An earlier version of this article was presented at Amdo at the Crossroads: 3^rd International Workshop of the Amdo Research Network (hosted by Oxford University; online; September 2021). Our thanks will go to Giulia Cabras, Keith Dede, and Shannon Ward for their initial editing of the draft with insightful comments. This work is part of the research outcomes of a Grant-in-Aid of the Japan Society for the Promotion of Science: “Investigation of undescribed languages in the eastern Tibetosphere and their geolinguistic research” (headed by Hiroyuki Suzuki; no. 17H04774).

Introduction

2Cone is located at the north-eastern corner of the Tibetosphere, generally known as Amdo (Tib. A mdo). Currently, Cone is a county located in Kanlho (Tib. Kan lho) Tibetan Autonomous Prefecture, Gansu Province. Historically, Cone had a more extensive territory under the local chieftain (Ch. tusi 土司) of the Yang family, which included the present adjacent counties of Thewo (Tib. The bo), nDrugchu (Tib. ’Brug chu), and Batse (Tib. sBra rtse) (’Brug thar 1996, pp. 188-199; ’Brug thar & Chab mgo Tshe ring 2009, pp. 278-283).

4The linguistic situation of Cone is complicated. Both Cone Tibetan and Amdo Tibetan are spoken in the county. There are several local vernaculars within Cone Tibetan that have low mutual intelligibility with one another. rNam rgyal tshe brten (2008) classifies Tibetic languages in Cone County into four sections: the north-western part of Klu chu, the southern part of Klu chu, the area alongside Klu chu, and the eastern part of Klu chu, among which the first group mostly corresponds to Amdo Tibetan (cf. fig. 2). Suzuki (2012) examines his argument and proposes a revised classification. Previous linguistic descriptions, such as dKon mchog rin chen and ’Brug mtsho skyid (2012) and Jacques (2014), describe the Nyinpa (Tib. Nyin pa) dialect, which, unfortunately, is a minority vernacular; Suzuki (2015a) puts Nyinpa in another group together with Thewo-stod (Tib. The bo stod) Tibetan. The majority of vernaculars of Cone Tibetan are described, for example, by Qu (1962), Yang (1995), Suzuki (2016b), bKra shis tshe ring (2019), and Zou (2021). The most striking features that distinguish Nyinpa from other varieties of Cone Tibetan are sound correspondences of the Literary Tibetan (henceforth LT) series c, ch, j, sh, and zh, realised as prepalatal affricates and velar fricatives in Nyinpa while primarily realised as retroflex affricates and fricatives in the others (Suzuki 2012). Figure 2 gives a provisional language classification of Tibetic languages in Cone County.

5Communications across native vernaculars of Cone Tibetan are difficult. Amdo Tibetan functions as a “lingua franca”, especially in schooling, whereas Chinese (Northwestern Mandarin) is used in everyday life in the city and most villages in the eastern half of the county. Many varieties of Cone Tibetan face severe sinicisation of the language use, and hence are highly endangered, as discussed by Bendi Tso and Turin (2019).

6In our chapter, we focus on the Dungrimdo (Tib. Dung ri mdo) variety of the Bragkhoglung (Tib. Brag khog lung) dialect, the mother tongue of the second author. It belongs to the “southern part of Klu chu” group (rNam rgyal tshe brten 2008; Suzuki 2012), spoken in Bragkhoglung Town in the western part of Cone County. Suzuki (2012) and Zou (2021) describe its phonology and dialectal features.

Folktales and theoretical backgrounds

7Folktales have characteristics differing from other genres, especially legends and religious narratives, as discussed by Lüthi (1947). Folktales in the Tibetosphere are known from various publications, such as those in an English translation by O’Connor ([1906] 1977), Shelton (1925), and Kajihama (2004), and in Chinese translations by Song (2004), Lin (2016), and sKal bzang sgrol ma (2020). Linguists, such as Bielmeier (1985), Kretschmar (1986, 1995), Causemann (1989), Herrmann (1989), Schwieger (1989), and Haller (2000, 2004), also publish folktales with German translations and an original phonetic transcription. However, since Tibetan is rarely written in an oral style, folktales published in Tibetan are written in Modern Literary Tibetan with some oral-style arrangements as, for example, in Mi ro rtse sgrung (1963), A khu ston pa (1980), and Tshe dbang rdo rje et al. (2006).

8Folktales in the Tibetosphere are thus not regarded as “oral literature”, that is, a literature that reflects a manner and scene of narration, because Tibetans usually do not write oral varieties in the Tibetan script. Rather, Tibetans generally use LT as their cultural norm in writing. The LT version of folktales can be regarded as a translation. However, we take issue with the practice of recording, documenting, and representing folktales in LT on theoretical grounds that underpin folktales and storytelling more generally.

9Ozawa (1999) argues the theoretical framework of folktales and practical issues in documenting and reproducing them by primarily examining the theory of Lüthi (1947). Ozawa raises an essential question to readers: “Where is a folktale?” His answer is “a folktale only exists during the time of being narrated by a storyteller”, and he reminds us of the fundamental idea that a folktale is aural literature with a temporal nature similar to music (Ozawa 1999, p. 11). For many people, enjoying music means enjoying listening to the music, not enjoying reading musical notes and scores. Similarly, enjoying a folktale should denote the joy of listening to a folktale, not the joy of reading books about it.

10Ozawa’s theoretical question is directly connected with how a local Tibetan community transmits its oral heritage. Today’s method of transmitting stories can be expanded to encompass reading storybooks and looking at picture books, but the essential issue is how to provide a narrative style in books. Writing an oral variety with the Tibetan script would function well if the community accepted it. Suzuki and Sonam Wangmo (2015, 2017a, 2017b, 2018, 2021a, 2021c) as well as Suzuki et al. (2015) have already begun to provide oral Tibetan writing of folktales and legends with a linguistic annotation. Additionally, to fulfil the objectives of language documentation, one needs to record the entire scene of folktale narration so that the complete act, including such things as pauses and gesture, are documented.

11According to the theory emphasised by Ozawa (1999), visualising folktales in any format is not the best choice to maintain and transmit them. However, we encourage making an online archive of audio-visual materials of folktales in addition to publishing folktale books to ensure the language community has practical access to the stories. Audio-visual materials would need subtitles that reflect the tales’ meaning in a given lingua franca as well as a given local variety to help an audience reproduce the story. Our focus is on issues that arise when one edits a folktale book, that is, how to write down and represent folktales in texts.

Writing folktales

12We have collected several folktales from the Dungrimdo hamlet of Bragkhoglung Town. The content of some of them is recognisable in previously published works collected in other places of the Tibetosphere. Table 1 displays folktales in Dungrimdo of the “animal-protagonist” type, compared with other versions.

Table 1. Similarities between the folktales in Dungrimdo and previous publications

Folktales in Dungrimdo	Previous publications’ version
The Hare and the Lion	O’Connor’s ([1906] 1977, pp. 51-55) The Hare and the Lions; Shelton’s (1925, pp. 81-85) How the Rabbit* Killed the Lion
Hare’s Wisdom	Shelton’s (1925, pp. 136-137) How the Wolf, the Fox, and the Rabbit Committed a Crime
The Hare and the Tiger	O’Connor’s ([1906] 1977, pp. 1-3]) How the Hare got his Split Lip** Suzuki & Sonam Wangmo’s (2021c) The Hare and the Tiger
The Ewe and the Wolf	O’Connor’s ([1906] 1977, pp. 56-59]) The Sheep, the Lamb, the Wolf and the Hare; Kajihama’s (2004, pp. 51-53) The Ewe and the Wolf; Lin’s (2016, pp. 320-321) The Old Woman and the Tiger (partially similar)
	* Shelton (1925) uses the word “rabbit”; however, as O’Connor (1906) does, it should have been translated as “hare”. ** The story corresponds to the first part of How the Hare got his Split Lip.

13Below we take a folktale The Hare and the Lion, and demonstrate how to practise writing an oral variety with the Tibetan script. The folktale’s content is summarised as follows: “The Lion, as the king of the beasts, should have persuaded others not to fight, but the Hare did instead. The Lion asked the Hare who intervened, and the Hare tricked the Lion, making him look into the sea, then threw him in there”. It concludes with the words “the way of the Hare” describing the cleverness of the Hare.

(1) ʈʂ^hẽ¹¹zã¹⁴kə⁵⁵ dʑa¹¹wo⁵⁵ s^hi¹¹ɡə⁵⁵ re⁵¹.
The king of the beasts is the lion.

(2) s^hi¹¹ɡə⁵⁵ tə⁵⁵ ts^hi¹¹ɣa⁵⁵ ʂ^he¹⁴nɑ¹¹ɡə⁵⁵mi⁵⁵ ʔɑ¹¹ra⁵⁵.
Lions have a special appearance, right?

(3) ʔə¹¹zə⁵⁵ ta⁵¹dɑ¹²… ʔo⁵⁵ ke¹⁴ ta⁵¹dɑ¹² ʈʂe¹¹mo⁵⁵ nə¹¹ɣɑ⁵⁵lo⁵⁵… ʔo⁵⁵ zi⁵¹dɑ¹² ʈʂe¹¹mo⁵⁵ nə¹¹ɣɑ⁵⁵ dʑa¹¹ri⁵⁵ ɕi⁵¹ta¹¹zə⁵⁵.
Then, Tiger and… no, both Tiger and Brown bear… no, both Leopard and Brown bear fought (with each other).

(4) ti⁵⁵ tõ¹¹dɑ⁵¹ tə⁵⁵ s^hi¹¹ɡi^{55 h}ʂe⁵¹ko¹¹ɣe⁵⁵ s^hi¹¹ɡə⁵⁵ ʔə¹¹zə⁵⁵ ʈʂ^hẽ¹¹zã¹⁴kə⁵⁵ dʑa¹¹wo⁵⁵ re⁵¹.
For that event [fight], that Lion needed to persuade (Leopard and Brown bear not to fight), (since Lion) is the king of the beasts.

(5) s^hi¹¹ɡi⁵⁵ ma^11hʂə⁵¹kə⁵⁵ ri¹¹wõ¹⁴ɣə^{55 h}ʂe⁵¹ɣa¹¹zə⁵⁵.
Lion did not intervene (in their affair), but Hare did.

(6) ʔə¹¹zə⁵⁵ s^hi¹¹ɡi⁵⁵ ʈʂi¹²ta¹¹zə⁵⁵.
Then, Lion asked (Leopard and Brown bear):

(7) tɕ^hi¹¹nɑ⁵⁵ ʔə¹¹zə⁵⁵ k^he¹¹ni¹¹ba⁵⁵ dʑa¹¹ri⁵⁵ ɕi⁵¹kə⁵⁵tə⁵⁵ti⁵⁵mi⁵⁵ tɕ^hi¹¹ne⁵⁵ tõ¹¹dɑ^{51 h}ʂə⁵¹mo⁵⁵ ka¹⁴ re⁵¹dze¹².
“Both of you were fighting each other the day before yesterday, who was the one that intervened in your business?” said (Lion).

(8) tə⁵⁵ ʔa¹¹kə⁵⁵ ri¹¹wõ¹⁴ re⁵¹dze¹².
“That was Uncle Hare”, said (both).

(9) ʔə¹¹zə⁵⁵ ʔa¹¹kə⁵⁵ ri¹¹wõ¹⁴ pi⁵¹ta¹¹zə⁵⁵.
Then, (Lion) called Uncle Hare (and asked):

(10) ʔe⁵⁵ ʔa¹¹kə⁵⁵ ri¹¹wõ¹⁴ tɕ^he¹¹ ʔa¹¹na⁵⁵ ti⁵⁵ tõ¹¹dɑ^{51 h}ʂə⁵¹nə⁵⁵ ʈʂ^hi¹¹ re⁵¹.
“Hey, Uncle Hare, why do you intervene in this business?”

(11) ʔe⁵⁵ k^ho⁵⁵ ma¹¹re⁵¹ kə¹¹ra⁵⁵na¹² tɕa¹¹ts^ho⁵⁵ nɑ¹¹na⁵⁵ ri¹¹wõ¹⁴zə⁵⁵ nɑ¹¹ɡə⁵⁵ tə⁵⁵ re⁵¹sei¹⁴.
“Eh, (that) is not me, there is a hare inside the sea, that is him”, said (Hare).

(12) tə⁵⁵ ka¹¹ra⁵⁵ ji⁵¹.
“Where is that [Hare]?” (said Lion).

(13) s^hi¹¹ɡi⁵⁵ro¹² k^ho¹¹ro⁵⁵kə⁵⁵ ta¹¹ ŋo¹¹ta⁵⁵ nɑ¹¹ɣə⁵⁵ɦa⁵⁵ s^hi¹¹ɡi⁵⁵ ŋo¹¹tə⁵⁵ nɑ⁵⁵ rɑ¹⁴ ʂ^ha¹¹ʨi⁵⁵ jə¹¹lo⁵⁵ tə⁵⁵ ma¹¹ʂ^he⁵⁵zə⁵⁵.
Lion himself did not know that he would appear in the mirror.

(14) s^hi¹¹ɡə⁵⁵ tɕ^hi⁵¹kə⁵⁵sə⁵⁵ s^hi¹¹ɡə⁵⁵ ʈʂa¹¹ⁿɡo⁵⁵ tɕ^hi⁵¹kə⁵⁵sə⁵⁵nə⁵⁵ t^hə¹¹ri⁵⁵ tɛ¹¹ɣe⁵⁵ʈʂə¹¹lo⁵⁵tə⁵⁵.
(Hare) took Lion, took him to the top of the cliff, and made him look downwards,

(15) k^ho¹¹ro⁵⁵ ʔə¹¹zə⁵⁵ tɕa¹¹ts^ho⁵⁵nɑ⁵⁵ t^hə¹¹ri⁵⁵ tɛ⁵⁵ti⁵⁵ ʔa¹¹na⁵⁵pa¹¹na⁵⁵zə⁵⁵ nɑ¹¹ɡə⁵⁵mi⁵⁵ te⁵⁵.
Then, when he (Lion) looked into the sea, he found the same (figure as him).

(16) ri¹¹wõ¹⁴le⁵⁵ ts^hi¹¹ɣa⁵⁵ mə¹¹nɑ⁵⁵mi⁵⁵.
Hare (in the sea) did not have any appearance.

(17) nã¹¹ʈʂ^ho⁵¹ ri¹⁴lo⁵⁵ me¹¹gə⁵⁵ ts^hi¹¹ɣa⁵⁵ ʈʂa¹¹ʈʂa⁵⁵ mə¹¹nɑ⁵⁵mi⁵⁵.
(He [Hare]), except for having long ears, did not have the appearance at all.

(18) s^hi¹¹ɡə⁵⁵ ha¹¹ɡe⁵⁵ ʂ^he¹⁴ pə⁵⁵ ri¹⁴lə⁵⁵ ts^hi¹¹ɣa⁵⁵ nɑ¹¹ɡə⁵⁵mi⁵⁵.
Lion has a really special appearance with long hair.

(19) t^hə¹¹ri⁵⁵ to¹¹ri⁵⁵ ʈʂi¹¹ma⁵⁵ ho¹⁴ la⁵⁵nə⁵⁵ tɕ^ho⁵⁵ tə¹¹ka⁵⁵ ŋa¹² ts^hi¹¹ɣa⁵⁵ t^hẽ¹¹ɡə⁵⁵.
(Lion) also got angry downwards and (said:) “You showed me a little your dignity”.

(20) ŋa⁵⁵ tɕ^ho⁵⁵ t^hə¹¹ɕ^hi⁵⁵ ts^hi¹¹ɣa⁵⁵ t^hẽ¹¹ne⁵⁵dɑ¹² lə¹¹ɕ^hi⁵⁵ ts^hi¹¹ɣa⁵⁵ t^hẽ¹¹ɡə⁵⁵.
“I showed you my dignity downwards, and I showed you my dignity upwards”.

(21) s^hi¹¹ɡə⁵⁵ tɕa¹¹ts^ho⁵⁵ nɑ¹⁴ ha¹¹to⁵⁵mo⁵⁵ ri¹¹wõ¹⁴ re⁵¹.
It was Hare that threw Lion into the sea.

• 9 The narrator mentions Tiger and Brown bear here; however, it should be Leopard and Brown bear who f (...)

(22) ʔə¹¹ze⁵⁵ ri¹¹wõ¹⁴ɣə⁵⁵ lo⁵⁵ ko⁵⁵nə⁵⁵ ta⁵¹dɑ¹² ʈʂe¹¹mo⁵⁵ nə¹¹ɣe⁵⁵ pa^{55 h}ʂu⁵⁵ta¹¹zə⁵⁵.
Then, Hare used his intelligence and persuaded both Tiger and Brown bear9.

(23) s^hi¹¹ɡə⁵⁵ tɕa¹¹ts^ho⁵⁵ nɑ¹⁴ ha¹¹ta⁵⁵zə⁵⁵ tɕ^he⁵⁵tə⁵⁵.
(It was) you [Hare] that threw Lion into the sea.

(24) ʔə¹¹zə⁵⁵ tə⁵⁵ hã¹¹sə⁵¹ t^hu⁵¹tsə⁵⁵kə⁵⁵ pan¹¹fa⁵⁵ re⁵¹.
Then, that is also the way of Hare.

16A minimum “formatting” is potentially necessary work to represent a piece of oral literature, and we can justify omitting the audience’s actions in the storytelling. However, it seems that it is contradictory that the utterances (1)-(24) include fillers (words like “uh”, and overused “then”), mistakes, and self-corrections, as in (3) [mistake and self-correction] and (22) [mistake]. Here we should consider the purpose of this kind of representation. The potential to make local peoples understand the form of the utterances (1)-(24) is unrealistic, because they hardly understand the IPA transcription. The form displayed as (1)-(24) is primarily for scholars as principal target readers. Therefore, other methods of representation should be considered for the purpose of preservation and transmission to the next generations.

1. The LT orthography of lexical words, especially nouns, adjectives, and verbs, are maintained as much as possible. For example, /ta⁵¹/ means “tiger” and “sign”, which are written as stag and rtags, respectively. This policy makes a given writing understood more explicitly. It depends on speakers how they pronounce the spellings or recall the local pronunciation with a hint of the spellings.

2. Grammatical words and particles (e.g. affixes and interjections) and local words that are not derived from LT lexical items or underwent drastic sound changes may be written as a phonetic transcription with the Tibetan script.

18For example, the sentence (1) is transcribed as in (25):

20Many Tibetans probably expect the form in (27) as the written representation. However, there is a major difference between (25) and (27): the final verb. (25) = (1) uses /re⁵¹/ (red) “be (statemental copulative verb)”, whereas (27) uses yin. In various Tibetic languages, (27) is strange as an “utterance”, because the grammar of LT differs from spoken varieties. On the other hand (26), an Amdo Tibetan version, has more potential to be accepted by local Tibetans. Poets and novelists from Amdo publish their work by using a pronunciation-based spelling nowadays. However, the language represented in (26) differs from Cone Tibetan. If we consider folktales a kind of oral literature, we need a method to reproduce the spoken forms in writing.

21As a component of the complete documentation of folktales for the purposes of preservation, we should edit the storyline even if there are mistakes in the storytelling itself. For example, in sentence (3), the output should be the form of (28) as a complete version:

22We only use the part outside the square brackets for our presentation. This is because the story does not need the storyteller’s interference.

23Careful consideration is needed for lexical issues. For example, sentence (24) contains Chinese words inserted in the utterance. With some additional analyses, we find the language use in (24) is “mixed” as an interpretation of the story, which can be put outside its body.

24In writing (29), we replace the Chinese words (hã¹¹sə⁵¹ “also” < haishi 还是; t^hu⁵¹tsə⁵⁵ “hare” < tuzi 兔子; pan¹¹fa⁵⁵ “way” < banfa 办法) with Cone Tibetan words (da rung; ri bong; bkod pa). The insertion of the Chinese words is due to the narrating style of the individual storyteller who is code-switching; the Chinese words are not loanwords that substitute missing native words. Hence, we select native words in (29) that are still used in everyday communications for the purpose of reproduction of the story as a local folktale.

25Table 2 shows a line-by-line re-writing of the story in the Tibetan script (Cone Tibetan style) that we propose and the phonetic description (1)-(24).

26The reading pronunciation varies depending on towns and townships, even hamlets, within Cone County. However, this writing method enables local Tibetans to understand stories and reproduce the narration style as they speak. Each Tibetan spelling functions as a hint to recollect its local sounds, with which they reconstruct a story. This role cannot be substituted by writing in LT or Amdo Tibetan. See Appendix, providing a full transcription of three styles: Cone Tibetan, Amdo Tibetan, and LT, as well as a full English translation.

27We have not widely employed the orthography in the local community, Bragkhoglung Town, yet. However, through our short interviews with locals, this approach is more or less welcomed; at least, it is better than nothing. However, as discussed by Zeisler (2008), Ladakh, a Tibetic-speaking community still has long debates among the speakers on the stance to writing reform. It is probable that similar discussions also emerge in Cone; however, no pilot experiments have been examined in a wider community of Cone County. The relevant discussion is beyond the scope of the chapter; however, we should note this issue as a future task.

Writing oral varieties versus norms

28Local language preservation is an important issue in communities worldwide, especially as it relates to interests in language endangerment and language diversity. Tibetan communities must come to terms with two issues: development of a standardised language and preservation of language diversity. The former principally arises throughout the Tibetosphere, and the latter is often discussed in communities using minority languages; that is, non-Tibetic languages spoken by Tibetans. Consequently, particular difficulties emerge in communities where Tibetans speak a marginal variety of a Tibetic language. Tibetans in such communities are simultaneously pressured by a majority variety and a language standardisation movement.

31Nevertheless, as far as oral literature is concerned, spoken forms should have written representations. Linguists can read and understand IPA and interlinear glossing; however, even scholars without knowledge of linguistics will have difficulty understanding linguists’ methods, to say nothing of local people. For the purpose of language preservation, reproduction, and transmission, writing with a native language script is an option, providing a reproduction of “reading pronunciation” according to each linguistic area. Reading pronunciation must be noted as “non-dialectal form” when linguists work on dialectology (Denwood 1999), since it does not always reflect spoken forms in a given dialect. However, if the Tibetan script is given following a spoken manner, there will be the possibility for it to serve as a “hint” to reproduce a series of utterances in the local people’s vernacular.

32When one plans to publish a picture book or an anthology of folktales as materials for language preservation, it is helpful to consider writing a spoken variety with the Tibetan script. Here we do not have enough space to discuss “how to reflect spoken forms with the Tibetan script” since the customs of reading pronunciation vary according to communities. However, we must first consider whether using the Tibetan script to reproduce a storyline of folktales is appropriate, and then we can discuss how to write it down.

34One should also pay attention to the intentional use of invented dialectal forms in a folktale, especially in a language possessing a standardised variety, as Ozawa (1999, 2016) points out. This invention enhances a plausible style as a folktale, often practised by storytellers who speak a standardised variety more than dialects. In Tibetic languages, conversely, we have to consider not inserting standardised forms in a dialect version. Writing spoken varieties has not widely been practised in the Tibetosphere; hence, we need a more careful approach. As shown in table 2, a favourable way is to write a Tibetan transcription according to the spoken pronunciation. For this purpose, a phonetic description is a basis for making a dialect version; following this method, we can escape from an unconscious insertion of standardised or literary forms.

Conclusion

36The present description is an essay on the practice of writing a spoken variety of the Bragkhoglung dialect of Cone Tibetan, a minority Tibetic language spoken in the Amdo region. By centring on the theoretical basis of folktales, we have argued how to transmit oral language cultures. First, we discussed the nature of folktale as a temporal art of narrating, and writing folktales is a secondary purpose of transmission and reproduction of them. This means that when we consider the preservation of oral literature, we do need audio(-visual) recording materials. Meanwhile, we can offer subtitles with the Tibetan script for those who do not understand the spoken variety in the recording as well as for those who want to learn and reproduce narrations. For these purposes, the writing has to reflect the oral variety. We conclude that writing oral languages with Tibetan script can help the preservation of oral cultures and transmission to the next generations. Such an essay should be understood in a different manner from the prestige of the literary language in the Tibetosphere.