may provide both support and social pressure, for younger children, peer groups are mainly influenced by the existing linguistic ideology. However, with a mother’s guidance and support, young girls like Haru may still be persistent in using BOKU in spite of the social pressure that they face in their community. This must be particularly true for children at the onset of the middle childhood (6-11 years old) when "coregulation" of power/involvement gradually shifts from parents to peers (Maccoby 1980; Parke & Buriel 1998).


4. Discussion

With a combination of quantitative and qualitative analysis, our results confirm that although young children acquire socially expected gender expressions, resistance to the social norm can happen as early as 6 to 7 years old (i.e., first graders in Japan). As we have seen in Haru’s utterance, the choice between BOKU and WATASHI is complex and intentional even for a first grader. Despite her preference for BOKU, Haru feels that she must conform to the social norm due to social pressure by her peers. By the age of 6 to 7, at the beginning of middle childhood, children not only have developed a high level of metalinguistic awareness but also "practice" it in their daily life (e.g. Eckert and McConnell-Ginet 1992). Moreover, polarization between boys and girls (ORE for boys and WATASHI for girls) among Haru’s classmates suggest that these children may feel more pressure to conform and reproduce the linguistic ideology than children younger than them or teenagers. Part of the pressure can be attributed to the influence of media, which tend to reproduce gender stereotypes (Inoue 2003).

At this point, Haru’s use of BOKU does not mark solidarity or resistance to school as observed in different communities of practice. Her practice is a rather individual/isolated case at her school. However, her use of BOKU raises a fundamental question: Why do women have to speak differently from men? Because of her practice, children in her community witnessed that the linguistic ideology can be challenged although most of them may not support her.(18) It can be speculated that this expression and possibly others may be reintroduced in their later stage (e.g. latter period of middle childhood and adolescence) when they start negotiating individual as well as group identities.(19) According to Haru’s mother, such negotiation seems to have started among Haru’s classmates - now in the 4 th grade.(20)

It was over 100 years ago when female college students were reproached because of their use of BOKU. Has the climate surrounding gender related speeches improved since then? It seems that the answer depends on which side of the coin we focus more. The answer can be negative, since the present study confirms that girls still face exactly the same negative comments by their speech as it happened 100 years ago. The answer can be positive, since the present study shows that Haru, a less than 7 year old girl, challenges linguistic ideology even though none of her peers support her. Her mother provides an alternative way to look at linguistic norms. In order to provide a comprehensive answer, we need to keep documenting and analyzing actual speech by various age groups in different social settings. It is hoped that the current study can make a small contribution to understanding gender related language issues in Japan.

© Rika Ito (St. Olaf College, MN USA)


(1) The citation is found in page 49 of the 1990’s reprint of the original paper published in 1979.

(2)UCHI originally means "inside" or "home" and it is used in some dialects but not in Tokyo. Thus, Miyazaki states that it is a new term and maybe due to its newness, it is a more gender neutral (Miyazaki 2002:363).

(3) It does not mean that boys never use non-traditional self-reference terms. See Miyazaki (2004) for boys’ negotiation.

(4) In Long’s perceptual study (1999: 182), Hokkaido was the only geographical area associated with standard Japanese, except Kanto (Tokyo and its surrounding areas) by his respondents.

(5) Examples of excluded tokens are: Immediate repetition (e.g. Otoosan, WATASHI ishu iranai wa. Watashi ishu iranai. "Dad, I don’t need a chair. I don’t need a chair" (Sumi 3;5) and tokens such as "WATASHImo watashi tte ii tai." (I also want to say "WATASHI.") Here, the first one is included but the second one is excluded.

(6) List of abbreviations are the following: PAST for past tense, TOPIC for topic marker. For simplicity, examples from (2) will only provide English translation without gloss.

(7) For simplicity, one token of ATASHI was combined with WATASHI.

(8) Decrease of BOKU from 2;11 to 3;5 in Yuta’s speech seems quite odd. However, once we consider other factors such as addressees (mother vs. twin sister) and speech contexts (playing house vs. non-playing house), we see an interesting distributional pattern. BOKU & ORE are used when talking to his mother and in playing house context with his twin sister. First name is used with his twin sister in regular conversations (as oppose to play house context). This suggests that the variation is not only related to age, but also interlocutors and speech context.

(9) In fact, the replacement of alveolar fricative /s/ with palatal / ʃ / is often associated with childishness.

(10) Why do boys acquire BOKU before girls acquire WATASHI? Clancy (1986:454-5) suggests that it must be due to the models given by mothers and caretakers. That is, BOKU or BOKU-CHAN can function as addressed’s term for little boys. E.g BOKU Ø onamae wa? (What’s your name?) [A stranger talking to a boy to ask his name.] But such usage is not allowed for WATASHI. (See details in Ide 1977; 1978-79).

(11) Although there is a sharp decrease in the use of WATASHI by Sumi from 3;5 to 3;7 (from 57% to 8%), it is due to the effect of speech context and addressee as observed in Yuta’s speech. Sumi uses WATASHI predominantly in play-house context (close to 80%). Her use of WATASHI is pretty consistent — hovering around 20% when she talks to her twin brother but absent with her mother in both periods. Her frequent usage of WATASHI in play-house context suggests that she is aware of playing social persona Andersen, Elaine Slosberg (1990). Speaking with style. London: Routledge.

(12) It was a recording made in July, 2001, when Haru was first grade, towards the end of the first semester.

(13) Here, "conscious decision" means that Haru is choosing one over the other. The author does not mean that Haru is making a decision based on gender ideology.

(14)Jibun ("self") is a reflexive pronoun and is used for any person and gender. Unlike English refexives, its antecedent does not have to occur in the same clause (See Shibatani 1990; Tsujimura 1996).

(15) Haru used BOKU most frequently before the conversation (two out of three times of the tokens of self-reference terms) and she used less during the discussion although it was still 50% of the time. There were only 4 tokens after the BOKU/WATASHI discussion by Haru and only one of them was BOKU.

(16) Her frequency of BOKU only drops to 45% from 48%. On the other hand, WATASHI declines quite significantly (52% to 36%), and she used first name as pronoun in the later period, which was not used at all in the earlier period. This may reflect another strategy suggested by her mother ("Well then why don’t you call yourself ‘Haru’?").

(17) Some readers may think that Haru’s mother is a rather strange individual who dismisses her role and even speaks differntly as depicted in various media (cf. the mother in Hirokazu Kreeda’s 2004 movie, Nobody Knows). In contrast to those rebellious or non-mainstream mothers, Haru’s mother seems to be quite well respected in the community. She has a college degree, used to have a respective career until giving child birth, and she is really active in the community (e.g. participating in PTAs and volunteering to be a community organizer for youth activity).

(18) In fact, one of Haru’s friends started using BOKU (reported in a phone interview with the mother 5 months after of last recording period). Haru’s mother told the author that mothers of her daughter’s friends commented that the girls must be feeling "cool" by the use of BOKU.

(19) Of course, non-traditional use of self-reference terms can be an experimental and age-graded practice (Uchida 1997 :89 ). Although girls will most likely abandon BOKU, it does not mean that they decide to conform to social norms. They may use other devices to express themselves (such as being assertive).

(20) According to Haru’s mother, it is quite popular for girls to use a new form, uCHI (with LOW-HIGH pitch, in contrast to the original dialectal form) while for boys Ore (with HIGH-LOW pitch in contrast to "traditional" oRE with LOW-HIGH pitch).


Andersen, Elaine Slosberg (1990). Speaking with style. London: Routledge.

Cherry, Kittredge (1987). Womansword: What Japanese words say about women. Tokyo and New York: Kodansha International.

Clancy, Patricia M. (1985). The acquisition of Japanese. In Dan Isaac Slobin (ed.), The crosslinguistic study of language acquisition: Volume 1: The data 373-524.

Eckert, Penelope and McConnell-Ginet, Sally (1992). Think practically and look locally. Annual review of anthropology 21, 461-490.

Endo, Orie (2001). Onnanoko no "Boku/Ore" wa okashikunai [Girls' "Boku/Ore" is not strange]. In Orie Endo (ed.), Onna to kotoba [Women and language.] 30-39. Tokyo: Akashi Shoten.

Hinds, John (1982). Ellipsis in Japanese. Carbondale: Linguistic Research.

Ide, Sachiko (1977). Sex and personal referents of Japanese children. Proceedings of the Symposium on Japanese sociolinguistics, Summer Institute of the Linguistic Society of America Honolulu: University of Hawaii.

Ide, Sachiko (1978-9). A sociolinguistic analysis of person references by Japanese and American children. Language sciences 1, 273-293.

Ide, Sachiko (1979). Onna no kotoba otoko no kotoba [Women's language, men's language]. Tokyo: Nihon Keizai Tsushinsha.

Ide, Sachiko (1990/1979). Person references of Japanese and American children. In Sachiko Ide and Naomi Hanaoka McGloin (eds.), Aspects of Japanese women’s language 43-61. Tokyo: Kuroshio Publishers.

Ide, Sachiko and McGloin, Naomi Hanaoka (1990). Aspects of Japanese women’s language. Tokyo: Kuroshio Publishers.

Ide, Sachiko and Yoshida, Megumi (1999). Sociolinguistics: Honorifics and gender differences. In Natsuko Tsujimura (ed.), The handbook of Japanese linguistics 444-480. Madlen, MA & Oxford: Blackwell Publishers.

Inoue, Miyako (2003). Speech without a Speaking Body: "Japanese Women's Language" in Translation. Language & Communication 23, 315-330.

Jugaku, Akiko (1979). Nihongo to onna [Japanese langauge and women]. Tokyo: Iwanami Shoten.

Kindaichi, Haruhiko (1957). Nippongo [The Japanese language]. Tokyo: Iwanami Shoten.

Maccoby, Eleanor E. (1980). Social development : psychological growth and the parent-child relationship. New York Harcourt Brace Jovanovich.

Miyazaki, Ayumi (2002). Relational shift: Japanese girl's nontraditional first person pronouns. In Sarah Benor, Rose Juie Sweetland and Qing Zhang (eds.), Gendered practices in language 355-374. Stanford: CSLI Publications.

Miyazaki, Ayumi (2004). Japanese junior high school girls' and boys' first-person pronoun use and their social world. In Shigeko Okamoto and Janet S. Shibamoto Smith (eds.), Japanese language, gender, and ideology: cultural models and real people 256-274. Oxford: Oxford University Press.

Nakamura, Keiko (2001). Gender and language in Japanese preschool children. Research on language and social interaction 34, 15-43.

Nakamura, Momoko (2001). Kotoba to jendaa [Language and Gender]. Tokyo: Keiso Shobo.

Okamoto, Shigeko (1997). Social context, linguistic ideology, and indexical expressions in Japanese. Journal of Pragmatics 28, 795-817.

Okamoto, Shigeko and Shibamoto Smith, Janet S. (2004a). Introduction. In Shigeko Okamoto and Janet S. Shibamoto Smith (eds.), Japanese language, gender, and ideology: Cultural models and real people 3-20. Oxford: Oxford University Press.

Okamoto, Shigeko and Shibamoto Smith, Janet S. (eds.) (2004b). Japanese language, gender, and ideology: Cultural models and real people. Oxford: Oxford University Press.

Okubo, Ai (1967). Yoji gengo no hattatsu [Development of child language]. Tokyo: Tokyodoo Shuppan.

Parke, Ross D and Buriel, Raymond (1998). Socialization in the family: Ethnic and ecological perspectives. In William Damon and Nancy Eisenberg (eds.), Handbook of child psychology, 5th ed. Vol 3: Social, emotional, and personality development. 463-552. Hoboken, NJ John Wiley & Sons, Inc.

Shibamoto, Janet S. (1985). Japanese women's langauge. New York: Academic Press.

Shibamoto, Janet S. (2003). Gendered structures in Japanese. In Marlis Hellinger and Hadumod Bußmann (eds.), Gender across languages: The linguistic representation of women and men 203-225. Amsterdam/Philadelphia: John Benjamins.

Shibatani, Masayoshi (1990). The languages of Japan. Cambridge: Cambridge University Press.

Tsujimura, Natsuko (1996). An introduction to Japanese linguistics. Malden, MA & Oxford: Blackwell Publishers.

Uchida, Nobuko (1997). Kaiwa kodo ni mirareru seisa [Sex differences in conversation]. In Sachiko Ide (ed.), Joseigo no sekai [The world of women's language] 75-93. Tokyo: Meiji Shoin.

1.4. Reproduktionen und Innovationen in Sprache und Kommunikation verschiedener Sprachkulturen / Reproduction and Innovation in Language and Communication in different Language Cultures

Sektionsgruppen | Section Groups | Groupes de sections

TRANS       Inhalt | Table of Contents | Contenu  16 Nr.

For quotation purposes:
Rika Ito (St. Olaf College, MN USA): BOKU or WATASHI: Variation in self-reference terms among Japanese children. In: TRANS. Internet-Zeitschrift für Kulturwissenschaften. No. 16/2005. WWW:

Webmeister: Peter R. Horn     last change: 19.6.2006     INST