Abstract
This paper analyses the loss of politeness markers across three generations in the Ryukyuan Islands of Japan. Honorific registers are first lost in endangered languages, and last speakers of an endangered language often state that they avoid using the language to semi-speakers because they do not use polite registers. We give an overview of language endangerment, analyse how individuals reflect on politeness markers, and how language loss and the awareness thereof manifest in language use. We find that the loss of politeness markers does not result in an affront to convention and social order. Japanese, the replacing language, is employed to construct social orders. Linguistically constructed orders in Japanese are different from those in Ryukyuan. In one of the two communities studied, politeness markers no longer function to construct social hierarchies but serve as a ‘we-code’ among locals. The experience of language endangerment works here as a social levelling mechanism.
1 Introduction
When studying pragmatics, it is assumed that the languages (and speakers) are vital and healthy. This is not the case in endangered languages, i.e., for one third of the world’s languages (Krauss, 1992). Endangered languages are not equally acquired between members of a speech community. These languages are also no longer adapted to societal, economic, or cultural changes, resulting in processes of language loss and attrition (Dorian, 1981). Using endangered languages in contemporary life necessitates putting linguistic forms to new uses.
In this paper, we look at politeness markers in Northern Ryukyuan from a comparative perspective. Honorific speech is among the first features that are lost in endangered languages (Schmid, 2011: 35), because the replacing language (in our case, Japanese varieties) starts spreading in so-called ‘higher domains’ where honorific speech is used (Sasse, 1992). Language endangerment results in reduced knowledge of language structures, starting with structures employed in polite registers (Tsitsipis, 2003). Last speakers of an endangered language often state that they avoid using the endangered language to younger semi-speakers, because these do not use polite language adequately (for Austronesian languages, see Florey, 2010). Avoidance of endangered languages towards younger speakers is also reported in the Ryukyus (Anderson, 2014). In this paper, we contrastively study the loss of politeness markers in Kumejima and Setouchi, two northern Ryukyuan communities in Japan.
Our discussion of these two communities is positioned at the junction of pragmatics and sociolinguistics. Holmes (2018: 11) summarizes the intersection between the two as sociolinguistics analysing “the linguistic resources available in speech communities” while pragmatics explains “how individuals use linguistic resources to produce and interpret meaning in interaction.” In this paper, we analyse how the social background (mainly locality, age, gender) impacts on the knowledge and use of politeness markers, i.e., linguistic elements used to show deference to the listener (House and Kasper, 1981).
We are interested in emic rationalization of the loss of politeness markers and in the pragmatic consequences that result thereof. Polite registers are fundamental for constructing social relations. In endangered language communities, generations of speakers have different access to and knowledge of politeness markers. We therefore engage in a study of variation of local linguistic practices that go “beyond the geographical diversity of culture” (Matsumoto, 2021: 14) by comparing two separate linguacultures (Chen et al., 2021).
We have chosen a comparative approach (Kumejima and Setouchi in the northern Ryukyuan Archipelago). Three language varieties are involved in both cases, and we distinguish these by fonts. Varieties of Northern Ryukyuan languages (Setouchi and Kumejima) are set in italics and long vowels are indicated by repeating the vowel in question twice. Standard Japanese, one of the replacing languages, is set in roman fonts (‘non-italics’). Ryukyuan Substrate Japanese, a local variety of Japanese that emerged in the course of language shift, is set in italics + bold fonts. The Kumejima variety (henceforth Kumejima) is part of South-Central Okinawan (ISO code: ryu). The Setouchi variety (henceforth Setouchi) is part of Southern Amami-Ōshima (ISO code: ams). There are altogether six Ryukyuan languages which comprise about 700 different local varieties. There exists no standard language for either Okinawan or Amamian. Both are so-called ‘unroofed spoken languages’, that is, they are constituted (solely) of various local varieties. Ryukyuan languages and Japanese are sister languages. Together, they form the Japonic language family (Shimoji and Pellard, 2010). All Ryukyuan languages are endangered (Moseley, 2009).
2 Language Shift and Loss in Kumejima and Setouchi
In this section we present the background of language shift and language loss in the two communities under study.
2.1 Kumejima
Kumejima is an island with 7,800 inhabitants. It is located approximately 100 km west of Okinawa Island. The form of South-Central Okinawan spoken on Kumejima shows considerable geographical variation, yet all varieties retain mutual intelligibility with each other as well as with other varieties of South-Central Okinawan. Other varieties of South-Central Okinawan are spoken in hamlets that were established by settlers from other parts of the Okinawan language sphere like Itoman, Naha and Tonaki Island in the late nineteenth century. These migrant varieties are still recognisable as such even though they have been influenced considerably by the Kumejima variety. The differences between the varieties of individual settlements are becoming less clear among younger speakers.
In Kumejima, ‘honorific language’ is referred to as uyameekutuba (literally, respect language) and ‘to use honorific language’ is referred to either as uyameekutuba sun (to do/use respect language) or as uu-fuu sun (to do/use uu and fuu) referring to the honorific language version of the affirmative word uu (yes), and the responsive word fuu (yes?). Kumejima speakers are aware that their language is part of Uchinaaguchi, the endonym for the Okinawan language in the broadest sense, including the varieties of northern Okinawa. Local varieties on Kumejima are referred to by place name plus either -munui or -kutuba. For example, the language of Janadō hamlet, Yararoo in Okinawan, is referred to as either Yararoo-munui or Yararoo-kutuba. Both munui and kutuba mean ‘language’ or ‘word’ when used in isolation, but in compounds with place names, munui has slightly more the connotation of ‘dialect.’
The language shift towards Japanese on Kumejima is taking place at a slightly slower pace than has been reported for other places in the Okinawan-speaking area (Heinrich, 2007: 6). Based on recent observations in the field, men born in the late 1960s form the last cohort of active native speakers that use the language amongst themselves. There are isolated cases of younger native speakers, some born as late as in the early 1980s.
2.2 Setouchi
Like all other Ryukyuan languages, the Setouchi variety of Southern Amami- Ōshima has been steadily replaced by Japanese and Ryukyuan Substrate Japanese. Setouchi’s population is 9,300, including non-speaker inhabitants. The Setouchi variety is spoken on four islands in the Amami Islands (Kakeromajima, Yoroshima, Ukeshima, and in the southern region of Amami-Ōshima). These regions fall administratively under Kagoshima Prefecture, but due to their shared history as part of the Ryukyu Kingdom, the Amami Islands are culturally more akin to Okinawa than to mainland Japan. The lines separating local varieties of Southern Amami-Ōshima are unclear, and also within Setouchi there is significant variation, particularly across islands. For this study, all consultants speak Setouchi as used on Amami-Ōshima Island.1
Southern Amami-Ōshima was reported to have approximately 1,800 speakers in 2004 (Eberhard, Simons and Fennig, 2019), but there are no estimates on speaker numbers for Setouchi. Intergenerational transmission for the language has been broken for at least two generations, and all inhabitants on the Amami Islands are fluent in Japanese as a second or first language. Most remaining fluent speakers of Setouchi are now in their seventies or older (Maeda, 2014). They rarely use formal registers, honorific, or humble speech in spontaneous discourse. Semi-speakers (born from the mid-30s to the mid-50s) do usually not have command of formal registers. The only possible consultants for honorific language are therefore the elderly. Younger speakers are likely to use Standard Japanese or a local variety of Japanese (Kagoshima Japanese dialect, Ryukyuan Substrate Japanese), when speaking to their elders (Anderson, 2015: 482). The nomenclature for ‘politeness’ or ‘polite speech’ in Setouchi is yawarasa.
3 Honorifics and Social Hierarchies
In this section, we discuss the two honorific language systems of Kumeijima and Setouchi and also how they are employed to construct social hierarchies. Just like in Japanese, honorifics in Ryukyuan languages are systematically encoded in grammar. In many situations, the use of honorifics has been mandatory before language shift, and elderly full speakers remember these uses and norms. Incomplete knowledge of honorifics by younger speakers is a problem. Florian Coulmas (2005: 93–94) has a point in writing that honorifics “are not frills that can be dispensed with at will, because there are no encounters that take place in a social vacuum. […] Speaking without honorifics is incompetent or childish at best, and otherwise an affront to convention and social order.” In what follows, we perceive politeness as a dimension of social practice, in which linguistic behaviour is organized in a way to not offend the expectations of others and to conform to norms of communication. Notions of politeness are contested, and their study requires emic assessments of polite behaviour (Eelen, 2001). We concur with Watts (2003: 11) that emic notions of politeness represent a “struggle over the reproduction and reconstruction of values of socially acceptable and socially unacceptable behaviour.” As an effect, we regard politeness markers not as ‘being polite in themselves’ but as resources to be employed in the discursive construction of behaviour that is perceived as polite. Given these perspectives on politeness and on language shift and loss in Kumejima and Setouchi, we can expect that incomplete command of politeness markers amounts to pragmatic problems. Before exploring these, let us first delineate and exemplifying the honorific language systems of both communities. These outlines are derived from the fieldworks of Gijs van der Lubbe and Martha Tsutsui Billins.
3.1 Kumejima
Honorific language in Kumejima consists of both morphosyntactic and lexical items. When applying Shibatani’s (2006: 381–384) classification, Kumejima honorifics can be sorted into the following five categories:
(I) Honorific titles can be added to a name and also professions or kinship terms that are considered to have a high social status. Most of these titles have been replaced by Japanese or Ryukyuan Substrate Japanese titles. For instance, men that are significantly older than the speaker can be addressed as otō (dad). Otō is the Ryukyuan Substrate Japanese equivalent of Japanese otōsan. It has replaced the Kumejima equivalent suu. Similarly, Kumejima yacchii (older brother) was replaced by Ryukyuan Substrate Japanese nīnī (onīsan in Japanese). Men of the youngest generation of Kumejima speaker indicate that they prefer to use shiija (older sibling) when addressing men slightly older than themselves, because using yacchii would make the addressee feel old. Older speakers indicate that shiija is exclusively used to refer to third persons, and that it is not used as a form of address. A possible explanation for younger speakers of Kumejima not using nīnī to address older men is that nīnī is also used to address men or boys younger than the speaker. In this way, nīnī has also replaced Kumejima niisee (young man).
(II) Pronouns: There is a noticeable east-west difference in the use of second person pronouns in Kumejima. The western part has a two-way honorific distinction between the honorific unzu and the non-honorific yaruu or yaa. In the east, unzu is used as a high-level honorific, but yaruu and yaa are used to reflect different honorific levels. Yaruu is thereby used as a mid-level honorific towards older siblings and people who are slightly older than the speaker but with whom the speaker is on friendly terms, or towards people who are younger than the speaker but to whom the speaker still wants to convey a certain level of respect. Yaa, on the other hand, is used towards people who are younger than the speaker or friends of the same age. The east-west difference in the use of yaruu can cause misunderstandings.
(III) Subject honorifics is understood as “a way of showing deference toward the referent of a subject nominal, namely, the actor” (Shibatani, 2006: 282). Kumejima uses the honorific suffix -misen/-nsen that attaches to verbs (V) and property concept verbs (PCV). The forms -misen/-nsen can also be attached to the copula (COP), where it expresses deference to a nominal predicate. A special subset of honorific verbs also exists. For instance, usagayun is used as an honorific for both kamin (to eat) and numin (to drink), and mensen is used as an honorific for itsun (to go), tsun (to come) and un (to exist). Usagayun can be further honorified by attaching the honorific suffix -misen/-nsen to it.
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
(IV) Humbling forms in Kumejima consist of a set of verbs that could also be referred to as object honorifics as they “express deference toward the referents of non-subject nominals by humbling the actor’s action directed toward them” (Shibatani, 2006: 383). For instance, unnukiyun is used as a humble form of ayun (to say) and katayun (to tell). Even though speakers of all ages recognize these words, no instances of their use were found in our natural conversation corpus of Kumejima.
(V) Addressee honorifics show deference to the listener, regardless of the subject, or nominal predicate of the sentence. Deference to the addressee is expressed by means of the suffix -bin, that attaches to verbs, property concept verbs and the copula.
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
3.2 Setouchi
Today, Amami honorifics can be divided into three larger categories, referent honorifics and two types of addressee honorifics. Shibatani (2006) calls these referent-controlled honorification and addressee-controlled honorification, respectively. Addressee honorifics and referent honorifics grammatically code deference towards the addressee or the referent (Comrie, 1976). There are two types of referent honorifics in Amami – subject honorifics (respectful forms which raise the status of the referent in the subject argument) and non-subject honorifics (humble forms), which lower the referent’s status. In the case of humble forms, the speaker or someone in the speaker’s in-group (e.g., a family member) is the referent in the subject position in the utterance. It is likely that the honorific system in Amami was once more complex, but due to stylistic shrinkage (in the sense of Dorian, 1981), knowledge and use of Amami honorifics within the community shows only three types, according to metalinguistic judgements of speakers.2
(I) In the collection of data addressee honorifics (polite speech) was most commonly manifested with the -yaon suffix. Fluent speakers prescribe that this form is used with elders or with people you do not know well (outgroup members) or with superiors. Based on forms used in experiments and in interviews, -yaon is the second most well-known form after the plain form (yun). Since -yaon does not mark the highest level of politeness, it has likely retained its place in usage amongst speakers to a certain degree (though the use has decreased more than that of the plain form verbs). According to two consultants, this form most closely corresponds to the desu/-masu form in Standard Japanese and can be used in most situations where the plain form would be too casual. The word final -n makes the verb non-past affirmative whilst -an creates non-past negative forms (Example 1).
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
Though most commonly collected with the -yaon suffix (Example 2), honorifics in this level can also be conjugated to end in -yaoddo. The -do ending add emphasis to an utterance, for example, ikyaoddo (go!) or aryaoddo (be!).
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
(II) In the case of referent honorifics, there is the case of subject honorifics (respectful speech). Both humble and respectful forms can be used together in the same situations. Conventionally, respectful forms are used when speaking about superiors. A conversation between a speaker and their superior might include both humble and respectful language (from the subordinate speaker). For example, before Amami was lost in the public domains, you would expect to hear these forms when serving a customer. If the shopkeeper is the speaker, they would use humble forms when referring to their own actions/themselves (thus lowering themselves), and the respectful forms when referring to the customer (thus raising the customer’s status). These forms are not used when speaking about oneself or something one has done themselves. An expression still widely used in Amami today is the respectful form (umore) by shopkeepers who greet customers. This form is generally characterized by the umo auxiliary verb (Example 3).
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
One interesting aspect here is that ‘go’ / ‘come’ / ‘exist’ are expressed with simply the auxiliary verb (umore), and other verbs contain a separate stem (the example ‘to see’ is expressed with michiiumore (verb stem of see + honorific/respectful aux verb).
(III) The second type of referent honorifics are non-subject honorifics (humble speech). It is used when describing one’s actions or the actions of a person in an in-group to others of a higher status. For example, a waiter would use humble language when referring to themselves or their actions to a customer. In Japanese, humble language implies that the speaker’s actions are in order to assist or benefit the listener (Pizziconi, 2011). These forms have been much more difficult to find in the Setouchi variety. Based on the collected data, they are scantly used outside of a few commonly-used lexical chunks. Humble language in Amami tends to have distinct lexical root verbs from polite verbs (though not always) and the -yaoro suffix attached to the verb (Example 4).
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
Some verbs in this category may have undergone some sound assimilation, such as ueshaoro (give). Commonly used/known phrases, such as arigassama ryoota (thank you) and kyaryoo (hello), fall into this category of honorific (Example 5).
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
4 Contemporary Attitudes towards Honorific Speech
In this section we investigate how individuals reflect on politeness markers in their endangered language variety. The views on the application of honorifics in Kumejima are based on interviews and informal conversations with about 20 speakers of Kumejima between the ages 31 and 104 during the period 2011–2019. The data was collected by Gijs van der Lubbe. Elicitation fieldwork with the goal of morphosyntactic description was carried out with a speaker from Tomari-Janadō, Nakazato (female, born 1936). The spontaneous conversation data was collected during informal meetings with the participants in the period 2016–2019. Additional data for Kumejima was from Van der Lubbe’s fieldnotes on unrecorded conversations.3
Data for Setouchi was collected by Martha Tsutsui Billins via fieldwork during a period of four months, split between two field trips in 2017 and 2018. Spontaneous conversational data was collected by consultants themselves using an audio recorder. Semi-structured interview data and observed communicative event data were also collected by her. 75% of consultants were aged in their 70s or older, while 25% were middle-aged, in their 40s, 50s and 60s.4
4.1 Kumejima
All Kumejima varieties are mutually intelligible with South-Central Okinawan. Sui-kutuba, the language of Shuri, which was the seat of government of the Ryukyu Kingdom, functions as a de facto standard towards which the use of uyameekutuba is oriented throughout the Okinawan language sphere (Ishihara et al., 2019). As an effect, conversations about honorific use in Okinawan imply references to Shuri. Consider the following statement by one of our consultants from Nakazato (female, 80s) when asked whether she uses a certain subject honorific construction (Example 6).
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
Shuri’s societal structure gave rise to a complex system of honorifics, and as a result, other varieties tend to be characterised as ‘lacking honorifics’ or ‘uncouth’ by those speakers themselves.
Speakers note differences between the different varieties of Kumejima also with regard to honorific use (e.g., the different use of second person pronoun use mentioned above). According to popular belief, the main dialectal divisions in Kumejima are between the former municipalities of Gushikawa in the west and Nakazato in the east. The fact that Gushikawa speakers are reputed to use the second person honorific unzu in more situations may contribute to the Gushikawa varieties as having a slightly higher status than the Nakazato varieties. A speaker (male, 50s) from Nakazato made the following statement upon being asked whether he uses the second person intermediate honorific yaruu (Example 7).
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
The fact that a speaker from Nakazato refers to the use of yaruu as ‘actually rather impolite’ may point to the higher status of Gushikawa varieties.
There is a marked difference between Kumejima and Japanese in the socio-pragmatic factors that influence the application of honorifics. Whereas ingroupness plays an important role in not using honorifics in Japanese (Brown, 2008: 381), seniority seems to play a more important role than in-group out-group distinctions in Kumejima. All consultants indicated using or having used honorifics towards their own uncles, aunts, and grandparents. Speakers are aware of these differences between Kumejima and Japanese. A consultant (female, 80s) indicated that her grandchildren do not speak Okinawan, however if they had, she would have wanted them to use honorific language to her.
Speakers are also aware of the influence that the shift to Japanese has had on their acquisition of Kumejima honorifics. It is often mentioned as an aspect in which younger speakers feel their Kumejima skills fall short. A speaker (female, 80s) mentioned the following upon being asked whether she uses honorifics often (Example 8).
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
The lack of confidence in honorific language use is that it can stop them from using Kumejima with people towards whom they ought to use honorifics, as this speaker (male, 50s) reports (Example 9).
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
Some younger speakers choose to use Kumejima regardless of their lack of confidence, and some of the older speakers (who are self-professed users of honorifics) seem to encourage that. Consider the following Example (10) from a conversation between a male, 40s, from Zenda (A) and another male, 60s, from Maja (B).
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
As we can see from these examples, metapragmatic accounts on Kumejima crucially involve two other language varieties, Japanese varieties and the Okinawan Shuri variety. Choice of and variation in language (Kumejima, Shuri, Japanese) is not simply an account about diversity or locality, but it is first and foremost hierarchical. In the minds of these speakers, this hierarchical ordering of varieties plays a role in reinforcing and legitimizing language shift away from Kumejima.
4.2 Setouchi
Like other speakers of Ryukyuan, Setouchi speakers have been stigmatized for using their languages in the past. During the Meiji period (1868–1912), local languages were heavily and systematically discouraged. The classic example is schoolchildren being singled out and humiliated for using their local languages by being forced to wear a dialect tag (hōgen-fuda) around their neck. Against the backdrop of linguistic discrimination, Setouchi has undergone linguistic attrition. Many semi-speakers do not draw on Setouchi honorifics. It is difficult to discern whether these speakers lack actual fluency or ability, or if they are more lacking in confidence to produce honorifics in front of more fluent seniors. For use of addressee and referent honorifics, speakers report that honorifics should be used with elders, and plain form should be used with equal-aged peers, friends, and younger people.
Many speakers can produce the casual or plain registers, but struggle or are unable to draw on honorific registers. Similarly, in semi-structured interviews a common comment from older people across the Ryukyus is ‘young people these days cannot speak local languages correctly anymore! They are not polite enough!’ Younger people in Setouchi admit that they do not like to speak Setouchi to their elders. They are embarrassed, lack confidence or are worried about being scolded (Example 11). We can see in this example that notions of correctness and appropriateness are relevant for the language choices of Setouchi speakers.
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
This semi-structured interview with HS (female, 45 years old a young semi-speaker of Setouchi) highlights two language ideological notions about Setouchi. First, it shows that younger speakers are nervous about their ability to use honorific registers. Secondly, it shows that Setouchi has the effect of signalling to others that they are in one’s inner circle. Y (male, 70s), a fluent Setouchi speaker, discussed accurateness and care in communication when asked for his views on politeness (Example 12):
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
Y thus sees politeness being equivalent to accurateness (or correctness) and these qualities are at odds with speakers’ diminished abilities and confidence to draw on honorific Setouchi registers when convention requires them to be polite.
In these examples we see how language shift and attrition results in linguistic insecurity on the parts of Setouchi semi-speakers. In order to deal with this linguistic insecurity, semi-speakers shift to language varieties where such insecurity is absent (Japanese). We also learn that politeness is not simply about the language system or about adequately addressing elders. Its use (correct or incorrect) serves as an index of the speaker’s character. In other words, politeness it is morally loaded, and this also plays a role in discouraging semi-speaker to try using polite registers.
5 Politeness Markers in Communities Undergoing Language Shift
In this section, we discuss data that was collected during period of fieldworks for language documentation. It is mostly of ethnographic nature, and it was at times also coincidentally recorded during fieldwork. While language documentation of endangered language is predominantly concerned with the endangered language only, a focus on politeness markers requires attention to both the endangered Ryukyuan varieties and the replacing Japanese varieties. In our data we find fixed and ritual use of set expressions but also cases of code-switching and mixed uses.
5.1 Kumejima
This section analyses three conversations where Kumejima and Okinawan Substrate Japanese was spoken. None of the conversations were recorded with the intent to document natural conversation, and speakers were not specifically asked to use Kumejima. These three conversations were chosen as we regard them as typical of conversations between residents of Kumejima who are able to speak the local variety.
The first conversation is between our consultant K (female, 80s) and P (a postman, male, 60s). It is the only recorded conversation where the consultants did not know each other. Honorifics are in bold type (Example 13).
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
This conversation represents what one would expect from two speakers of Kumejima. P, the postman, greets using the formal Shuri Okinawan greeting chaabira sai, and K, who does not see him, replies in informal Kumejima (turn 4). After some initial formalities in Japanese (turns 5–8), the younger consultant in this conversation P takes the initiative to speak Kumejima. From that point on, Japanese is used only for limited and isolated lexical items. Subject honorifics and addressee honorifics are used throughout the conversation.
The second conversation (Example 14) was between our consultant K (female, 80s) and O (female, 100s), an acquaintance of hers. Van der Lubbe was also present, as well as a family member of O (also female, 80s). Almost the entire conversation between K and O was in Kumejima, with only some isolated Japanese lexical items interspersed. One would expect K to consistently use honorifics towards O, because of a twenty-year age difference. However, this was not consistently the case. K kept using addressee honorifics inconsistently for about the half the time, while using plain forms for the other half. For instance, K does not use addressee honorifics in the turns 1 and 6, but she does in turn 2 and 3. When addressing O, K uses the Okinawan Substrate Japanese kinship term obā (grandma).
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
K does use honorifics consistently when she refers to actions by O. Consider the following fragment where K points at the rotary dial telephone in the house of O (Example 15).
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
Also, K consistently uses subject honorifics when referring to the actions of O (eating, drinking, coming, going). Consider the following fragment (Example 16), where K uses the subject honorific usagayun (to eat/drink).
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
The third conversation (Example 17) is between S (male, 60s) and T (male, 80s). Again, there is a twenty-year age difference between the two speakers. One would expect S to use honorifics towards T. S inconsistently uses addressee honorifics in some instances, however the only consistent use of honorifics by S consisted of him using the honorific second person pronoun unzu. Another way in which S addressed T was with the Okinawan Substrate Japanese kinship term otō (dad). Consider the following fragment, where S uses unzu but does not use an honorific verb doe ‘going’ when T is leaving.
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
While ‘eating/drinking’ is consistently not honorified, there is one instance, S talks to L (Gijs van der Lubbe) and uses the subject honorific usagayun (eat/drink) when referring to T in the third person (Example 18).
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
In another conversation, referred to in (9), between Y (male, 40s), S (male, 60s) and (I) (male, 60s), the age difference would again suggest honorific use by Y. However, according to the fieldnotes of Van der Lubbe, no honorifics could be attested except for the sporadic use of unzu. All speakers indicated that their knowledge of honorific language in Kumejima was limited.
Taking the four conversations into consideration, a picture emerges of inconsistent use of honorifics. Also, the younger the speakers are, the fewer honorifics they use. Only the youngest speaker, Y (male, 40s), admitted that the only honorific he ever used was the second person pronoun unzu. Even though there were speakers that said they would not use Kumejima towards elders because of their lack of confidence in honorific use, no instances were found of this actually happening. The fact that Kumejima was the principal language used in the four conversations, may point to a strong language loyalty of Kumejima speakers despite lack of knowledge and confidence in using honorific language.
Younger speakers of Kumejima appear to have internalised that they are lacking sufficient proficiency in honorific language (part 4.1), but they produce more honorific language that could be expected from their own metapragmatic accounts. This inconsistency between metapragmatics and actual language behaviour could be explained by the former being a pan-Ryukyuan discourse (‘younger speakers no longer know how to speak politely!’), while the local sociolinguistic situation in Kumejima does not fully correspond to this characterization. This discrepancy calls for more contrastive research.
5.2 Setouchi
In this section we discuss data that was recorded by employees working at a retirement home in Setouchi Town. The younger employees (Setouchi semi-speakers) overwhelmingly favoured Japanese to communicate with the retirement home residents, all of whom are fluent Setouchi speakers. T (male, 54), an employee at the retirement home, recorded himself with residents (elderly speakers). T was instructed to turn on the recorder and converse freely with the residents on any topic he chooses. In this setting, polite speech would be expected, as the residents are T’s elders and because he is at work and the residents are considered ‘clients’ or ‘customers’ at the retirement home.
In Example (19), retirement home worker T (male, 54) recorded himself speaking to residents OH (female, 94) and IM (female, 82), who are fluent Setouchi speakers. T is asking the two women about their lives, and he is using formal Japanese (indicated by underlining), despite the fact that he is a Setouchi speaker and could theoretically speak Setouchi to the two women if he chose to.
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
In this exchange, we can clearly see that T is using formal, though not honorific Japanese. T-san uses the formal desu/-masu form when he asks (OH) where she was born, when he asks about her husband’s hometown, and he also uses the polite suffix -san to refer to her husband in Japanese.
In Example (20) T interviews two women who live at the retirement home: TF1 (female, 93) and IM2 (female, 95). They are both from Nishikomi Village in Setouchi. Again, in this conversation, we see T using Japanese for the majority of the exchange, rather than Setouchi. Language in Amami Substrate Japanese (called ton-futsūgo in Amami) are indicated in bold.
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
In this session, T is primarily in Japanese, except for one word, nankya, which is the polite plural second person pronoun in Setouchi. Besides this, he uses only Japanese (though not polite Japanese), mixed with some Amami Substrate Japanese. When T was asked about this session, he said that he knows these speakers very well, and they are more like his friends, so sometimes he will use a bit of local language with them. With residents he does not know well or who are new in the retirement home, he reports to always use Standard Japanese. He also reports that he speaks Setouchi to his peers. This indicates that polite Setouchi (in this case the polite plural second pronoun) may be used to show solidarity or friendship, as T demonstrated in this example.
In this next Example (21), T is at work speaking to resident (IM2) (female, 95). He turned off the recorder suddenly after asking (IM2) to wait a moment, likely being called away for some work-related tasks. T uses Standard Japanese to speak to IM2, despite the fact that they are both Amami speakers and replying both in Setouchi and Japanese (formal Japanese is again indicated by underlining).
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
In these sessions, we can observe that the younger care-takers use Japanese to speak to the elderly residents, despite both being Setouchi speakers.
We can see that as domains of Setouchi use have become limited. Setouchi has undergone “stylistic shrinkage”, a process that is common in endangered languages (Kuipers, 1998; Gal, 1979). This stylistic shrinkage, which has been reported across the Ryukyuan Archipelago, has particularly affected formal registers (Anderson, 2014: 123). As an effect, knowledge of the full range of Setouchi registers is decreasing. Stylistic shrinking may eventually develop into “monostylism” (Gal, 1979), where speakers can only use the language in casual domains and contexts. While many speakers cannot draw on polite speech in Setouchi, they can use the replacing language (in our case Japanese) whenever such a register is required (see de Cillia et al., 1998: 29). The data in Setouchi shows a complementary assignment of social functions between the languages involved. When speakers need to use honorific language or need to be polite, they use Japanese. Setouchi honorifics, however, are still present in small tokens, as in T’s use of the polite pronoun nankya.
Formulaic phrases and elements such as nankya are referred to as ‘lexical touchstones’ by Tsutsui Billins (2021).5 They are known and understood also by speakers with little fluency in Setouchi. Many of these lexical touchstones represent a rich source of honorific morphemes. They include greetings and apologies, such as umore (welcome), mishore (please eat), kyaoryo (hello), sumiyoran (sorry), ugaminshooran (welcome, hello). Considering that most speakers cannot use honorifics productively in spontaneous speech, it is noteworthy that at the same time, these phrases are commonly found in the community’s repertoire and also in the linguistic landscape in Setouchi.
Honorifics are not inherently polite, and they signal many other meanings. Additionally, in the endangered language context, honorifics may not be used as expected in languages with full vitality. The Setouchi speech community exhibits variation regarding honorifics. For example, consider the case where consultant YG (female, 98) is using the imperative addressee honorific tabore to ask her neighbour K (male, 45) to cut her grass (Example 22).
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
According to prescribed honorific use, this use of the tabore honorific is unusual. YG is much older than K, so the expected phrase would be plain (Example 23):
Citation: Contrastive Pragmatics 3, 2 (2022) ; 10.1163/26660393-bja10030
Another session that demonstrates unexpected Setouchi honorific use is found below, where OK (male, 65), a fluent Setouchi speaker uses the honorific imperative whilst coaxing his baby granddaughter to eat. OK uses imperative addressee mishore (honorific IMP; please eat) with his baby granddaughter to encourage her to eat her dinner whilst he is spoon-feeding her. When Tsutsui Billins asked OK about it, he was surprised. His first response was ‘did I say that?’ After thinking about it, he said that he used the honorific to be kind, and that if he were speaking Japanese in that situation, he would have said ‘tabenasai’ (eat!) but in a nice way – the tone of voice is important, he says. This is interesting because it implies that the baby register in Setouchi has disappeared due to stylistic shrinkage. It also implies that Setouchi honorifics are encoding different meanings besides deference (as it is unexpected to use an honorific to a baby).
Besides variation in actual language use, speakers’ opinions on the meaning of honorifics are also varied. Based on semi-structured interviews, for less-fluent speakers the difference between the imperative forms shore and tabore appear to be opaque. This is most likely due to imperfect acquisition of these forms (in the sense of Palosaari and Campbell, 2010: 111). Very old and highly fluent speakers are aware that tabore is the politest form. We see it, for example, being employed in religious rituals when addressing gods. However, many speakers Tsutsui Billins interviewed did not have a distinction for the two forms and her that they could be used interchangeably. It is possible that there is some re-grammaticalization of auxiliary verbs (e.g., tabore) occurring here. Younger speakers use tabore with their equals and their inferiors (rather than reserving it for elders and superiors). Endangered language speakers may be adapting forms that remain in their repertoire to be used in a new context. There is also data that suggests that some forms are merging in the process of language loss. For example, speakers that were studied in a local beauty parlour reported that shiitabore and shichitabore (“please do”, or shite kudasai in Standard Japanese) can be used interchangeably, while Prof. Maeda (personal communication), a specialist of the Setouchi variety, and other more fluent speakers maintain that tabore is the politest form.
The sociolinguistic and pragmatic situation in Setouchi is different to that of Kumejima, despite the fact that we find similar metapragmatic comments on politeness on the endangered language varieties (part 4.2). In order to compensate for incomplete knowledge about polite registers, younger generations in Setouchi shift to Japanese varieties or employ lexical touchstones. The latter provide for resources to (also) express politeness. One reason for the difference between Kumejima, where we found more polite registers than could be expected, is that Okinawan language varieties enjoy in general higher language vitality than the Amami language varieties to which Setouchi belongs (Niinaga, Ishihara and Nishioka, 2014). More research is required in future to understand whether declining language vitality in Kumejima will result in patterns similar to those in Setouchi, or whether the two communities compensate for language attrition in different ways.
6 Conclusions
Community members in both Kumejima and Setouchi belief that adequate levels of politeness are a problem due to language shift and language attrition among the middle and the young generation. While this is noted and regretted in both communities, it does not amount to “an affront to convention and social order” as one could expect by following Coulmas (2005: 93–94). In multilingual communities such as Kumejima and in Setouchi, community members simply employ the replacing language (Japanese or Ryukyuan Substrate Japanese) to linguistically construct social orders. Note, however, that linguistically constructed hierarchies in Japanese are different from those in Ryukyuan. We have seen that seniority is more important than in-group and out-group distinctions in Ryukyuan. In Ryukyuan varieties, politeness is also required to address older family members. Language shift from Ryukyuan varieties to Japanese varieties imply therefore cultural and social change. The remaining knowledge of Ryukyuan varieties is used to mitigate such change, and this includes putting ‘old forms to new uses’. Such strategies underline that honorific marker are not in themselves ‘polite’ but that they simply constitute ‘playing material’ to construct social meaning, including honorification.
We can also conclude from our analysis that language shift and loss results in more ‘leeway’ to stray from conventions and norms as they once (reportedly) existed. This has implications how politeness markers are concretely employed to construct social relations in linguistic interactions. Language endangerment and its pragmatic consequences are acknowledged and accepted in both communities (part 4). This implies that norms in endangered languages are changing, and they do so differently between the two communities we studied here (part 5). Language shift, loss and attrition is more advanced in Setouchi. The use of Japanese, and with that the acceptance of social and cultural change, is more widely spread there. In Kumejima, language vitality is higher than expected, but the differences we find in accepting different uses of honorifics (east-west, between generations) can also dynamically be interpreted as harbingers of linguistic, social, and cultural change. In Setouchi, on the other hand, we find situations where token knowledge (lexical touchstones) of the endangered language variety plays a new and potentially important pragmatic role. Lexical touchstones often include politeness markers, but these are now used differently and are put to new functions. They no longer primarily function as means to denote and reproduce Ryukyuan social hierarchies but serve to create a ‘we-code’ among primarily Japanese-speaking locals. The experience of language endangerment works here as a social levelling mechanism. This mechanism also deserves more attention in future research in contrastive pragmatics and the sociolinguistics of endangered languages.
References
Anderson, Mark. 2014. Language shift and language loss. In: Mark Anderson, and Patrick Heinrich (eds.), Language Crisis in the Ryukyus. Newcastle upon Tyne: Cambridge Scholars Publishing, 103–139.
Anderson, Mark. 2015. Substrate-influenced Japanese and code-switching. In: Patrick Heinrich, Shinsho Miyara, and Michinori Shimoji (eds.), The Handbook of the Ryukyuan Languages: History, Structure, and Use. Boston: de Gruyter Mouton, 481–509.
Brown, Lucien. 2008. Contrasts between Korean and Japanese honorifics. Rivista degli studi orientali 81(1): 369–385.
Chen, Rong, Juliane House, and Dániel Z. Kádár. 2021. Ritual frame indicating expressions: An interview with Juliane House and Dániel Z. Kádár. Contrastive Pragmatics 2(2): 259–268.
Comrie, Bernard. 1976. Aspect. Cambridge: Cambridge University Press.
Coulmas, Florian. 2005. Sociolinguistics. The Study of Speakers’ Choices. Cambridge: Cambridge University Press.
de Cillia, Rudolf, Florian Menz, Wolfgang U. Dressler, and Petra Cech. 1998. Linguistic minorities in Austria. In: Donald Peckham, and Christina Bratt Paulston (eds.), Linguistic Minorities in Central and Eastern Europe. Clevedon: Multilingual Matters, 18–36.
Dorian, Nancy. 1981. Language Death: The Life Circle of a Scottish Gaelic Dialect. Philadelphia: University of Pennsylvania Press.
Eberhard, David M., Simons, Gary F. and Fennig, Charles D. (eds.), 2019. Ethnologue: Languages of the World (22nd edition). Dallas: SIL International.
Eelen, Gino. 2001. A Critique of Politeness Theory. Manchester: St. Jerome Publishing.
Florey, Margaret. 2010. Endangered Languages of Austronesia. Oxford: Oxford University Press.
Gal, Susan. 1979. Language Shift. Social Determinants of Linguistic Change in Bilingual Austria. New York: Academic Press.
Heinrich, Patrick. 2007. Look who’s talking: Language choices in the Ryukyu Islands. LAUD Working Papers (691): 1–21.
Holmes, Janet. 2018. Sociolinguistics vs pragmatics: Where does the boundary lie? In: Cornelia Ilie, and Neal R. Norrick (eds.), Pragmatics and its Interfaces. Amsterdam: John Benjamins, 11–32.
House, Juliane, and Gabriele Kasper. 1981. Politeness markers in English and German. In: Florian Coulmas (ed.), Conversational Routine. The Hague: Mouton, 157–185.
Ishihara, Masahide. 2014. Language vitality and endangerment in the Ryukyus. In: Mark Anderson, and Patrick Heinrich (eds.), Language Crisis in the Ryukyus. Newcastle upon Tyne: Cambridge Scholars Publishing, 140–168.
Ishihara, Masahide, Katsuyuki Miyahira, Gijs van der Lubbe, and Patrick Heinrich. 2019. Ryukyuan sociolinguistics. In: Patrick Heinrich, and Yumiko Ohara, (eds.), Routledge Handbook of Japanese Sociolinguistics. London: Routledge, 25–42.
Krauss, Michael. 1992. The world’s languages in crisis. Language 68: 4–10.
Kuipers, Joel C. 1998. Language, Identity and Marginality in Indonesia: The Changing Nature of Ritual Speech on the Island of Sumba. Cambridge: Cambridge University Press.
Maeda, Tatsuro. 2014. Amamian language life: Experiences of migration and dialect correction. In: Mark Anderson, and Patrick Heinrich (eds.), Language Crisis in the Ryukyus. Newcastle upon Tyne: Cambridge Scholars Publishing, 236–254.
Matsumoto, Yoshiko. 2021. Pragmatics of understanding: Centrality of the local cases from Japanese discourse and Alzheimer’s interaction. Contrastive Pragmatics 2(1): 24–51.
Moseley, Christopher (ed.), 2009. Atlas of the World’s Languages in Danger (3rd edition). Paris: UNESCO.
Niinaga, Yūto, Masahide Ishihara, and Satoshi Nishioka. 2014. Kita-Ryūkyū shogo no sonzokuryoku to kikido [Language vitality and endangerment in Northern Ryukyuan]. In: Michinori Shimoji, and Patrick Heinrich (eds), Ryūkyū shogo no hoji o mezashite [In Pursuit of Ryukyuan Language Maintenance]. Tokyo: CoCo Shuppan, 96–142.
Palosaari, Naomi, and Campbell, Lyle. 2010. Structural aspects of language endangerment. In: Peter K. Austin, and Julia Sallabank (eds.), The Cambridge Handbook of Endangered Languages. Cambridge: Cambridge University Press, 100–119.
Pizziconi, Barbara. 2011. Honorifics: The cultural specificity of a universal mechanism in Japanese. In: Dániel Kádár, and Sara Mills (eds.), Politeness in East Asia. Cambridge: Cambridge University Press, 45–70.
Sasse, Hans-Jürgen. 1992. Language decay and language-induced change. In: Matthias Brenzinger (ed.), Language Death: Factual and Theoretical Explorations with Special Reference to East Africa. Berlin: Mouton de Gruyter, 59–80.
Schmid, Monika S. 2011. Language Attrition. Cambridge: Cambridge University Press.
Shibatani, Masayoshi. 2006. Honorifics. In: Keith Brown (ed.), Encyclopedia of Language & Linguistics (vol. 5, 2nd edition). Oxford: Elsevier, 1600–1608.
Shimoji, Michinori, and Thomas Pellard (eds.), 2010. An Introduction to Ryukyuan Languages. Tokyo: Research Institute for Languages and Cultures of Asia and Africa.
Tsitispis, Lukas D. 2003. Implicit linguistic ideology and the erasure of Arvanitika (Greek-Albanian) discourse. Journal of Pragmatics 35: 539–558.
Tsutsui Billins, Martha. 2021. A Sociolinguistic Documentation of Honorific Verbs in Southern Amami Ōshima. Ph.D. dissertation, SOAS, University of London.
Watts, Richard. 2003. Politeness. Cambridge: Cambridge University Press.
Biographical Notes
Gijs van der Lubbe is a part-time lecturer at the University of the Ryukyus and Okinawa Christian University. His research interests focus on the grammatical description and sociolinguistics of Northern Ryukyuan languages, as well as the construction of learning materials for these languages.
Martha Tsutsui Billins is a lecturer in the Anthropology Department at California State University, Fresno. She recently received her Ph.D. in Linguistics from SOAS, University of London. Her work focuses on Ryukyuan language Amami Oshima sociolinguistics, particularly honorifics and politeness strategies.
Patrick Heinrich is professor of sociolinguistics and Japanese Studies at Ca-Foscari University of Venice. His present research is focussed on language and wellbeing in endangered language communities.
We use the term ‘consultant’ in this paper instead of ‘informant’ or ‘participant,’ because this term captures best that many of the insights presented here have been developed in close collaboration with our consultants.
Further research examining fixed registers may reveal more types, such as those that are described in this paper for the case of Kumejima.
Gijs van der Lubber informed the consultants the conversations would be recorded, and that the data would be used for research purposes.
All consultants gave informed consent to be recorded and for the data to be used in publications and archived. Participants also signed consent forms.
While Anderson (2014) and Ishihara (2014) have noted the existence of “formulaic phrases” in Ryukyuan languages, no in-depth study has explored the pragmatic functions of these phrases.
