 Hello, I'm Shirai Satoko. It is a great pleasure for me to be here and share with you some of the findings from my field work in the Andrapa language. Here are the contents of this talk. First, I provide some basic information about the language, geography, and culture of the Andrapa people. Next, I will profile features of the Andrapa language. The last section will discuss some of the distinctive linguistic phenomena. In this talk, I will focus on the directional prefixes and related phenomena. OK, the first part is language and people. Andrapa is spoken in these mountainous areas where steep mountains and deep valleys meet. It is located in the western part of Sucram province, China. Andrapa is a Tibetan term for the people who speak this language. In Chinese, it is termed Drapa, pronounced as Drapa in Sucram Mandarin. In their old name, the people is referred to as Drapi, and the language is called Drakke. Drapa is one of the Tibetan-Berman languages which belong to the Sino-Tibetan language family. A number of languages with similar typological features are distributed in the vicinity of Drapa, forming a language area. We call the area as the Chiangik, language area, or the western Sucram ethnic corridor. These languages are also hypothesized to form a genetic branch called Chiangik. In particular, genetic relationships are being established for the gharongik group of languages indicated with the blue triangles. The Chiangik language area is located at the eastern end of the Tibetosphere or the Tibetan cultural area. The Chiangik languages are surrounded by three major languages with a written tradition and numerous speakers, that is Chinese dialects, Tibetan or Tibetan languages, and Ii or Loloish languages. Drapa is one of such minor languages. And in this area, villages are scattered among steep mountain valleys, and so they themselves say each village has its own dialect. According to preceding studies, approximately 10,000 people speaks Drapa as their first language. There is no writing system or traditional writing in this language. Tibetan has been the traditional lingua flanca in this area, and Chinese is now the predominant language. So Drapa has been under the strong influence of these languages. Dialect differences exist even in this small language, and it is considered that there are three dialect groups, northern, southern, and central. In this talk, I will use data from my fieldwork on the metro dialect, the most northern variety. Drapa people live in stone houses as seen in these pictures. They are known for their matrilineal society with so-called walking marriages. They live by cultivating a small field, raising a few livestock, and gathering wild vegetables. Men often go to distant towns to work. This is a traditional kitchen. They sit on a wooden floor, and they have a big hearth in their kitchen. Okay, I hope you're interested in them. Then let's move on to the linguistic part. Phonology. This slide shows the inventory of consonants and vowels. In addition to these basic segments, we can find pre-nasalized consonants. For example, belly, all, or pay, ice, and so on. Pre-aspirated consonants, such as abu to learn, dopu to carry, and so on. And other consonant clusters. Drapa has word tones that is distinctive pitch patterns that are carried by phonological words. They are categorized into four tones according to the pitch pattern of the first two syllables. High level, which is realized as high high of the first two syllables. Second is high falling, high falling, and realized as high and low of the first two syllables. Third, low rising, and that is low high in the first two syllables. And low rising falling, which is realized as rising and low of the first two syllables. And the rest of those are realized as low. However, the lexical tone may be overridden by the intonation or post-lexical prosody of the topic or focus. For example, the word for teacher is ge ge. So this is a playing pattern. Lexically, this word has tone three, low rising tone. However, when it comes to the topic position, it may be pronounced as ge ge, as high high. Or it is focused on, if it is focused on, it can be said as ge ge, as tone four, rising and low. From a viewpoint of morphological typology, DRAPA mostly shows a cognitive features. It employs both prefixes and suffixes. For example, verbs can take a directive prefix, a negative prefix, an aspect suffix, and so on. In this example, dotte and ching mo xia, this means nothing other than the harsh harsh have remained. And actually, it means everything has lost. In this case, the word stem is xia, remain. And two prefixes are touched before the stem, no, outward directive, mo, the negative, and suffix, suffixes attached here are is a factual and perfective suffix. Adjective stems can be reduplicated productively and form nominalizations. That is, adjective stems have verbal features, since they can take verbal affixes, but the reduplicated adjectives show nominal features, such as being an argument of a sentence. For example, zo is this stem, that means the sour. And this stem can have verbal affixes, such as oza. This means it has become sour, or it has rotten. And upward directive prefix is attached, and the perfective suffix is also attached. These are verbal affixes. On the other hand, when it is reduplicated as zozo, it is nominalized. So it can take classifier, noun classifier, and it can mean a sour one, something sour. Syntactically, it is a vowel final language. The basic constant order is s, o, v. Grammatical relations are indicated by case markers, and case markers are post-nominal and critiques. So for example, nga mitsu shishiro, this means I am putting a cat. And the subject of transitive verb is not marked, just nga. And the object is marked with wu, mitsu. However, such accusative marker can be dropped if the grammatical relation is clear from the context. But in this case, nga is not marked, and the object mitsu is marked with the accusative marker wu, and then shishiro, the verbal predicate, comes. In the structure of a noun phrase, the head noun can be modified both from before and from after. So there are pre-head modifiers and post-head modifiers. Pre-head modifiers include demonstratives, nouns, and nominalizations. Whereas post-head modifiers include adjectives in the replicated form and quantifiers. And this is an example of a noun phrase, me de kimura ncho loi, one vowel that his mother gave him. So the head noun is this one, ncho lo, a vowel. And mother gave me de kimura. This is nominalized with suffix m, and genitive is attached to this addonominal clause, or maybe this is similar to a so-called relative clause. And this is a verbal-based nominalization, and it modifies the head noun. Before the head noun, and then head noun comes, and after the head noun, e-classifier is attached. And in this case, this classifier means one. So this is one of the quantifiers. And this means one vowel in cho loi that his mother gave me de kimura. There are a number of sentence final particles for mood and modality, like many other East Asian languages have. For example, te is for hearsay, pa is for inferential, mo for conformative, re is for factual or alophoric, a or ra for double-h questions, me or me for polar questions, sa for admirative, and so on. So for example, when they say it is raining, the form with no sentence final particle is used, for example, mok atere. It is raining. This pattern is used if the speaker directly knows the fact. But if the speaker learns about it from someone else, for example, his friend said that it's raining and he now knows it's raining, then he says, mok atere de. The last te is the hearsay sentence final particle. And so this means it's raining, but implies that they say it's raining or I heard it's raining. OK, now let's talk about some linguistic topics. Here I'd like to focus on directional prefixes and related phenomena. First, please take a look at this example. This is a quote from the folklore. In the situation is like this, a rich man and uncle Tempa, a very famous wise man, went out to do business with a lot of grain. And the next night, they stayed at the temple. But the next morning, they found the grain disappeared. But actually, it was uncle Tempa's plan. OK, then the servants of the rich man went everywhere outside the temple to look for the grain. In this part, your plane will be to the chariot. The servants outside look for a go and go. In this part, the verb cho, which means go, takes the prefix do. OK, in the next part. But uncle Tempa went up into the temple and did something. This part, uncle Tempa knows everything. Genbe kebi ochare, temple inside of temple, he went. In this part, the verb cho, go, takes o, prefix o, different prefix than the last part. And then he went down to where the rich man was. Ho, ngoro akutemba, that akutemba. Gebudo achore, as a rich man's place, went. And here, cho, the verb stem, takes a, prefix a. So the verb takes different prefixes in accordance with the context or something like that. OK, these prefixes show the direction of movement. As now you know, they live in the steep mountains and deep valleys. And when they go upward, the prefix o is attached. And this prefix alternated to o in the previous example. Similarly, when they go downward, a is attached before the verb stem. And this vowel is not altered. And upstream or inward, go is used. And outward or downstream, ng is used. And t, prefix t, indicates indefinite or neutral direction. The next example is ngara-niu-niu-pan-japi-at-diare. And in this example, japi, here, denotes the japa region. And the verb takes the upward directional prefix. Since this sentence was uttered in the town of Stau, which is at a higher attitude than japa. Japa is the origin of the movement, upward movement. Japa is lower and Stau is upper. And the directional prefix suggests the direction of movement. It is worth noting that the noun phrase japi do not have a case marker. Of course, in japa, the nominative case marker is zero form. So the subject, gara-niu-niu-pan, do not have any case marker. But japi, the origin of movement, do not take any case marker. Actually, japa do not have proper case marker for the origin of movement or for form in English. But they have directional prefixes. So they know the direction of movement. And they know which of Stau or on japa is located upper position. So they know whether it is the origin or goal of the movement. So if the verb implies a movement, the directional prefixes indicate the direction of movement. So these verbs may take any of the directional prefixes to indicate the direction of movement. For example, basic movement verbs or run, flow, please, push or arrive. Then, what about verbs that can't seem to choose a direction? For example, to eat or to die. They also take one of the directional prefixes, but the connection is fixed. Most verbs are associated, lexically associated with only one of the directional prefixes. Such directional prefixes are obligatory in the perfective and imperative. For example, kotsu is eat the imperative. And this inward directional prefix is obligatory in this case, before the word eatzu. And for example, me dosunda asoare. The mother told the children what they should do later before she died. In this case, the verb she takes the neutral directional prefix and so order or told takes the downward and directional prefixes. These are fixed connections. So for example, eat cannot take, for example, downward the directional prefix. Or die cannot take, for example, upward the directional prefixes. It is fixed in this language. Some semantic shifts are found on the directional prefixes, especially on the outward directional prefix. We consider this was originally the outward directional prefix. For example, ngeuki, give birth or to be born, can take this prefix only. This is a lexically fixed connection. And this is also used for the downstream movement. So, for example, ngeoli means a gentle breeze blows from upstream to downstream, toward downstream. So the prefix nge is used. Since this verb stem, a breeze blows, li is one of the movement verbs. So it can choose any of the directional prefixes in accordance with the direction of movement or blowing. So if they say gali, it means a breeze blows from downstream to upstream. G is inward and upstream directional prefix. However, for a strong wind like a storm, only the outward directional prefix is used, ngechi, a strong wind blows. This verb cannot choose the directional prefix. So this is not directive, but this implies something drastic, drastic movement or drastic event. Nge is often found with such verbs, implies a drastic change or drastic event. Finally, I'd like to point out the area features. Having a set of directional prefixes is one of the area features shared by the languages of the Changi language area. In the Changi languages, a directive marker can fall into the preverbal slot of motion or orientational verbs according to the direction of motion. However, other features vary among languages. For example, drapa shows the feature that it is obligatory in perfective and imperative and semantic shift from outward to drastic. But these features are different in other languages. These are references and more references. Lastly, I'd like to thank the consultants who taught me the drapa language. Thank you all for listening to the end. Akka, thank you in drapa.