Ikema Ryukyuan: Investigation Past Experience and the Current State through Life Narratives

Ikema Ryukyuan: Investigation Past Experience and the Current State through Life Narratives

1. Introduction

There are two goals in this paper. The first is to describe the current endan-gered status of one of the endangered languages spoken in Okinawa, Japan by identifying a set of factors that has affected this language. The second is to promote the life narrative approach that we have adopted to arrive at a reliable basic description which is necessary for documentation. We pro-pose that the life narrative approach is an important framework for docu-menting all endangered languages.

The online edition of UNESCO’s Atlas of the World’s Languages in Danger (2009)² identifies four Ryukyuan languages as ‘definitely endan-

• We would also like to thank Maggie Camp and Yuka Matsugu for their help in the prepara-tion of this manuscript.

Japanese/Korean Linguistics 19. Edited by Sang Yee Cheon. Copyright © 2010, «GreetingLine»

/ SHOICHI IWASAKI & TSUYOSHI ONO

gered’ (Miyako, Okinawa, Kunigami, and Amami) and two as ‘severely endangered’ (Yonaguni and Yaeyama) according to the following defini-tions. ³

Definitely endangered: Children no longer learn the language as mother ton-gue in the home.

Severely endangered: Language is spoken by grandparents and older genera-tions; while the parent generation may understand it, they do not speak it to children or among themselves.

In this paper, we examine the current state of endangerment surround-ing one of the dialects of the definitely endangered language of Miyako Ryukyuan, Ikema. Ikema is named after the island where people speaking this dialect originally came from.⁴ It is often reported that at this point only people in their 60s or older (the grandparents’ generation) speak Ikema fluently. This suggests that the current parental generation does not speak the language anymore. If this is true, Ikema is moving from the ‘definitely endangered’ state to the ‘severely endangered’ status rapidly.

There are, however, somewhat conflicting reports being made. In 2005 and 2006, Heinrich, for example, surveyed speakers’ use of their own local variety when addressing different types of speakers (i.e. spouse, children, parents, grandparents, neighbors, and colleagues), and found out that the vitality of Miyako, as well as Amami and Yonaguni, is stronger than that of Okinawa and Yaeyama (reported in Fija et al. 2009).⁵ Some researchers also suggest that Ikema is relatively healthy (Shigehisa Karimata, personal communication).

Based on the above-mentioned survey by Heinrich, Fija et al. (2009) suggest that the vitality of Yonaguni is due to the strong sense of commu-nity resulting from its small size and isolation. In the case of Amami, they suggest that because its US occupation ended much earlier (in 1953) than other Ryukyu islands (in 1972), the impact of language shift there was less drastic, i.e. speakers on other islands felt shifting to Standard Japanese was

• http://www.unesco.org/culture/ich/index.php?pg=00206

• http://www.unesco.org/culture/ich/index.php?pg=00139

• Besides Ikema, there are three to four major dialects of Miyako. Previous research on the Miyako language is mostly based on its central dialect spoken in Hirara on the island of Mi-yako (Shibata 1980, Izuyama 2002, Nevsky 2005), and descriptions of Ikema are limited to preliminary analyses of vocabulary, phonetics, phonology, morphology, and syntax (Hirayama 1983, Nakama 1992).

• The positions of Yonaguni and Okinawa are reversed when compared to the UNESCO re-port.

IKEMA RYUKYUAN / 3

essential for social mobility once they fully became Japanese citizens after the reversion in 1972. When it comes to the situation of Miyako, however, Fija et al. do not have a clear explanation.

Miyako also did not experience radical language shifts, but for quite different reasons. … While a detailed account for this is not yet possible and would re-quire detailed field work, the reasons seem to include the absence of in-migration and continuance of subsistence farming. (emphasis added)

While we leave to future research the ‘absence of in-migration’ and ‘con-tinuance of subsistence farming’ as possible causes for slower language shift in Miyako, we take this statement as an appeal for a detailed sociolin-guistic field work for the situation in Miyako.

Using a broad labeling such as a ‘definitely endangered language’ as a useful starting point, we propose to begin a more detailed examination of the situation. More specifically, we attempt to identify factors that influence the community’s and individuals’ language use.

The UNESCO report suggests the following six major factors:⁶ (1) Int-ergenerational Language Transmission, (2) Absolute Number of Speakers,

• Proportion of Speakers within the Total Population, (4) Trends in Exist-ing Language Domains, (5) Response to New Domains and Media, and (6) Materials for Language Education and Literacy. The UNESCO Ad Hoc Expert Group on Endangered Languages, however, stresses that ‘(n)o sin-gle factor can be used to access a language’s vitality’ (2003: 7). To under-stand the intricacy of the interaction of factors, we have taken an approach in which we engaged speakers in extended life narratives during an inter-view session. We will discuss this approach in detail in section 3, but we will first introduce the community of Ikema speakers.

2. Ikema Speaking Communities

Currently, Ikema is spoken in three main communities. The original loca-tion of Ikema speakers is Ikema Island, which has a land area of 2.62 square miles and a current population of about 736. It is located north of Miyako Island, which is itself located 170 miles southwest of the main Okinawa Island and just 180 miles northeast of Taiwan. Until a 4675-foot bridge between Ikema and Miyako was opened in 1992, the island was only accessible by sea. It flourished as a port for deep sea bonito fishing until the mid 1970’s. However, fishery is no longer a vital industry on the island. In

• The report also mentions the local and majority communities’ attitudes towards an endan-gered language and the amount of available documentation materials as something that should be considered when the degree of endangerment is assessed.

• / SHOICHI IWASAKI & TSUYOSHI ONO

1874, some residents were forced to relocate from Ikema Island to a new location in the northern part of Miyako. This second community was named Nishihara. The current population of Nishihara Village is 1034. The major industry is sugar cane farming. The third location where Ikema is spoken is Sarahama Village on Irabu Island, a community created approximately 300 years ago. The current population is 3264. Here fishing is still strong, unlike on Ikema Island. Irabu Island is currently only accessible by sea, but a bridge is under construction with the opening date set for 2013. Even after a long period of separation, speakers of the three disparate communities still identify themselves as Ikema Minzoku ‘Ikema Race’, and probably due in part to their strong identity, the profile of the Ikema language is distinct from other varieties spoken on Miyako. Our research team⁷ has been con-ducting field work mostly in Ikema and Nishihara, but recently we have also started to explore the situation in Sarahama.

Before we conducted the current research, we made the following cas-ual observations regarding the types of speakers. There is no monolingual Ikema speaker, though some speakers over 80 years old are more comfort-able speaking in Ikema. People between 60 and 80 years old are bilingual speakers of Ikema and the local variety of Standard Japanese, and most people in this group can switch between the two languages smoothly de-pending on the social situations. People between 40 and 60 are diverse in terms of their linguistic ability, but most, if not all, have a high level of comprehension ability in Ikema. People younger than 40 are stronger in Standard Japanese, and the younger members may be completely monolin-gual in Japanese.

The current populations in the three communities, published in 2010, are 736 (Ikema), 1034 (Nishihara), and 3264 (Sarahama). ⁸ Among the community members, however, the number of children is small; only 8% of the population attend elementary or middle schools.

Table 1.

• The team consists of researchers from universities in Canada, Japan, and the U.S.

• http://www.city.miyakojima.lg.jp/site/view/contview.jsp?cateid=29&id=437&page=1

IKEMA RYUKYUAN / 5

In contrast, the number of elders is extremely high. According to the 2005 census, on average the percentage of the population over 65 years old is greater than 33% (49.5% in Ikema, 37.2% in Nishihara, and 29.5% in Sara-hama). These statistics show that the basis for continuing cultural and lin-guistic tradition is very weak. Our task in this paper, however, is not to pre-dict the future of Ikema, but to explore whether our casual observations given above are confirmed. More importantly, we attempt to find out how the current language situation came about.

3. Life Narrative Approach

As found in a quote by Himmelmann, the ultimate goal of our current re-search is to fully document one dialect of an endangered language:

A language documentation … aims at the record of the linguistic practices and traditions of a speech community (Himmelmann 2002: 9)

To achieve this goal, it is not sufficient to work with a single ‘good’ speaker as is often done in a language description study. Instead we set our target as documenting Ikema used by the whole community. As part of the Ikema documentation project, we have been conducting in-depth interviews with community members regarding their language use in various situations throughout their lives, including language choice, code mixing, school poli-cies, fluency, and bilingualism, among others. This has provided us with a rich array of information unobtainable in quantitative sociolinguistic sur-veys based on questionnaires.

Two related issues which we explore in the present study are:

1. Current state and degree of endangerment (e.g., the number of flu-ent speakers of Ikema and the status of Ikema/Japanese bilingual-ism)

1. Factors for language shift and the way in which they contributed to the current linguistic situation

In order to gather information regarding these issues, we initially pre-pared a set of questions concerning topics like age; birthplace and places of residency; family structure; primary caretaker; schooling; occupation; life-style; available language media including newspaper, radio and TV; and language use at various stages/places. Our original plan was to go over these interview questions quickly with a large number of community mem-bers (similar to the way data collection in quantitative sociolinguistics is typically conducted), but when we began interviewing people, it became immediately clear that we could effectively gather information only by en-

• / SHOICHI IWASAKI & TSUYOSHI ONO

gaging the interviewees in a conversation in which they talk about their experiences in the form of life narratives.

Obviously, reflecting actual language use, especially from several dec-ades ago, is a daunting task. This is particularly difficult in determining which language variety was used in specific situations since people seem to switch between Ikema and Japanese rather unconsciously. In fact, we found out at the very beginning that our interviewees, in trying to answer our questions, recounted their experience in the form of conversational narra-tives. For this reason, we abandoned our original plan and switched to the standard methodology of oral history where we let the interviewees tell a narrative while we played an active role as conversational partners (Labov 1984: 32-42).

In the new approach, we normally started each interview by asking some of the questions mentioned above, but when the interviewees started telling a narrative, we let them take the floor. We tried to elicit answers to our questions at places which seemed appropriate. Naturally, as we spent more time talking to community members, relevant factors became increas-ingly clear. We then dealt with these factors by implementing a new set of more specific questions. This made it necessary for us to constantly revise our questions, and to have (sometimes multiple) follow-up sessions with our interviewees and new sessions with other community members (see the next section for more about this).

The actual format of the interview varied from session to session. When we talked with an interviewee alone, we used Japanese as all the interview-ees can speak it. When there were one or more other Ikema speakers at the session, interviewees sometimes switched to Ikema when they needed to solicit information from others. Thus, this approach has made the process of data collection much slower than we had originally planned but allowed us to access more detailed and accurate information concerning language use in various situations, which is indispensable in identifying relevant factors and determining how those factors might have interacted among themselves to create a unique situation for each person in particular and to result in the current endangered state of Ikema in general.

In 2009, we interviewed 23 community members (11 from Ikema and 12 from Nishihara; 12 male and 11 female) who were between 39 and 69 years old. Unless otherwise noted, ages mentioned in this paper are from the summer of 2009. These interviews lasted a minimum of 45 minutes, with follow-up sessions as discussed above. These longer interviews were supplemented by additional shorter interviews with approximately 25 inter-viewees, including several from Sarahama. The shorter interviews were conducted to reconfirm factual information and to fill in information gaps.

IKEMA RYUKYUAN / 7

4. Factors

In this section, we will discuss several factors which we have identified as relevant to the current endangered state of Ikema. As will become evident, it is often not easy to separate these factors because they co-occur and in-teract among themselves creating a unique situation for each individual, resulting in the current condition of endangered state of Ikema. The rele-vance of some of these factors became apparent only through the course of the conversational narratives which we had our interviewees engage in.

It is important at this point to define fluency as we refer to it throughout the paper. We determine fluency based on two criteria: a) assessment by community (typically elder and multiple) members and b) our observation of the person’s ability to speak only in Ikema without code mixing.

4.1. Schooling

Education is always a strong means to implement governmental language policy. Even before Standard Japanese was officially defined by the Meiji government, local institutions were established in Okinawa in the 1880’s to train teachers in the Tokyo dialect. The central government policy was en-dorsed so strongly by people in Okinawa that even a regulation to prohibit the use of ‘dialects’ on campus was established (Hokama 1971: 80).

Our interviewees received primary education during and after World War II, when language regulation was also strongly implemented. Their regular exposure to Standard Japanese started when they attended kinder-garten in the village.⁹ Since our oldest interviewee (born in 1939) attended kindergarten, we can conclude that it was already operational during the war, if not earlier. Thus children who were around five years old heard lo-cal teachers (from the same village) speaking Japanese, while at home most, if not all, of our interviewees spoke Ikema.

Our interviewees all attended local elementary and junior high schools during the post-war period in the village. With the defeat in the war, Japan began to transform itself into a modern democratic nation. The government saw as its responsibility educating its people for full participation in society, which was also strongly desired by people in Okinawa which was sched-uled to return to Japan in 1972 after the post-war US control. A manifesta-tion of the strong sentiment for this desire was the use of infamous hogen-fuda ‘dialect tag’ at school (Hokama 1971: 59; Itani 2006; Kondo 2008).¹⁰

• We checked this with most of our main interviewees and found that they all attended kinder-garten.

• The first use of hogenfuda was recorded in the late Meiji period (early 20^th century), and reappeared before the Pacific War. Hogenfuda in the post-war period can be considered to be part of a more comprehensive measure adopted by schools and communities. Other measures

• / SHOICHI IWASAKI & TSUYOSHI ONO

Hogenfuda was made out of a piece of wood or cardboard usually in a rectangular shape (2” x 6” to 7”). Some were painted red. On them, the phrase hogenfuda ‘dialect tag’ or hogen shiyoosha ‘dialect user’ was writ-ten. A student who was caught using even a single word of the ‘dialect’ was given the tag and had to wear it around his/her neck until a different student spoke a word of ‘dialect.’ This practice does not seem to have been regu-lated closely and different teachers appear to have used it in somewhat dif-ferent ways, i.e. some were more strict and others more lenient. Although some interviewees reported that there was an element of playfulness,¹¹ most students took it seriously.

Although it may have actually been detrimental to young students’ de-veloping identities, this practice most likely resulted from the concern that parents, teachers, and community leaders had for their children’s future well-being. In other words, it was not a top-down mandate in the education system, but rather a local practice that emerged only in Okinawa. However, the fear of being a non-fluent Japanese speaker is not without foundation as many elders experienced linguistic discrimination during the war and at locations of immigration.¹² This concern became more real when young speakers received strong incentive to seek employment in major cities dur-ing the 1960’s when Japan was transforming itself from a war-torn nation into an economically powerful modern nation. Thus the language shift in the Okinawan context was motivated by a mixture of fear, an inferiority complex, a desire to be part of the mainstream society, and an effort to-wards upward social mobility.

The hogenfuda was already in use in 1920’s on Ikema Island.¹³ How-ever, this infamous practice seems to have disappeared sometime in the first half of 1960’s based on our interviewees’ accounts. On Ikema Island, Ms. Y who started going to elementary school in 1961/2 experienced it, but Mr. F who started going to school in 1966 didn’t. In Nishihara, Mr. I who

include honoring households that use Japanese daily and appointing a student to watch over other students’ language use. Except for unconfirmed report in the Tohoku (northeast) region in Japan (Itani 2006), the use of hogenfuda was reported only in Okinawa. This unique form of punishment is believed to have its origin in the ‘punishment tag’ used traditionally throughout Ryukyu in the Ryukyu Kingdom period (Itani 2006).

• Several speakers reported that when they had the tag, they would step on another student’s foot to make her/him exclaim in Ikema, agai ‘ouch’, thereby forcing her/him to wear the tag.

• ‘Dialect’ speakers were viewed as spies during the war (Yakabi 2007: 161), and ‘dialect’ speakers were ridiculed by other Japanese immigrants in places like Hawaii and South America (Hokama 1971: 83).

• This information was obtained from an elderly woman on Ikema introduced to us by one of our consultants, Mr. Morio Iraha (2009). According to Itani (2006: 42), Kondo (1999: 49) also found a record of the earliest use of hogenfuda in Nishibe elementary school (in Nishihara) in the 1930’s and 40’s.

IKEMA RYUKYUAN / 9

started school in 1957 experienced it while Mr. SS who started school in 1962 did not.¹⁴

Based on this information, the following picture begins to emerge. At least until the early to mid 1960’s, children entered elementary school still primarily speaking Ikema and the extreme measure of using hogenfuda was necessary to change this situation.¹⁵ Because of this severe language con-trol, students gradually became comfortable with Japanese after the mid 1960’s, and perhaps this to some extent satisfied the parents’ and teachers’ desire to provide a better future for their children. This also explains the fact that people who are younger than 50 years old as of 2009 in general do not have the speaking facility of Ikema. The 1960’s seems to be a critical decade for the language shift.

4.2. Media

Though hogenfuda disappeared in the early to mid 1960s, media in the form of TV seems to have taken over the role of promoting the use of Stan-dard Japanese. The decline of Ikema is particularly apparent starting with people who are around 50 years old, which may be explained in part by the introduction of TV to the area in the late 1960s (most likely 1967 in Ikema and a few years earlier in Nishihara). Although TV sets were originally acquired by only some households, TV programs in Standard Japanese be-came immediately available to the community as a whole because TV sets were often placed facing the street to show the programs (and perhaps to show off the sets!) to neighbors. People who are 50 years old in 2010, for instance, had regular exposure to Standard Japanese through TV starting when they were first graders or younger. Researchers on endangered lan-guages have discussed TV as being a ‘cultural nerve gas’ (Krauss 1992: 6) which eliminates minority languages, and the spread of this technology in Japan seems to have been very effective to this end, even on the remote islands of Okinawa.

• This estimate of time corresponds with a statement in Itani (2006: 161). Itani considers the emergence of Okinawan identity as the cause of the disappearance of hogenfuda, but if this is the case, the revitalization of Okinwan languages should have been voiced more strongly. On the contrary, Japanese became the language of choice.

• This does not mean, however, that all the children were using Japanese regularly even after the mid 1960’s. We often heard mention of the slogan (“Let’s use the Standard Language”) written on the blackboard and specially selected students to told to watch over the use of lan-guage among students. One interviewer reported that she promised to promote the regular use of Japanese once she was elected class leader in the mid 1970’s. Also, though hogenfuda may have disappeared, students were sometimes forced to stand in the back of the room, sit on the floor, or were pinched on the chest when they slipped into ‘dialect’.

• / SHOICHI IWASAKI & TSUYOSHI ONO

4.3. Primary Caretaker

Our interviewees and others often commented that those who were raised by grandparents (especially grandmothers) speak Ikema fluently. We do have several speakers who fit into this category: Mr. N (62 years old), Ms. O (61 years old), Ms. M (56 years old), and Ms. K (54 years old). However there are several exceptions as well. Mr. S (69 years old), for instance, is considered to be fluent even though he was raised by his parents. One might suggest that Mr. S is old enough to be fluent even without having been raised by his grandparents. Ms. Y (53 years old) and Ms. S (44 years old), on the other hand, are not fluent even though they were raised by grandparents. Perhaps they fall below this critical age for fluency. Interest-ingly, we found out that Standard Japanese was regularly used at Ms. S’s home because her father came from Yaeyama where a different Ryukuan language is spoken and did not speak Ikema.

This suggests that what is more directly relevant to fluency in Ikema might be the way the speaker lived their life (i.e. lifestyle), which includes one’s regular interaction with fluent speakers, perhaps especially when they were children. This will be further discussed in the following section.

4.4. Lifestyle

Four of our younger fluent speakers, two of them our youngest, were not raised by grandparents. Closer examination of their lifestyles reveals more direct causes for shaping speakers’ fluency.

Mr. G (55 years old) in fact recalls the difficulty he had speaking Japa-nese when he first moved to Tokyo to work after high school, at the age of 18.¹⁶ It turns out that his high school, located in the regional political and commercial center of Miyako Island, was a fishery school attended mostly by men, many of whom were Ikema speakers. As a result, Ikema was widely spoken at this high school. Mr. SS (53 years old), another fluent speaker, attended the same high school and had a similar linguistic experi-ence there.

Mr. F (49 years old) went to college in Okinawa where he lived for sev-eral years with Ikema speakers in a place other Ikema speakers regularly gathered, a situation which might be best described as a mini-Ikema ghetto. The life of Mr. TS, the last speaker in this category, is the most intriguing. He is the youngest (47 years old) among the four and just like the other three was not raised by his grandparents. However, he did grow up in an Ikema immersion environment where he was looked after by Ikema speak-ing apprentices who worked for his father, a master carpenter. He later

• N (62 years old) recounts a similar experience when he moved to Okinawa to go to college.

IKEMA RYUKYUAN / 11

worked on mainland Japan when he was 17-23 where he was again im-mersed in an Ikema speaking environment while living with Ikema speak-ing relatives.

These accounts suggest that the way one lived her/his life is more di-rectly relevant than the single factor of the grandparent’s influence in shap-ing one’s fluency of Ikema.

4.5. Gender

The four fluent speakers not raised by grandparents, discussed in the last section, are 47-55 years old, well below the commonly believed fluency cutoff age of 60. Interestingly, they are all men, which is in clear contrast with the two women (44 and 53 years old) discussed in 2.3. Those women were both raised by their grandmother but are not fluent in Ikema, suggest-ing that gender is another factor playing a role in shaping fluency, perhaps one which is more influential than merely living with grandparents.

Interestingly, we found that male students broke the rules and spoke Ikema at school more often than their female counterparts. As noted in sec-tion 4.4 above, some male students continued to use Ikema even at high school where Standard Japanese was the standard medium of communica-tion. Perhaps for this reason, some men in fact had trouble speaking Japa-nese when they moved to the main island of Okinawa and mainland Japan after high school where they had to use Japanese. In general such an ex-perience was not reported by female interviewees.

Several interviewees also noted that this type of gender difference still continues today. For instance, Mr. TS (47 years old), introduced in the last section, noted that he talks with his male agu ‘neighborhood peer group’¹⁷ in Ikema, but added that his female agu can’t speak Ikema. Several other interviewees reported similar experiences. In fact, the use of Ikema seems to have a clear social implication or stereotype, which is most vividly ex-pressed in a comment by Mr. F (also introduced in the last section): hogen da to josei rashiku nai ‘if you use the dialect, you are not feminine.’ It is particularly interesting to note that Mr. F, though relatively young (49 years old), himself is a fluent speaker.

Finally, we recently met three junior high school students in Sarahama who are surprisingly fluent in Ikema.¹⁸ Although the exact fluency of these students needs to be determined, they are all first year students in junior high school (12 or 13 years old). Interestingly, they are all male students and spend much of their time interacting with local fishermen, which again

• People in the Ikema community, starting from childhood, practice a number of social activi-ties with their agu ‘neighborhood peer group’.
• There are likely to be more students with some fluency of Ikema yet to be discovered.

• / SHOICHI IWASAKI & TSUYOSHI ONO

supports the roles of both gender and lifestyle in shaping the fluency of Ikema.

The faster switch among women than men from Ikema to the dominant language of Japanese fits with the observation that, in some societies, wom-en are “more sensitive to the social significance of social-class-related lin-guistic variables” (Trudgill 1974: 93). The situation of Ikema is reminiscent of Monchak (a minor Mongorian language) where “young Monchak men continue to speak the language in far greater numbers and with greater flu-ency that do females” (Harrison 2007: 97).

4.6. Social/Religious Events

The three Ikema speaking communities of Ikema, Nishihara, and Sarahama are known for having numerous religious and social events. These include both community-based official events and more smaller scale and/or private gatherings, including those organized by the associations for elderly people and the above-mentioned agu groups. In these events and gatherings, the use of Ikema is encouraged, or even required. Based both on the interviews which we conducted for the current study and on our own experience in participating in many of these events/gatherings, we feel that these occa-sions positively influence the maintenance of Ikema: participating in them and using Ikema mutually reinforce peoples’ identity as belonging to each community in particular and Ikema Minzoku ‘Ikema Race’ in general.

Obviously, not everyone in the community participates in these activi-ties, and people’s lifestyles undoubtedly play a role in their level of partici-pation, but it is our impression that a large portion of the community mem-bers do actively participate in them perhaps partly due to social obligation. It is worth pointing out that these activities seem to be much more actively practiced in the two new settlements of Nishihara and Sarahama than the original community of Ikema, and interestingly it is the impression of both community members and ourselves that Ikema is better maintained in these new settlements than on Ikema,¹⁹ suggesting that cultural and language maintenance might be influencing each other. It should be remembered here that the three junior high school students who we discussed in the last sec-tion are all from Sarahama, whose fluency in Ikema points to the possible connection between lifestyle, gender, and social/religious events.

• A higher maintenance level of Ikema in Sarahama may be related to the following three factors: 1) Sarahama is more geographically isolated than Ikema and Nishihara from the re-gional political and commercial center of Miyako Island because it is on Irabu Island which can only be accessed by sea. 2) Sarahama hosts a much larger population than Ikema and Nishi-hara. 3) Finally, Ikema speaking people in Sarahama are a linguistic majority on Irabu Island compared to a smaller number of speakers of Irabu, another dialect of Miyako.

IKEMA RYUKYUAN / 13

5. Conclusions

Our ultimate goal of the larger Ikema project is “to record the linguistic practices and traditions” of this speech community (Himmelmann 2002: 9). To this end, gathering information from community members at large was a necessary step. We believe our life narrative approach is a successful me-thod to achieve this goal.

It has provided us with insight into the influencing factors for the cur-rent situation surrounding Ikema, which include schooling, including kin-dergarten, exposure to television, primary caretakers, gender, and lifestyle. We have found that there are still some fluent speakers of Ikema who are in their late 40s and 50s, which goes against the common belief that fluent speakers are limited to 60 years old or above. These younger speakers are often inconspicuous and hard to find. Only through our narrative-based approach have we been able to obtain these findings.

In the process of engaging community members in conversation through life narratives, we have also been provided with a web of new and useful information for our continuing project, including (a) who we might want to work with in collecting and analyzing data in terms of who is flu-ent, available, and appropriate and (b) what kind of data we might want to collect which is relevant not just to linguists but to the community. In par-ticular, considering that our ultimate goal is to come up with a good repre-sentation of “the linguistic practices and traditions of a speech community” (Himmelmann 2002: 9), deciding on the type of data to collect and who to represent in it in discussion with the community is essential.

We can also summarize important additional (and somewhat unex-pected) outcomes of our life narrative approach as follows:

1. Better Rapport: In the process, we were able to establish a good rapport with community members, which we believe is a critical step toward fully documenting a language. This was achieved be-cause the life narrative approach created opportunities where in-terviewees and researchers actively interacted as conversational partners. Obviously, there is nothing surprising about this because we create a relationship with others by actually interacting with them.

1. Better Understanding: It seems that people whom we interviewed for the present project now have a better understanding of what we are interested in and what we are trying to accomplish, which is again crucial for our documentation project to become successful. Again, this was because our questions about interviewees’ past and current experiences were what non-linguists could easily relate to.

• / SHOICHI IWASAKI & TSUYOSHI ONO

We thus suggest that if one’s goal is to document a language, they should start a project by asking questions similar to ours rather than jumping into questions purely of linguistic nature (such as sound inventory, clause structure, etc.).

1. More Interest: Through talking with us, our interviewees seem to have developed a curiosity in their past and current experiences, particularly of their language use. Some of them have started dis-cussing their experience with their family members and friends and volunteering further information to us.

IKEMA RYUKYUAN / 15

References

Fija, B., M. Brenzinger, and P. Heinrich. 2009. The Ryukyus and the New, but En-dangered, Languages of Japan. The Asia-Pacific Journal: Japan Focus 3138.

Harrison, D. 2007. When Languages Die. Oxford: Oxford University Press. Himmelmann, N. 2002. Documentary and Descriptive Linguistics. Unpublished

manuscript.

Hirayama, T. (ed.) 1983. Ryukyu Miyako shotoo hoogen kiso goi no soogooteki kenkyuu. Tokyo: Oofuusha.

Hokama, S. 1971. Okinawa no gengoshi. Tokyo: Hoseidaigaku shuppankyoku.

Itani, Y. 2006. Okinawa no hogenfuda. Naha: Borderink.

Izuyama, A. 2002. A Study on the Grammar of Miyako Hirara Dialect of Luchuan. Grammatical Aspects of Endangered Dialects in Japan (1), ed. S. Sanada, 35-97. Tokyo: ELPR.

Kondo, K. (ed.) 2008. Hogenfuda. Tokyo: Shakaihyooronsha.

Krauss, M. 1992. The World’s Languages in Crisis. Language 68: 4-10.

Labov, W. 1984. Field Methods of the Project on Linguistic Change and Variation. Language in Use, eds. J. Baugh and J. Sherzer, 28-53. New Jersey: Pren-tice Hall.

Nakama, M. 1992. Ryukyu hogen no kosoo. Tokyo: Daiichishoboo.

Nevsky, N. 2005. Miyako hogen nooto. Miyako, Japan: Okinawaken Hirarashi kyooikuiinkai.

Shitaba, T. 1980. Okinawa miyokogo no goitaikei. Gekkan Gengo 9:1-9:12.

Trudgill, P. 1974. Sociolinguistics. London: Penguin.

UNESCO Ad Hoc Expert Group on Endangered Languages. 2003. Document sub-mitted to the International Expert Meeting on UNESCO Programme Safe-guarding of Endangered Languages. Paris, 10–12 March 2003.

Yakabi, O. 2007. “Nihongo” “nihonminzoku” no hensei de ikani honroo saretaka: Okinawa no kyoodoshika Shimabukuro Zenpatsu no kiseki. Taiwan, Kan-koku, Okinawa de nihongo wa nani o shita no ka, eds. C. Furukawa et al., 155-173. Tokyo: Sangensha.